University of Ghana http://ugspace.ug.edu.gh
SCHOOL OF PUBLIC HEALTH 
COLLEGE OF HEALTH SCIENCES 
UNIVERSITY OF GHANA, LEGON 
 
 
A STUDY OF COMMUNITY PERCEPTIONS AND STRATEGIES TO 
IMPROVE WOMEN’S DIETARY QUALITY IN SAVELUGU-NANTON 
DISTRICT, GHANA 
 
 
BY 
 
SAWUDATU ZAKARIAH 
(10030800) 
 
 
 
THIS THESIS IS SUBMITTED TO THE UNIVERSITY OF GHANA, LEGON IN 
PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE AWARD OF 
PHD PUBLIC HEALTH DEGREE. 
 
 
 
 
 
2018
 
University of Ghana http://ugspace.ug.edu.gh
DECLARATION 
I, Sawudatu Zakariah, hereby declare that this thesis was written by me in the School of 
Public Health of University of Ghana under the supervision of Prof. Richmond Aryeetey, 
Prof. Augustine Ankomah and Prof. Kwesi Torpey. All references cited in this document 
have been duly acknowledged. 
 
 
……………………………     ………………… 
Sawudatu Zakariah      Date 
(Student) 
 
……………………………     ………………… 
Prof. Richmond Aryeetey     Date 
(Supervisor) 
 
……………………………     ………………… 
Prof. Ankomah      Date 
(Supervisor)  
 
……………………………     ………………… 
Prof. Kwesi  Torpey      Date  
(Supervisor)  
 
 i 
University of Ghana http://ugspace.ug.edu.gh
DEDICATION 
To all women whose life of hard work and sacrifices enable their children and future 
generations to have a healthier and successful life  
 
 
 ii 
University of Ghana http://ugspace.ug.edu.gh
ACKNOWLEDGEMENTS 
My unreserved gratitude goes to my supervisors, Dr. Richmond Aryeetey and Professors 
Augustine Ankomah and Kwesi Torpey, for their kind mentorship and assistance through 
this Ph.D. process. My sincerest gratitude goes to both academic and non-academic staff 
of the School of Public Health for their assistance in diverse ways. To Noguchi 
Memorial Institute for Medical Research and College of Health Sciences, University of 
Ghana, I am profoundly grateful for granting me study leave with pay to pursue this 
course. To Professors Margaret Armar-Klemesu (my unofficial supervisor) and Clara 
Fayorsey as well as Dr. Lydia Aziato, thank you for your continuous support and 
directions in all aspects of my academic life. 
I am equally grateful to all members of the two study communities – Yilkpene and 
Kpachilo –for their unreserved co-operation during all the phases of the study. For the 
same reason, the schedule officers responsible for agricultural extension services and 
nutrition in Savelgu-Nanton District deserve my unqualified gratitude. My fieldworkers 
and implementation team deserve my sincerest compliments for their creditable 
performance as my research assistants and local resource persons. 
Finally, but to no mean degree, my sincerest compliments go to the following for their 
spiritual, moral, financial, intellectual and other forms of support: my beloved husband, 
Dr. Harry Akoto, my elder brother (and father-figure), Dr. Mamadu Seidu Zakariah and 
his wife (my foster-mother), Dr. Cynthia Banneman, all my other siblings and my 
trustworthy friends, Tina and Mr. Kwaku Abankwa. To Susan and Sumaila, your 
continuous support has been phenomenal! Araba, Evelyn, Emefa and Victoria, thank you 
for your friendship and encouragement. And to my three little “Akotolets”: - Swetor, 
Akpene and Nuna - we did this together…. thank you for understanding Mummy. 
 
 iii 
University of Ghana http://ugspace.ug.edu.gh
I cannot fail to express my unqualified gratitude to all others who contributed one way or 
another towards the success of the study. All the individuals, groups and organizations 
whose works I consulted, as acknowledged in the references to the study, equally 
deserve my unqualified gratitude. 
May the Benevolent, Omnipotent, Omnipresent and Omniscient Father above reward all 
my benefactors with life-long showers of blessings.  Notwithstanding the assistance I got 
from all the acknowledged sources, all the lapses in the thesis are wholly mine. 
 
 
 
 
 
 
 
 
 
 
 
 
 iv 
University of Ghana http://ugspace.ug.edu.gh
ABSTRACT 
Background: The diets of women in reproductive age (WRA) are an important 
determinant of their health outcomes and quality of life especially in northern Ghana, 
where women’s diets remain persistently suboptimal. However, socio-cultural barriers, 
one of the multiple influences on WRA’s diets, are rarely addressed in existing 
interventions. This study identified the socio-cultural barriers to WRA achieving optimal 
diets, which, herein, represent beliefs, attitudes and actions that are derived from the 
existing native and emergent culture of the community and tested the feasibility of a 
community-based pilot intervention developed in consultation with the community to 
address these barriers. 
Methods: Socio-cultural barriers to women’s diets were identified and characterized 
through a qualitative study involving the exploration of the local food system, dietary 
knowledge, attitudes and practices at Yilkpene and Kpachilo. Data was collected through 
key informant interviews, focus group discussions, qualitative 24-hour dietary recalls 
with in-depth interviews and observations at the household and community levels. A 17-
week attitudinal and behaviour change intervention was designed and implemented at 
Yilkpene in consultation with key community stakeholders. The first component, 
nutrition education, emphasized the importance of women’s diets and their reproductive 
health. The second component, advocacy, concerned empowering women to have more 
control over household food resources and modifying food taboos affecting women. 
Messages were delivered in the local language through participatory activities including 
community durbars, small group meetings, home-visits and food demonstration sessions. 
In the post-evaluation of the intervention, data was collected from 182 participants using 
the same tools used at baseline. The two sets of data were compared to assess the impact 
of the intervention at Yilkpene. 
 v 
University of Ghana http://ugspace.ug.edu.gh
Results: Study participants, all aged at least 15 years, were mostly WRA (73%), married 
Muslims (87.4%) without formal education whose source of livelihood was farming 
(84.1%) and had households with sizes ranging between 6 and 40. At baseline, gender 
and socio-cultural factors significantly influenced all aspects of the food system, limiting 
women’s access to quality foods. Animal-source food taboos, which are mostly gender-
based and affect the quality of women’s diets, were identified. At endline, these practices 
did not change but members of Yilkpene community were sensitized about them. Prior to 
the intervention, the dietary knowledge and attitudes among members of both 
communities were sub-optimal; but, at endline, improved knowledge and attitudes were 
observed in both communities even though the scope of improvements was more in the 
intervention community. There were also less reported beliefs about plant-source food 
restrictions at Yilkpene compared to Kpachilo. At baseline, close to half (45%) of 
women in either community could not meet their minimum dietary diversity but diversity 
deficit declined at endline (25% at Yilkpene and 10% at Kpachilo). 
Conclusion: Attitudinal and behavioural change communication interventions on 
entrenched socio-cultural issues pertaining to women’s diets need more prolonged and 
sustained durations to enhance their scopes of feasibility. Ghana Health Service and 
individuals should organize similar interventions. 
 
 
 
 
 
 
 
 
 vi 
University of Ghana http://ugspace.ug.edu.gh
TABLE OF CONTENTS 
Content         Page 
DECLARATION ................................................................................................................. i 
DEDICATION ................................................................................................................... ii  
ACKNOWLEDGEMENTS .............................................................................................. iii 
ABSTRACT ....................................................................................................................... v 
TABLE OF CONTENTS ................................................................................................ vii  
LIST OF FIGURES ......................................................................................................... xii 
LIST OF TABLES .......................................................................................................... xiii 
GLOSSARY .................................................................................................................... xiv 
LIST OF ABBREVIATIONS AND ACRONYMS ......................................................... xv 
PREFACE- A PERSONAL REFLECTION ................................................................. xvii 
ORGANIZATION OF THE THESIS .......................................................................... xviii 
 
CHAPTER ONE ................................................................................................................. 1 
1.0 INTRODUCTION ........................................................................................................ 1 
1.1 Background to the study ........................................................................................... 1 
1.2 Statement of the problem ..................................................................................... 5 
1.3 Theoretical framework ......................................................................................... 6 
1.4 Conceptual pathway of change for socio-cultural intervention on women’s diets 
  ............................................................................................................................ 10 
1.5 Justification of the study ......................................................................................... 11 
1.6 Objectives of the study ............................................................................................ 13 
1.6.1 General objective .............................................................................................. 13 
1.6.2 Specific objectives ............................................................................................ 13 
1.7 Research questions .................................................................................................. 13 
 
CHAPTER TWO .............................................................................................................. 15 
2.0 LITERATURE REVIEW ........................................................................................... 15 
2.1 Women’s dietary practices ...................................................................................... 15 
2.1.1 WRA’s dietary practices in Ghana ................................................................... 16 
2.2 Consequences of inadequate dietary intake ............................................................ 18 
2.3 Determinants of women’s dietary practices ............................................................ 19 
 vii 
University of Ghana http://ugspace.ug.edu.gh
2.3.1 Personal Factors ............................................................................................... 19 
2.3.2 Socio-cultural factors ....................................................................................... 21 
2.3.2.1 Women’s social status and decision-making power .................................. 22 
2.3.2.2 Food taboos and restrictions ...................................................................... 24 
2.3.2.3. Preferential food allocation practices ....................................................... 27 
2.3.3 Environmental (physical) factors ..................................................................... 28 
2.4 Interventions addressing challenges to women’s dietary practices ........................ 35 
2.4.1 Interventions addressing challenges to women’s dietary practices in Ghana .. 39 
2.5 Qualitative Research ............................................................................................... 41 
2.5.1 Ethnography ..................................................................................................... 43 
2.5.2 Participatory Action Research (PAR) .............................................................. 44 
2.5.3 Qualitative evaluation of interventions ............................................................ 44 
2.5.4 Analysis of qualitative data .............................................................................. 47 
2.6 Behaviour change communication (BCC) interventions ........................................ 52 
2.6.1 Social cognitive theory ..................................................................................... 56 
2.6.2 Community Readiness Model (CRM) .............................................................. 57 
2.6.3 Principles of adult learning ............................................................................... 62 
2.7 Gaps in the reviewed literature ............................................................................... 63 
 
CHAPTER THREE .......................................................................................................... 65 
3.0 METHODS ................................................................................................................. 65 
3.1. Study location ........................................................................................................ 65 
3.1.1 Selected study communities and justification for selection ............................. 68 
3.1.1.1 Yilkpene (Intervention Community) ......................................................... 68 
3.1.1.2 Kpachilo (comparison community) ........................................................... 70 
3.1.2 Target population ............................................................................................. 71 
3.2 Study design ............................................................................................................ 72 
3.3 Study process .......................................................................................................... 75 
3.3.1 Phase 1: Qualitative exploration of the socio-cultural practices ...................... 75 
3.3.2. Phase 2: Analysis of gaps by community consensus ...................................... 77 
3.3.3. Phase 3: Development of intervention together with community ................... 81 
3.3.4. Phase 4: Implementation of intervention ......................................................... 86 
3.3.5 Phase 5: Evaluation of intervention .................................................................. 88 
3.4 Methods of data collection ...................................................................................... 89 
 viii 
University of Ghana http://ugspace.ug.edu.gh
3.4.1 Sampling of participant categories ................................................................... 93 
3.5 Data recording, management and analysis .............................................................. 96 
3.6 Emerging themes and sub-themes........................................................................... 97 
Table 3.7: Themes and their sub-themes ...................................................................... 98 
3.7 Monitoring the process............................................................................................ 99 
3.8 Quality assurance in the study (trustworthiness) .................................................. 100 
3.9 Ethical considerations ........................................................................................... 102 
 
CHAPTER FOUR .......................................................................................................... 105 
4.0 RESULTS OF THE STUDY .................................................................................... 105 
4.1 Study participants’ socio-demographic characteristics ......................................... 106 
4.2 Pre-intervention (baseline) results ........................................................................ 107 
4.2.1 The local food system and gender-based issues ............................................. 107 
4.2.1.1 Gender-driven access to productive resources ........................................ 108 
4.2.1.2 Gender differences in types and use of foods produced .......................... 110 
4.2.1.3 Women’s role in processing, preservation and storage of food ............... 114 
4.2.1.4 Gender-based roles in food distribution and intra-household allocation . 115 
4.2.1.5 Gender dynamics in household food provision, preparation and sharing 119 
4.2.1.6 Food shortage and coping strategies ........................................................ 121 
4.2.2 Community dietary knowledge, attitudes and sources of dietary information
 ................................................................................................................................. 124 
4.2.2.1 Knowledge of nutritive qualities of foods and sources of dietary 
information .......................................................................................................... 124 
4.2.3 Women’s dietary practices ............................................................................. 136 
4.2.3.1 Women’s eating pattern and dietary diversity ......................................... 136 
4.2.4 Drivers of women’s dietary choices ............................................................... 139 
4.2.5 Community readiness to address challenges to women’s diets ...................... 144 
4.2.5.1 Knowledge of women’s sub-optimal diets .............................................. 145 
4.2.5.2 Community knowledge of efforts addressing challenges to women’s diets
 ............................................................................................................................. 145 
4.2.5.3 Community climate regarding women’s sub-optimal diets ..................... 147 
4.2.5.4 Community leadership ............................................................................. 148 
4.2.5.5 Community resources .............................................................................. 149 
4.2.6 Description of implementation of intervention .............................................. 151 
 ix 
University of Ghana http://ugspace.ug.edu.gh
4.2.6.1 Exposure to intervention activities .......................................................... 152 
4.2.6.2 Modification and adaptation of initial plan of implementation ............... 153 
4.3 Post-intervention results ........................................................................................ 154 
4.3.1 Post intervention changes in community dietary knowledge and attitudes at 
Yilkpene .................................................................................................................. 154 
4.3.1.1 Improved dietary knowledge ................................................................... 155 
4.3.1.2 Changes in dietary perceptions at Yilkpene ............................................ 157 
4.3.1.3 Modifications to plant-based food restrictions at Yilkpene ..................... 161 
4.3.1.4 Changes in women’s dietary practices .................................................... 163 
4.3.2 Participants’ perceptions of intervention ........................................................ 168 
4.3.2.1 Messages recalled and practised at intervention meetings ...................... 169 
4.3.2.2 Messages practised by participants .......................................................... 172 
4.3.2.3 Benefits of intervention to participants .................................................... 173 
4.3.2.4 Issues associated with intervention messages .......................................... 175 
4.4 Summaries of results ............................................................................................. 176 
 
CHAPTER FIVE ............................................................................................................ 180 
5.0 DISCUSSION OF FINDINGS ................................................................................. 180 
5.1 Traditional norms and gender affect the local food system .................................. 182 
5.2 Community members’ dietary knowledge and attitudes have implications for 
quality of women’s diets ............................................................................................. 191 
5.3 Women’s dietary practices are likely to improve if the drivers of their dietary 
choices are addressed .................................................................................................. 198 
5.4 Community readiness and intervention coverage ................................................. 200 
 
CHAPTER SIX ............................................................................................................... 207 
6.0 SUMMARY, CONCLUSIONS AND RECOMMENDATIONS ............................ 207 
6.1 Summary of main findings .................................................................................... 207 
6.2 Implications of findings for women’s diets .......................................................... 209 
6.3 Conclusions ........................................................................................................... 209 
6.4 Recommendations ................................................................................................. 211 
6.5 Contributions to knowledge .................................................................................. 216 
6.6 Limitations of the study ........................................................................................ 218 
 
 x 
University of Ghana http://ugspace.ug.edu.gh
REFERENCES ............................................................................................................... 219 
 
APPENDICES ................................................................................................................ 230 
Appendix A: Key Informant Interview Guide on the Food System ........................... 230 
Appendix B: Key informant and FGD interview guide on community perceptions of 
intervention activities .................................................................................................. 233 
Appendix C: Assessment of community readiness to address women’s dietary 
challenges .................................................................................................................... 234 
Appendix D:  Qualitative 24-hour recall guide ........................................................... 236 
Appendix E: In-depth interview guide on food ownership and utilization at the 
household level ........................................................................................................... 238 
Appendix F: Focus group discussion guide (Community men and women) .............. 239 
Appendix G: Observations of women’s time use (household level: from 6am to 6pm)
 ..................................................................................................................................... 241 
Appendix H: Observational Checklist (Community level) ......................................... 242 
Appendix I: Participant’s Consent Form .................................................................... 243 
Appendix J: Child Assent Form .................................................................................. 245 
Appendix K: Parental Consent Form .......................................................................... 247 
VOLUNTEER AGREEMENT ....................................................................................... 248 
Appendix L: Intended Action plan for implementing attitudinal and behaviour change 
to improving women’s diets ........................................................................................ 250 
Appendix M: Implemented action plan for attitudinal and behaviour change to 
improve women’s diet ................................................................................................. 253 
Appendix N: Messages disseminated ......................................................................... 259 
Appendix O: Cooking demonstration sessions ........................................................... 264 
Appendix P: Certificate of ethical clearance from Noguchi Memorial Institute for 
Medical Research ........................................................................................................ 268 
Appendix Q: Certificate of ethical clearance from Noguchi Memorial Institute for 
Medical Research (Renewed) ..................................................................................... 269 
 
 
 xi 
University of Ghana http://ugspace.ug.edu.gh
LIST OF FIGURES 
Figure 1.1: Adapted framework from the Social Cognitive Theory (Bandura, 1997). ...... 7 
Figure 1.2: Pathway for attitudinal and behaviour change on women’s diets .................. 10 
Figure 3.1: Map of Savelgu-Nanton District .................................................................... 66 
Figure 3.2: Study design ................................................................................................... 74 
Figure 3.3: Illustration of quasi-experimental design for implementing attitudinal and 
BCC intervention ............................................................................................ 87 
Figure 4.1: Drivers of women’s dietary choices ............................................................. 139 
Figure 4.2: Food less preferred by Yilkpene and Kpachilo at endline ........................... 159 
Figure 4.3: Consumption of groundnut .......................................................................... 166 
Figure 4.4: Consumption of legumes.............................................................................. 167 
Figure 4.5: Consumption of animal-source proteins ...................................................... 167 
Figure 4.6: consumption of fresh green-leafy vegetables ............................................... 168 
Figure 4.7: Messages recalled by men and women from intervention ........................... 171 
  
 
 
 
 
 
 
 
 
 
 xii 
University of Ghana http://ugspace.ug.edu.gh
LIST OF TABLES 
Table 2.1: Comparative schedule of the Nine and Ten food group classifications .......... 51 
Table 2.2: Description of the nine stages of community readiness .................................. 60 
Table 3.1. Identified challenges to women’s optimal diets .............................................. 79 
Table 3.2. Challenges, current state of behaviour, interventions and their respective key 
messages ........................................................................................................... 82 
Table 3.3: Channels for delivering  intervention .............................................................. 84 
Table 3.4 Data collection methods, tools, phases in which used and information collected
 .......................................................................................................................... 90 
Table 3.5:  Methods, participant categories, number of interviews, samples of 
participants in baseline endline and community readiness assessment (CRA) 
phases of the study ........................................................................................... 94 
Table 3.6: Number of interviews by methods at baseline, CRA and endline ................... 95 
Table 3.7: Themes and their sub-themes .......................................................................... 98 
Table 4.1: Participants’ socio-demographic characteristics ........................................... 106 
Table 4. 2: Types of foods produced .............................................................................. 111 
Table 4.3: Foods preferred and less preferred by communities at baseline .................... 129 
Table 4. 4: Food taboos and restrictions at baseline at Kpachilo and Yilkpene ............. 133 
Table 4. 5: Frequency of meals at baseline ..................................................................... 137 
Table 4. 6: Minimum Dietary Diversity of Women (MDD-W) at Yilkpene and Kpachilo 
in May-June and November 2016 at baseline ......................................................... 138 
Table 4. 7: Community readiness scores obtained from key informants ....................... 145 
Table 4. 8: Frequency of participation in group meetings .............................................. 152 
Table 4. 9: Coverage of group meetings during implementation ................................... 153 
Table 4. 10: Frequency of meals at endline (May-June and November, 2017) .............. 164 
Table 4. 11: Minimum Dietary Diversity of women at baseline and endline................. 165 
Table 4. 12: Matrix of messages recalled and practised ................................................. 170 
Table 4. 13a: Summary of results –objective 1- baseline and endline ........................... 176 
 Table 4.13b: Summary of results- Objective 2 -baseline and endline …………..……177 
Table 4.13c: Summary of results - objective 3- baseline and endline……………..…..178 
Table 4.13d: Summary of results –objective 4…………………………………..…….178 
Table 4.13e: Summary of results –objective 5…………………………….…….…….178 
Table 4.13f: Summary of results – Community perceptions of intervention………….179 
 xiii 
University of Ghana http://ugspace.ug.edu.gh
GLOSSARY 
Definitions of commonly used terms and concepts 
Term/concept Definition 
Attitude The tendency to respond or act in a certain way to 
an event or situation based on one’s belief or 
perception 
Belief An idea considered true or real 
Community-based participatory Research conducted in collaboration with the 
action research people mostly affected by an issue with the goal of 
devising strategies to address it  
Community strategies Community activities or interventions aimed at 
addressing a problem 
Diets The sum or the different kinds of food consumed 
by an individual or a population or a combination 
of foods with a series of nutrients reflecting the 
psychological and cultural attitudes of individuals 
and populations (Etievant et al., 2010). 
Dietary practices Actions or practices of people in respect of their 
food sources and choices; the context of food; 
preparation, variety, quality and quantity of foods 
consumed as well as pattern of meals 
Formative research Research that occurs before and during the 
implementation of a programme to understand and 
facilitate the creation of activities to meet the 
specific needs of the population 
Knowledge Right information about an issue based on a 
certain standard 
Women’s minimum dietary An assessment of the quality of women’s diets by 
diversity  using the number of food groups consumed in their 
meals - FAO (2012) recommends that women who 
consume a minimum of five food groups in their 
meals can be classified as meeting their nutrient 
requirements for the day 
Perception Awareness of something and how one interprets it 
or one’s opinion about something 
Quasi-experimental design A kind of intervention that seeks to test an 
association between two variables measured by 
pre-specified indicators- assignment of participants 
to intervention or comparison is based on selection 
by the researcher rather than randomization as in 
the case of Randomized Control Trials (RCT). 
Socio-cultural People’s relations and interactions with the 
elements of culture 
Women of reproductive age Women aged 15 -49 years 
 
 xiv 
University of Ghana http://ugspace.ug.edu.gh
LIST OF ABBREVIATIONS AND ACRONYMS 
Abbreviation and acronyms Full versions 
ADRA Adventist Development and Relief Agency 
ADVANCE Agricultural Development and Value Chain Enhancement 
AGRA Alliance for a Green Africa 
ANC Ante-natal care 
BCC Behaviour Change Communication 
BMI Body Mass Index 
CBE Complimentary Basic Education 
CHPS Community-owned Health Planning and Service 
CRA Community readiness assessment 
CRM Community readiness model 
DDS Dietary diversity score 
DHS Demographic and health survey 
FAO Food and Agricultural Organization  
FGD Focus Group Discussion 
GDHS Ghana demographic and health survey 
GHS Ghana Health Service 
GOG Government of Ghana 
GSS Ghana Statistical Service 
HIV/AIDS Human-Immuno Deficiency/Acquired Immune-Deficiency 
Syndrome 
IPA Innovation for Poverty Action 
KII Key –informant interview 
LDDS Low dietary diversity score 
MDD Medium dietary diversity  
 xv 
University of Ghana http://ugspace.ug.edu.gh
MDD Minimum dietary diversity 
MDD-W Minimum dietary diversity -women 
MDG Millennium development goals  
NGO Non-governmental Organization 
NMCCSP National malaria control for child survival project 
NMIMR Noguchi Memorial Institute for Medical Research 
PNC Post- natal care 
RAINS Regional Advisory Information and Network Systems 
RCT Randomized control trials 
RING Resilience in Northern Ghana 
SCT Social cognitive theory 
TBA Traditional birth attendant 
TZ Tuo zaafi 
UNICEF United Nations International Children’s Emergency Fund 
USAID United States Agency for International Development 
WDD Women’s dietary diversity 
WFP World Food Programme 
WHO World Health Organization 
WIFA Women In Food and Agriculture 
WRA Women of reproductive age 
 
 
 
 
 
 xvi 
University of Ghana http://ugspace.ug.edu.gh
PREFACE- A PERSONAL REFLECTION 
The researcher has a social science background and works in a medical research 
institution. As a research assistant, her mandate was to assist in fieldwork activities on a 
number of projects including infant and young child nutrition (IYCN), food security in 
deprived urban communities and maternal mortality studies. In all these projects, she was 
assigned to collect qualitative data using multiple qualitative methods, which helped in 
her understanding of socio-cultural influences on behaviour outcomes. 
The researcher’s interest in the area of public health intervention stemmed from one of 
the projects she worked on, in which a research participant needed an intervention 
addressing her child’s malnourished status. It was at this point she felt the burden and 
need for the research community to empower research participants by making known to 
them their research findings even if the former do not intend to implement any 
intervention. In the opinion of the researcher, therefore, research findings must be made 
known to research participants and efforts made to help communities through the 
building of their capacity to address some of the challenges confronting them. 
The researcher’s background in nutrition-related research, therefore, informed her 
decision to undertake this study on the dietary practices of the rural woman and to design 
a context-specific attitudinal and behaviour change intervention addressing socio-cultural 
practices that are detrimental to women’s dietary quality. 
 
 
 
 
 
 xvii 
University of Ghana http://ugspace.ug.edu.gh
ORGANIZATION OF THE THESIS 
This study is presented in six chapters. Chapter One is the introductory chapter, which 
gives a background to the study and contextualizes the problem related to women’s diets. 
The theoretical underpinnings and justification for the study are highlighted together 
with the aim and objectives as well as the research questions guiding the study. 
Chapter Two situates the study within existing and related literature on women’s dietary 
situation and interventions aimed at addressing it in low-income countries and Ghana 
specifically. Additionally, a review of some methodological approaches and theories 
used in researches related to such interventions are reviewed. 
The third chapter presents the methods for the study. The study location is presented 
alongside the study design and the study process. The data collection methods, 
instruments and variables investigated are presented with the analysis plan. 
The fourth chapter presents the results of the study structured on the basis of the 
objectives. Participants’ socio-demographic characteristics, results of pre-intervention 
(baseline study), community readiness assessment and implementation of intervention 
are presented.  The last section presents results of the effect of the intervention. 
Chapter Five presents a discussion of the main findings based on the objectives and the 
last chapter presents the summary of findings, conclusions and recommendations drawn 
from the findings. The theoretical, methodological and policy contributions, areas for 
further research as well as the limitations of the study are presented in this chapter. There 
are also appendices including the instruments used for the study. 
 xviii 
University of Ghana http://ugspace.ug.edu.gh
CHAPTER ONE 
1.0 INTRODUCTION 
1.1 Background to the study  
Undernutrition among women of reproductive age (WRA) is an important public health 
challenge in low-income countries. Women’s undernutrition is particularly important in 
sub-Saharan Africa and South/South-East Asia, where at least 10% of WRA are reported 
to be under-nourished (USAID, 2012). In Ghana, an estimated 6% of women are 
currently under-nourished (GSS, GHS, & DHS, 2015; GSS, GHS, & ICF, 2015a; GSS, 
GHS, & Macro, 2009). Anaemia, which is linked with sub-optimal dietary quality and 
contributes to poor nutritional status, remains highly prevalent (42%) among women 
(GSS, GHS, & ICF, 2015b). The evidence demonstrates that these outcomes of 
nutritional status are more prevalent in Northern Region of Ghana, where 11. 2% of 
women are under-nourished and 47.5% are anaemic (GSS, GHS, et al., 2015b; GSS et 
al., 2009). 
 
An important determinant of women’s nutritional status is their diets (Global Panel on 
Agriculture and Food System for Nutrition, 2016). Diet, defined as the pattern of 
consumption involving the combinations of foods that reflect the cultural attitudes of 
people (Etiévant et al., 2010) is linked with the food system, which comprises all 
processes involved in keeping populations fed (Global Panel on Agriculture and Food 
System for Nutrition, 2016). The system is important for providing good-quality diets to 
meet the nutritional needs of communities (Kuhnlein & Pelto, 2014) and consists of all 
activities and techniques for production, acquisition, processing, preservation, storage, 
distribution and consumption of food (Kuhnlein & Pelto, 2014; Baden, Green, Otoo-
Oyortey, & Peasgood, 1994). In Ghana, women play active roles in all aspects of the 
 1 
University of Ghana http://ugspace.ug.edu.gh
system; women contribute about 47% of the agricultural labour force and account for 
about 70% of total crop production (Braimah, Dukuh, Oppong-Sekyere, & Momori, 
2014; Mohammed, Solikua, & Mansega, 2013). In spite of their vital contributions to 
food security, women’s diets and their implications for their health and reproductive 
health outcomes are often sub-optimal (USAID, 2012). 
 
Women in northern Ghana have a unique experience of food and nutrition insecurity. It 
is estimated that 60% of households in northern Ghana consume inadequate quantities of 
food and 20% consume poor quality diets consisting mainly of staples, vegetables and 
oils (Saaka, Oladele, Larbi, & Zeledon, 2017; WFP, 2012). Seasonal effects on food 
access have been explained as a major contributory factor for the food insecurity 
situation in the region (WFP, 2012). A report by “Feed the Future project” further 
estimated that 40.5% of women in the three northern regions consume low-quality diets -  
an indication of low diversity in the diet (Zereyesus et al. 2014). 
 
WRA’s physiological needs make them more vulnerable to poor dietary intake. This 
vulnerability arises because their need for nutrient intake increases during adolescence, 
pregnancy and lactation but, as compared to men, women’s energy intake is low 
(Galbete et al., 2017; FAO, 2009), indicating the food insecurity situation among them. 
When women are unable to meet their increased dietary needs, their dietary status is 
compromised (USAID, 2012). 
 
Women’s sub-optimal dietary practices including inadequate frequency of meals and 
poor qualities of diets are the commonest causes of their under-nourished status(USAID, 
2012). Poor quality of women’s diets is linked with multiple factors including personal 
 2 
University of Ghana http://ugspace.ug.edu.gh
and environmental factors, agricultural practices, national policies, food security, socio-
cultural and economic factors (Martínez Pérez & Pascual García, 2013; Haileslassie, 
Mulugeta, & Girma, 2013; Gadegbeku, Wayo, Badu, Nukpe, & Okai, 2013; Koryo-
Dabrah, Nti, & Adanu, 2012; Puoane, Matwa, Bradley, & Hughes, 2006; Ogunjuyigbe & 
Ojofetimi, 2006). 
 
Women are constrained by socio-cultural factors such as their social status and decision-
making power over household food resources and control over them (Braimah et al., 
2014). Cultural definitions of food, what constitutes food, what is edible, food taboos 
and food restrictions, food preferences and preferential food allocations, knowledge and 
perceptions about the health benefits of food and ease of acquiring food constitute some 
of the social norms affecting women’s diets (Kuhnlein & Pelto, 2014). Other socio-
cultural factors include the value placed on food, their perceptions of the relationship 
between food and body-size and the effects of urbanization (Puoane et al. 2006). 
 
Women’s diets have implications for their own well-being, productivity and health 
including their reproductive health and those of their children, community and nation 
(USAID, 2012; FAO, 2012a; Lartey, 2008; Johnson-Welch, 1999). Women with 
improved diets contribute positively to the development of society. Improvements in 
women’s nutrition could further reduce hunger, mal-nutrition and maternal and infant 
morbidities and mortalities (Ransom & Elder, 2003). Effective nutrition-related 
interventions aimed at addressing challenges to women’s nutrition and diets include 
nutrient supplementation and food-based strategies (Black et al., 2013; MacDonald, 
Neequaye, Namarika, & Yiannakis, 2009). In Ghana, these strategies have included 
fertility control, preventing infections, micro-nutrient supplementation for pregnant 
 3 
University of Ghana http://ugspace.ug.edu.gh
women (iron and folate), improving agricultural productivity, food fortification, diet 
diversification as well as nutrition and health education (Zereyesus, Ross, Amanor-
Boadu, & Dalton, 2014; GOG, 2013). Interventions have focused on addressing the 
direct causes of maternal undernutrition with emphasis on changing the individual’s 
behaviour but less attention has been paid to the social context of nutrition, which has 
been identified (Aikins, 2014; Delormier, Frohlich, & Potvin, 2009; Lartey, 2008). 
 
Delormier et al. (2009) have defined the social context of nutrition as the social nature of 
food and eating which is influenced by culture and experiences. The need for 
interventions to address societal norms linked with dietary behaviour has, therefore, been 
suggested (Delormier et al., 2009). Participatory interventions have demonstrated 
positive outcomes particularly for child nutrition in Ghana (Colecraft et al., 2012). Other 
studies have also found that interventions targeted at poverty, education and 
empowerment of women with productive resources contribute to improved maternal and 
child nutrition (FAO, 2012b; Underwood & Galal, 2004). 
 
Slow progress in addressing women’s undernutrition including their sub-optimal diets is 
partly attributable to limited attention to interventions that focus on the social context of 
women’s diets. Even though community-based participatory approaches have been 
linked with improved dietary practices among children aged two to five years (Colecraft 
et al., 2012; Colecraft et al., 2006), there are currently few interventions in Ghana aimed 
at addressing socio-cultural barriers to women’s diets through the use of community-
based participatory approaches. Between 2004 and 2009, Colecraft and her colleagues 
conducted a participatory formative study to identify barriers to the consumption of 
animal-source foods among children aged 2 to 5 years old in Ghana. Results of this 
 4 
University of Ghana http://ugspace.ug.edu.gh
initial assessment were fed into the design and implementation of an integrated 
intervention using participatory processes. The intervention activities involved micro-
credit, entrepreneurship training and nutrition education and the beneficiaries of the 
intervention were caregivers (mostly women) with children two to five years old.  The 
aim of the intervention activities was to improve and empower caregivers’ access to and 
use of animal-source foods in children’s diets. The outcomes of the intervention included 
improvements in caregivers’ knowledge on child nutrition and their incomes and 
children’s intake of animal-source foods. Similar participatory studies have yet to be 
conducted to address barriers to women’s optimal diets. 
 
1.2 Statement of the problem 
Malnutrition is common among women in northern Ghana and this is partly because 
almost half of women in the region eat diets with low diversity (Zereyesus et al., 2014). 
In Savelugu-Nanton District, the study location, the prevalence rate of anaemia among 
WRA is very high (58%) (GHS, 2018) confirming the poor dietary situation in the 
district and the need for an intervention addressing this challenge. 
 
Socio-cultural factors limit women’s ability to access adequate quantities and qualities of 
food (Martínez Pérez & Pascual García, 2013). Existing interventions hardly focus on 
socio-cultural barriers to women’s diets but on factors such as the immediate causes of 
malnutrition including access to and intake of a variety of foods (Lartey, 2008).  The 
problem for the study, thus, is that in northern Ghana, there are unsolved questions 
relating to women’s diets as follows: first, what are the socio-cultural influences on 
WRA’s diets in northern Ghana, second, what is the readiness of communities to address 
 5 
University of Ghana http://ugspace.ug.edu.gh
these socio-cultural barriers and, finally, what strategies can address these socio-cultural 
challenges? 
 
1.3 Theoretical framework 
Interventions influencing health behaviour are most likely to benefit their target audience 
when they are guided by appropriate theories of health behaviour (Glanz, Rimer, & 
Viswanath, 2008). Generally, behaviour change theories propose that, for an intervention 
to be successful, the determinants of the undesirable behaviour must be assessed to 
understand why people do the things they do and to use this understanding to develop an 
intervention targeting the undesired behaviour (Bandura, 1997). A number of health 
behaviour change theories have been propounded (Glanz et al., 2008); while some of 
these models focus on changing individual behaviour, others are inter-personal, focusing 
on changing group behaviour and bringing about social change. Examples of individual 
behaviour change theories include the health belief model, theory of reasoned action, 
theory of planned behaviour, transtheoretical model (theory of change) and the 
precaution adoption process model. The social cognitive theory and the theories of 
organizational change are examples of inter-personal theories. Among these theories, the 
social cognitive theory was the choice for the study mainly because it addresses inter-
personal health issues confronting communities or groups of people. 
 
The Social Cognitive Theory (SCT) (Bandura, 1997) was the main theory underpinning 
the study. The theory explains that human behaviour is a result of personal, social and 
physical (environment) factors (McAlister, Perry, & Parcel, 2008). The theory posits that 
these three interacting factors influence one another to bring about social change through 
human capacity for learning and adaptation. The theory further emphasizes that even 
 6 
University of Ghana http://ugspace.ug.edu.gh
though the social and physical environments contribute to shaping people’s cognition 
and behaviour, individuals have the ability to also influence the environment to bring 
about desirable social change through a collective action of the group. Figure1.1 
illustrates the dynamic interaction between personal as well as environmental factors and 
behaviour as proposed by the Social Cognitive Theory. 
 
Figure 1.1: Adapted framework from the Social Cognitive Theory (Bandura, 1997) 
 
The personal factors include sex, age, education, physiological status, awareness, 
knowledge, attitudes and self-efficacy. The social factors include group or social norms 
that determine what is acceptable behaviour such as food taboos and the physical factors 
include the physical resources and technology the community is endowed with. 
 
In this study, the three interacting constructs of the Social Cognitive Theory were 
operationalized by distinguishing between modifiable and non-modifiable personal 
factors. The non-modifiable factors, which the study did not intend to change, included 
age, sex, marital status, education and income. The modifiable factors were community 
 7 
University of Ghana http://ugspace.ug.edu.gh
knowledge and perceptions on diets and self-efficacy, which the intervention tried to 
influence. The physical environment comprised the food system, which included inputs 
for food production; kinds of foods produced; processing, preservation and storage; 
distribution and consumption of food at the household level.  
 
The social norms included women’s control over household food resources, food taboos 
and restrictions and the preferential food allocation system. Behaviour was 
operationalized to include preparation of food, access to food and dietary quality; 
however, because behaviour usually takes a long time to change, the study concentrated 
on improved dietary knowledge and perceptions as well as modification of food taboos 
as the immediate outcomes with the hope that the improved outcomes would eventually 
lead to improved dietary quality as depicted in the broken arrows in Figure 1.2. 
 
The concepts self-efficacy, collective efficacy, observational learning and facilitation of 
the social cognitive theory were used in this study to facilitate change in the intervention 
community. Self- efficacy is the belief about one’s ability to change or bring about a 
desirable behavioural outcome. The theory maintains that if human behaviour is a 
product of what people have learnt, their perceptions of the environment and their 
intellectual and physical capacities, then behaviour can equally be changed or modified 
through new learning experiences, guidance on people’s perceptions and support for the 
development of their capacities.  
 
Increasing individual self-efficacy was achieved through two mediating strategies: 
observational learning and verbal persuasion. Observational learning is the process of 
learning to perform a new desired action by being exposed to inter-personal display of 
 8 
University of Ghana http://ugspace.ug.edu.gh
that behaviour while verbal persuasion is a form of encouragement or advocacy to 
motivate a person or group to adopt a new behaviour. Facilitation implies providing tools 
and other resources that make the new behaviour easier to perform. These concepts were 
operationalized through the use of nutrition and dietary education to members of the 
community. Collective efficacy is the belief that the group has the ability to use 
concerted action to bring about desired change for the benefit of the whole group 
(McAlister et al., 2008). Collective efficacy was enhanced through the participatory 
nature of the intervention whereby all stakeholders and the community at large were 
involved in the intervention process. 
 
The SCT was deemed appropriate in informing the current study because its focus was 
on addressing socio-cultural barriers to women’s optimal diets. Socio- cultural practices 
are aspects of group/social norms that influence and shape women’s diets. There was, 
therefore, the need to employ the concepts self-efficacy and collective efficacy to 
address issues relating to knowledge and perceptions and at the same time get the 
community or group involved in a concerted effort to begin to think about and possibly 
address some of the socio-cultural barriers to women’s diets.  
 
The SCT does not provide a pathway for determining community readiness for change or 
modifying socio-cultural influences that constitute barriers to women’s optimal diets. 
The community readiness model (CRM) was, therefore, used as a tool to determine the 
intervention community’s readiness to implement strategies to address women’s dietary 
challenges. The CRM is a participatory tool used to assess community knowledge, 
attitudes, efforts and resources pertaining to specific community issues that need to be 
addressed (Edwards, Jumper-Thurman, Plested, Oetting, & Swanson, 2000). There are 
 9 
University of Ghana http://ugspace.ug.edu.gh
five key dimensions for measuring community readiness for action: first, community 
knowledge about the issues to be addressed; second, community efforts and knowledge 
of efforts addressing the problem; third, leadership within the community; fourth, 
community climate and, fifth, resources related to the issues. These dimensions are 
scored against nine stages of community development (Stanley, 2014).  
 
1.4 Conceptual pathway of change for socio-cultural intervention on women’s diets 
Informed by the social cognitive theory and using the community readiness model as a 
tool, a conceptual pathway to attitudinal and behaviour change was developed to address 
identified socio-cultural barriers to women’s diets in one of the study communities. 
Presented in Figure 1.2 is an illustration of the pathway of change. 
	
Community Assessment Development of Implementation of 
assessment  of community intervention with intervention addressing 
readiness for active participation identified socio-cultural 
change of community barriers 
Improved knowledge 
Changed perceptions 
Modified beliefs about 
diet 
Improved 
women’s diet 
quality 
 
Figure 1.2: Pathway for attitudinal and behaviour change on women’s diets 
 
 10 
University of Ghana http://ugspace.ug.edu.gh
Figure 1.2 demonstrates that, to introduce an intervention addressing barriers to women’s 
dietary quality, the determinants of women’s diets must be assessed and understood. In 
this study, the personal, social (and cultural) and environmental (physical) factors were 
assessed. The personal determinants included participants’ socio-demographic 
characteristics as well as their dietary knowledge and perceptions. The socio-cultural 
factors included issues of women’s empowerment such as their control over productive 
and food resources as well as beliefs and discriminatory food practices. The 
environmental (physical) factors included availability of and access to productive 
resources such as land and labour, climate and seasonality as well as methods of 
processing, preservation, storage and preparation of food. 
 
The assessment of the determinants of women’s diets was followed by an assessment of 
the intervention community’s readiness to change. Because the intervention was 
participatory, community and stakeholder engagement processes were utilized to assess 
its feasibility for addressing identified socio-cultural barriers to women’s diets. The 
intervention was then designed with the active participation of members of the 
community and implemented. It was envisaged that improving community members’ 
knowledge and perceptions as well as modifying their beliefs and attitudes regarding 
women’s diets could lead to changes in behaviour that would enhance women’s capacity 
to observe improved dietary practices and, ultimately, improved dietary quality 
(indicated in Figure 1.2 with broken arrows). 
 
1.5 Justification of the study 
Improving women’s diets can contribute to addressing the high burden of hunger and 
malnutrition (FAO, 2012b). Women’s role as producers of food for their families and 
reproduction of future generations makes it important for their dietary needs to be 
 11 
University of Ghana http://ugspace.ug.edu.gh
provided (USAID, 2012; Johnson-Welch, 1999). In Northern Region, women’s 
nutritional indicators including their dietary quality lag behind those for the other 
regions. Exploring the basic factors would yield new evidence on socio-cultural barriers 
to women’s optimal diets in the communities studied and others with similar 
characteristics.  
 
Second, in Ghana, the use of the community readiness model in designing interventions 
addressing dietary challenges has not been reported in the scientific literature. This 
model, however, has been used in some countries including the United States of 
America, South Africa and Liberia in HIV/AIDS and obesity related interventions 
(Pradeillis, Rousham, Norris, Kesten, & Griffiths, 2016; Kostadinov, Daniel, Stanley, 
Gancia, & Cargo, 2015; Findholt, 2007) to demonstrate the level of readiness of 
communities to address these public health challenges. Results of this study would 
therefore further demonstrate the relevance of the model and indicate the level of 
readiness of communities to address socio-cultural influences on women’s diets. 
 
The study would further add to the existing body of knowledge on the kinds of 
interventions that are relevant in addressing socio-cultural-related challenges to women’s 
diets. For example, working with members of the community to give women the 
autonomy to make household food choices could contribute to improve women’s 
contribution to household food security and their own dietary needs. Additionally, 
working with members of the community to modify taboos or beliefs that restricted some 
categories of women from consuming certain foods would lead to improvements in the 
quality of their diets. Ghana Health Service and other community-based organizations 
working in nutrition-related activities in rural settings would utilize the evidence from 
the study. 
 12 
University of Ghana http://ugspace.ug.edu.gh
1.6 Objectives of the study  
1.6.1 General objective  
This study aimed to describe barriers to consuming nutritious diets among WRA and to 
test the feasibility of an attitudinal and social-behaviour change communication 
intervention to address socio-cultural barriers to optimal diets among women at Yilkpene 
in Savelugu-Nanton District of Ghana. 
 
1.6.2 Specific objectives 
The specific objectives of the study were: 
1. To describe women’s interaction with the food system in Savelugu-Nanton District 
2. To assess dietary knowledge and attitudes (perceptions and beliefs) at Yilkpene and 
Kpachilo in Savelugu –Nanton District 
3. To describe women’s dietary practices at Yilkpene and Kpachilo 
4. To describe the drivers of the dietary choices among women at Yilkpene and Kpachilo  
5. To assess community members’ readiness to modify socio-cultural barriers to 
women’s optimal diets at Yilkpene  
6. To assess the post-intervention change in community dietary knowledge and attitudes 
(perceptions and beliefs) at Yilkpene.  
 
1.7 Research questions  
The study sought to answer the following research questions to achieve its objectives:  
What is women’s role in the local food system and how does it influence their diets in 
Savelugu-Nanton District? 
 
 13 
University of Ghana http://ugspace.ug.edu.gh
What personal, household and community-wide beliefs and practices influence women’s 
dietary behaviour in the district? 
 
Who are the change agents in the community and what is their capacity to modify 
community norms that affect women’s diets in the community? 
 
Are there any changes in community dietary knowledge, attitudes and practices after the 
social-attitudinal and behaviour change intervention?  
 
 
 
 
 
 
 
 
 14 
University of Ghana http://ugspace.ug.edu.gh
CHAPTER TWO 
2.0 LITERATURE REVIEW 
This chapter contains six parts: dietary practices of women of reproductive age, 
determinants of their dietary practices, interventions addressing barriers to optimal diets, 
methodological and theoretical considerations and gaps identified in the review. Each 
part is sub-divided into general information and specific information about Ghana. The 
two main types of literature reviewed were theoretical and empirical. 
 
2.1 Women’s dietary practices  
Women of reproductive age are females aged 15-49 years. They comprise pregnant 
women, lactating mothers as well as women who are neither pregnant nor lactating 
(Zereyesus et al., 2014). In Ghana, women constitute 51.7% of the total population 
(GSS, 2012).  
 
A diet is a pattern of consumption involving the combinations of foods reflecting the 
cultural attitudes of people (Etiévant et al., 2010). Dietary behaviour describes the 
characteristics of usual dietary intake including types of food, quality (measured as 
variety of food groups consumed over a period of time and meal composition), frequency 
of consumption and quantity consumed in a given period. Dietary behaviour is 
influenced by multiple factors that drive food choices including the food system- that is, 
food production and supply processes, access, sources of food, methods of food 
preparation and meal time practices and food preferences (Global Panel on Agriculture 
and Food System for Nutrition, 2016; Etiévant et al., 2010).  
 15 
University of Ghana http://ugspace.ug.edu.gh
Dietary behaviour is an important indicator for public health intervention because it is 
often measured as an outcome of how social, physical environmental and personal 
factors interact to affect nutritional health and status (Stok et al., 2016; Glanz et al., 
2008). Women’s dietary quality and intake have been reported to be sub-optimal 
globally but particularly in low-income countries (Martin-Prevel et al., 2017; Amugsi, 
Lartey, Kimani-Murag, & Mberu, 2016; Global Panel on Agriculture and Food System 
for Nutrition, 2016). The Global Panel on Agriculture and Food Systems for Nutrition 
(2016) indicates that globally, about three billion people consume low-quality diets with 
important implications for multiple forms of malnutrition with negative consequences for 
global health.  
 
The authors attribute the low-quality diets to rapid population growth, climate change 
and urbanization. In addition, women’s low dietary quality and intake have been 
attributed partly to the physiological demands of adolescence, pregnancy and lactation 
(Underwood and Galal, 2004). Pregnancy increases the nutrient needs of women up to 
about 300 extra calories a day compared with non-pregnant women of similar age but 
most WRAs are unable to meet this increased nutrient demand (Ransom & Elder, 2003). 
As a result, micronutrient mal-nutrition, also known as “hidden hunger”, is very common 
among WRA (Ramakrishnan, 2002). Women’s access to adequate and nutritious foods 
as well as their dietary practices also determines their dietary intake and dietary quality 
(Underwood & Galal, 2004; Gittelsohn & Vastine, 2003).  
  
2.1.1 WRA’s dietary practices in Ghana 
Women’s dietary situation in Ghana has been reported to be sub-optimal (Armar-
Klemesu et al., 2014; Kobati, Lartey, Marquis, Colecraft, & Butler, 2012; Koryo-Dabrah 
 16 
University of Ghana http://ugspace.ug.edu.gh
et al., 2012). The majority of non-pregnant and non-lactating women in both the coastal 
and guinea savannah zones have low energy intake; the deficiency is even greater among 
women in the latter zone. They also consume insufficient quantities of protein, vitamin 
A, vitamin C and niacin (Seshadri, 2001). The sub-optimal consumption of nutrients is a 
result of women’s dietary practices throughout the life-cycle (Nana & Zema, 2018). 
 
Women’s dietary pattern in Ghana is influenced by socio-demographic characteristics 
such as residence, age, income and social status. While urban women were more likely 
to consume foods prepared from grains, proteins, fat and oils, rural women were more 
likely to consume foods prepared from legumes, roots and tubers and vitamin A-rich 
foods. In addition, younger women or adolescents (15-19 years) consume more sugar-
based foods and less food prepared from milk, fats and oils and butter. Fewer women 
from poorer socio-economic backgrounds tended to consume foods from all food groups 
(FAO, 2009).  
 
In Brong Ahafo Region, women’s diets were more diverse, dominated by maize, roots 
and tuber-based staples and consumed with soups, stews and sauces. On the contrary, in 
Northern Region, women’s diets were reported to be monotonous and dominated by 
grain-based foods with low consumption of fruits and vegetables(Armar-Klemesu et al., 
2014). Saaka et al., 2017 confirmed that the main staple foods in northern Ghana 
including Savelugu-Nanton comprise maize, sorghum, millet and yam, which are 
predominantly energy-giving foods. In their study, 60% of households in the region were 
classified as food insecure and therefore, less likely to consume fleshy foods, dairy 
products and vegetables among others (Saaka et al., 2017). As a result of the poor 
diversity in their diets, women in Northern Region tended to be more under-weight than 
 17 
University of Ghana http://ugspace.ug.edu.gh
their counterparts in Brong-Ahafo Region (Amoateng, Doegah, & Udomboso, 2017; 
Armar-Klemesu et al., 2014). 
 
2.2 Consequences of inadequate dietary intake 
The most common micronutrient deficiencies among WRA include iron, vitamin A, 
iodine, folate and zinc (Harika et al., 2017). Inadequate intake of food leads to increased 
risk of a mother becoming under-nourished, and this, in turn, can lead to low weight- 
gain during pregnancy. Foetal malnutrition can lead to low birth-weight and pre-term 
births. Studies have shown that the more variety there is in an individual’s diet, the better 
the quality of the food in terms of its micronutrient adequacy (Saaka, 2012). In a study in 
northern Ghana, women who consumed low-diversity diets had babies with lower birth-
weights than mothers who consumed high-diversity diets (Saaka, 2012). 
 
In Ghana, the prevalence rate of low birth-weight is estimated at 11% (FAO, 2009; GSS 
et al., 2009), an indication of women’s poor dietary intake. Women’s poor dietary status 
has negative effects on themselves and their unborn babies. Children born with low 
birth-weight are at great risk of increased mortality and morbidity while those who 
survive into adulthood are likely to enter a lifetime of reduced developmental potential 
and increased risk of chronic diseases. Stunted growth in infants also contributes to 
reduced mental capacity, stunted adolescence and, when the child is female, she stands a 
higher risk of experiencing obstetric complications during pregnancy owing to having a 
small pelvis, and the vicious cycle can continue unless broken by an appropriate 
intervention (Levay, Mumtaz, Rashid, & Willows, 2013b; Underwood & Galal, 2004). 
The effects of women’s undernutrition are higher maternal mortality, impaired mental 
development and increased risk of maternal morbidity including chronic adult diseases 
 18 
University of Ghana http://ugspace.ug.edu.gh
(Levay, Mumtaz, Rashid, & Willows, 2013a; Ransom & Elder, 2003). The FAO‘s 
nutrition profile of Ghana (FAO, 2009) identified the three northern regions as those 
with the most serious burden of undernutrition and food insecurity. The poor state of 
food security in the these three regions was supported by a report of the World Food 
Programme and Saaka and hid colleagues who confirmed that while 20% of households 
in northern Ghana consumed low quality diets, 60% consumed inadequate quantities of 
food (Saaka et al., 2017; WFP, 2012).  
 
2.3 Determinants of women’s dietary practices 
Three sets of determinants of women’s dietary practices have been identified. These 
determinants are personal, socio-cultural and environmental factors.  These determinants 
contribute to the creation of barriers that make the consumption of healthy diets difficult 
for women and other population groups. A good understanding of these determinants is, 
therefore a prerequisite for designing effective social and behaviour change interventions 
addressing women’s dietary challenges (Farahmand, Tehrani, Amiri, & Azizi, 2012).  
 
2.3.1 Personal Factors 
Personal factors, that characterize the individual himself or herself, include knowledge, 
beliefs, perceptions and attitudes towards food and nutrition-related issues; educational 
attainment as well as their food preferences and physiological statuses. These factors 
influence their dietary quality and intake (WFP 2005). Food is often characterized based 
on perceived quality and appropriateness for consumption (Gittelsohn, Haberle, Vastine, 
Dyckman, & Palafox, 2003). Knowledge, perceptions and meanings of food linked with 
dietary behaviour have been reported in a qualitative study in South Africa (Puoane et 
al., 2006). Accordingly, high educational attainment is linked with improved knowledge 
 19 
University of Ghana http://ugspace.ug.edu.gh
about food and nutrition, which contributes to improved diets and, ultimately, nutritional 
status (FAO, 2012b). Formal education is associated with increased knowledge about the 
benefits of good nutrition and healthy dietary behaviour including the intake of fruits and 
vegetables (Amoateng et al., 2017; Dunneram & Jeewon, 2015; FAO, 2012b; WFP, 
2005).  
 
Farahmand et al. (2012) indicate that women’s and other family members’ food 
preferences were important contributing factors to consumption of poor quality diets 
among women in an Iranian community in a qualitative study on barriers to healthy 
nutrition among women. Additionally, women’s inadequate knowledge or information 
on healthy diets, their inability to distinguish between what is healthy and unhealthy as 
well as their lack of knowledge on appropriate methods of food preparation contribute to 
their poor dietary choices.   
 
In Ghana, Nti et al. (2002) found that even though 83% of pregnant women had some 
knowledge of the nutritive qualities of foods (energy-giving, body-building and 
protective) they consumed, some women were unable to translate this knowledge into 
their dietary habits owing mainly to financial constraints; thus debunking the notion that 
good knowledge of the nutritive qualities of food leads to improved dietary intake.  
Farahmand et al. (2012) confirmed the assessment that a good dietary knowledge does 
not always translate into consuming healthy diets. According to de-Graft Aikins (2014), 
women’s knowledge about food was drawn from a number of sources termed “lifeworld” 
sources - that is, family and friends, educational settings, health professionals, mass 
media and women’s unique pregnancy-related expectations. Although women may not 
have “expert knowledge” of the nutritive qualities of foods, their “social or lay 
 20 
University of Ghana http://ugspace.ug.edu.gh
knowledge” of nutrition was similar to expert knowledge. Expert knowledge implies the 
scientific or medical recommendations of foods grouped into high- or low-nutrient 
densities. Physiological factors such as loss of appetite, nausea and vomiting have been 
found to influence dietary practices among pregnant and lactating women (Aikins, 2014; 
Armar-Klemesu et al., 2014; Koryo-Dabrah et al., 2012; Nti, Larweh, & Gyemfua-
Yeboah, 2002). 
 
2.3.2 Socio-cultural factors  
Socio-cultural factors refer to the way family, community and extra- community systems 
are organized for the management and consumption of food in the community and at the 
household levels. For instance, a study among Iranian women revealed that some social 
and cultural barriers to women consuming nutritious diets include unhealthy behavioural 
modelling in the form of unhealthy eating pattern of the family, peers and community 
that are imitated by other members and inappropriate prioritization of the family budget. 
Here, non-food items may be prioritized above nutritious foods.  
 
Other social barriers identified were time limitations to prepare nutritious diets partly as 
a result of women working outside the home and poor hygiene around food 
environments (Farahmand et al., 2012). Socio-cultural factors also include social norms 
that govern the way resources are controlled at both the community and household levels 
and their impact on behaviour. Socio-cultural issues discussed in this section include 
women’s social status and decision-making power, food taboos and restrictions and 
preferential food allocation system. 
 
 21 
University of Ghana http://ugspace.ug.edu.gh
2.3.2.1 Women’s social status and decision-making power 
In Ghana, women’s participation in labour-intensive farming activities and household 
chores places a lot of demands on their energy. The heavy workload has been identified 
as a major factor for women’s poor diets (Ransom & Elder, 2003; Nti, Inkumsah, & 
Fleischer, 1999). The heavy workload necessitates their leaving home early and, thus, 
compromising their diets and child care practices (Saaka et al. 2017. Women’s multiple 
roles create many physiological problems for them especially general weakness and 
tiredness, body pains; headaches, dizziness, arthritis and malaria (Nti et al., 1999). 
Insufficient energy intake among WRA could be met by a reduction in pregnant 
women’s physical activity (IFAD, 2013). On the contrary, this reduction in women’s 
physical activity is difficult to achieve considering the social norms in low-income 
countries. Most women are unable to meet their daily nutrient requirements and are, thus, 
prone to micro-nutrient mal-nutrition (Martin-Prevel et al., 2017; Underwood & Galal, 
2004). 
 
A positive link has been established between women’s decision-making power on the 
one hand and their dietary diversity and quality on the other hand (Amugsi et al., 2016; 
Levay et al., 2013a; Niloufer, Syed, & Rahat, 2003). It is believed that women’s control 
over food purchases is linked with the quality and quantity of food available to the 
household. Levay et al. (2013) linked maternal mal-nutrition in Bangladesh to many 
factors including the patriarchal gender-order that inhibits mobility and decision-making 
among women including decision-making regarding food consumption. Women’s lack 
of decision-making power has also been reported in Ghana (Mohammed et al., 2013). 
The evidence indicates that in spite of their efforts to achieve food security, most women 
in Ghana face enormous challenges in accessing and controlling productive resources to 
 22 
University of Ghana http://ugspace.ug.edu.gh
carry out their roles. Land rights in the patriarchal societies are held by men and women 
may only be able to access land through their male relatives such as fathers, husbands 
and brothers. In the exceptional situations in which women are self-employed farmers, 
many of them are widows, divorcees or women whose husbands have migrated to seek 
employment elsewhere (Bambangi & Abubakari, 2013; Mohammed et al., 2013).   
 
Women’s participation in decision-making in household food purchasing and other 
aspects of household decision-making (money spending, ability to refuse sex and opinion 
on wife-beating) was found to be significantly associated with higher dietary diversity 
(Amugsi et al., 2016). The assessment further revealed that women from monogamous 
homes had higher dietary diversity scores than those from polygamous homes. The 
authors concluded that improving women’s decision-making autonomy to buy nutritious 
foods could have a positive impact on their dietary quality. 
 
Social status within a household influences dietary behaviour within it. Men in most 
cultures are usually those who attain the highest status in most households (Gittelsohn & 
Vastine, 2003). Leroy et al. (2008) reported that in northern Ghana, extended households 
comprise multiple family units. Each family unit in turn, comprises a male, his wives and 
children and the oldest male assumed the role of head of the household. The head of the 
household has the responsibility of making resources such as land accessible to other 
members of his household in addition to allocating food resources from its stores for 
consumption by the whole household. Leroy et al. further reported that the first wife of 
the families making up a household wields more decision-making power than her co-
wives - thus, giving her more access to better quality food (Leroy, Razak, & Habicht, 
2008). The strategies to improve women’s diets should focus on promoting the decision-
 23 
University of Ghana http://ugspace.ug.edu.gh
making power at the household level. Issues at stake here included improving household 
wealth, promoting female education and encouraging monogamous marriages (Amugsi 
et al., 2016). 
 
2.3.2.2 Food taboos and restrictions 
Two sets of food prohibitions have been described: permanent and temporary or 
transitory food prohibitions (Gittelsohn et al., 2003). Permanent food prohibitions are 
usually associated with cultural or religious groups such as the avoidance of pork by 
Muslims. Adherents of Islam constitute the majority population in Northern Region of 
Ghana (GSS, GHS, & DHS, 2015). The permanent avoidance of selected foods creates a 
group identity and differentiates them from other groups (Senah, 2003). Usually, 
violation of such prohibitions attracts sanctions. Temporary or transitory food 
prohibitions affect individuals or sub-groups of populations such as pregnant women, the 
sick and children or during the post-partum period (Gittelsohn & Vastine, 2003). 
 
Animal-source foods have been found to be the most commonly-prohibited foods 
(Gittelsohn & Vastine, 2003). The existence of food taboos in most cultures, which aim 
at creating a unique identity for people who belong together or are from the same clan, 
has been reported (Senah, 2003). Food taboos pertain to putting restrictions on the 
consumption of certain foods considered to have adverse effects either on the persons 
who consume them or on other members of the group. Food taboos mostly affect 
children, the sick and women of child-bearing age (Gittelsohn & Vastine, 2003). Food 
restrictions during pregnancy and lactation exist in Pakistan and many other cultures. 
Hot and oily foods are among the major foods restricted during pregnancy while cold 
and heavy foods are the main ones restricted during lactation in Pakistan. Restrictions in 
 24 
University of Ghana http://ugspace.ug.edu.gh
other societies include meat and fish in Sudan, buffalo milk in India, fish, curd, grape, 
pineapple, mango, coconut, melon, sugar-cane and long banana in Vietnam (Niloufer et 
al., 2003). 
 
The role of food taboos has further been emphasized in reviews of literature 
(Ogunjuyigbe & Ojofetimi, 2006), facility-based cross-sectional surveys (Niloufer, 
2003) and qualitative studies (Monterrosa, 2017; Arzoaquoi, 2014b; Martínez Pérez & 
Pascual García, 2013). Even though poverty and ignorance are major contributory 
factors to malnutrition, the vital role of food taboos could not be underrated in sub-
Saharan Africa. In Nigeria, pregnant and lactating women are prohibited from 
consuming foods such as cow pea, certain categories of leafy and green vegetables and 
fruits, mutton, bush meat and snail (Ogunjuyigbe & Ojofetimi, 2006). 
 
It is perceived that the consumption of the foods leads to adverse outcomes such as an 
unusually large baby, which affects normal delivery (cowpea and mutton), nausea 
(guava), worms in mother’s stomach (mango), dizziness and slippery/slimy stools (some 
leafy and green vegetables) and animalistic behaviours in children (bush meat). Because 
these foods are nutrient-dense, prohibition affects women who need more energy and 
nutrients to facilitate their physiological functioning (Martínez Pérez & Pascual García, 
2013). 
 
Reported food taboos linked to pregnancy in Ghana include snails in southern Ghana, 
groundnut and meat in Kassena-Nankana District of Upper-East Region and egg in all 
regions of the country (Senah, 2003). The rationale for the taboo against the 
consumption of snail is to prevent the drooling of the child. The egg taboo is to prevent 
 25 
University of Ghana http://ugspace.ug.edu.gh
the baby from becoming a thief. The meat and groundnut taboo is to prevent the birth of 
“spirit children”. Pregnant women are prevented from purchasing vegetables such as 
tomato, pepper, okro and garden-egg from the market in order to protect the unborn child 
from developing skin rashes and physical disability later in life (Senah, 2003). 
 
In Ghana, Arzoaquoi (2014) has reported the presence of food taboos and beliefs during 
pregnancy in Yilo Krobo District of Eastern Region. He reported universal awareness of 
food taboos and beliefs during pregnancy which included snail, rat, “hot foods” and 
animal lungs. Arzoaquoi’s list of prohibited foods was similar to that of Senah (2003). 
The study further indicated that the study population understood food taboos as 
forbidden foods and that the desire for healthy pregnancy outcomes and respect for the 
ancestors and elders of the community were the main reasons for adhering to them. 
 
In spite of the negative influences of food taboos on the dietary practices among lactating 
and pregnant mothers in Ghana, it is usual for lactating mothers to be encouraged to 
consume foods that are perceived to give strength and blood such as palm soup, 
groundnut and light soup, mashed kenkey, beverage and porridge (Armar, 1989). These 
foods are perceived to enhance production of breast-milk. Other practices such as 
massaging the breasts with shea-butter and herbs and the consumption of herbal 
preparations are encouraged for lactating mothers to enhance production of breast-milk. 
Armar’s findings have been corroborated by Haileslassie et al. (2013) in a cross-
sectional study in Ethiopia. 
 
Dietary restrictions contribute to anaemia, malnutrition, low birth-weight and intra-
uterine growth retardation while consequently worsening pregnancy outcomes (Niloufer 
 26 
University of Ghana http://ugspace.ug.edu.gh
et al., 2003). Perez and Garcia (2013) have concluded that food taboos among the Fullas 
and other peoples in Africa contribute to protein-energy malnutrition among children as 
well as pregnant and lactating mothers (Martínez Pérez & Pascual García, 2013). On the 
contrary, the influence of food taboos on women’s dietary practices is considered low 
(Levay et al., 2013a). Levay et al. explained that in Bangladesh, the taboos were not 
generally followed although most women avoided a specific type of fish called “mrigal 
maach” because they believed consuming it caused “mrigy” (epilepsy) in the baby. 
 
2.3.2.3. Preferential food allocation practices 
Preferential food allocation limits the intake of food in a population (Gittelsohn & 
Vastine, 2003). The authors explained preferential food allocation to involve the 
distribution of food at the household level on the basis of how some household members 
were valued above others including favouritism of males over females and adults over 
children. Valuation of some household members above others is linked to their economic 
contributions, social statuses, social relationships and power relations. Gittelsohn and 
Vastine indicated that preferential food allocation has more impact than cultural food 
restrictions or taboos on household food allocation. 
 
Foods that are differentially allocated are luxurious, scarce and expensive such as 
animal-source foods. The natures of these preferential food allocations vary by context: 
while adults are favoured over children in one society, the reverse is the case in another 
society. In some settings, men are favoured over women while in others, women are 
favoured (Gittelsohn et al., 2003). Preferential food allocation has been observed among 
women especially during situations of severe food insecurity in some parts of Ghana. 
Armar-Klemesu et al. (2014) reported that women of reproductive age may restrict food 
 27 
University of Ghana http://ugspace.ug.edu.gh
consumption during periods of food insecurity as a coping strategy to ensure that other 
members of their households, especially children, get enough food to eat. In some cases, 
women serve the other members of their households first and, if there is some food left, 
they could then eat – thus, limiting the quantity and quality of food available for them. 
 
2.3.3 Environmental (physical) factors 
The physical environment includes the climate, soil as well as mineral and water 
resources the society is endowed with to meet human needs for survival including the 
nutritional needs. The physical environment, for instance, influences the kinds of 
agricultural activities and the variety of crops and animals available to the population as 
food (Kuhnlein & Pelto, 2009; Kuhnlein & Receveur, 1996). 
 
The food system and women’s involvement in northern Ghana 
According to FAO (2007), a food system is an interaction between and within the 
physical and human environments that influences activities and outcomes of the food 
chain beginning from food production through processing, preservation, storage, 
preparation, distribution to consumption at the household level (FAO, 2007). The food 
system consists of all processes mankind undertakes to ensure populations are fed 
(Cassidy & Patterson, 2008); thus, food supply, environments and consumers are all part 
of the system (Global Panel on Agriculture and Food System for Nutrition, 2016). 
Interaction with the food system, through the inter-related activities of production to 
consumption of food, enhances a population’s ability to address their nutritional and 
health needs while at the same time promoting the socio-economic and environmental 
well-being of their communities (Kuhnlein & Pelto, 2009). Owing to rapid population 
growth, climate change and urbanization, however, food systems are failing to need the 
 28 
University of Ghana http://ugspace.ug.edu.gh
nutritional needs of population by focusing more on producing large quantities of food 
while compromising on quality. The result is that consumers are unable to make healthy 
dietary choices with optimal nutritional outcomes (Global Panel on Agriculture and Food 
System for Nutrition, 2016). There is, therefore, the need to reverse the current trends in 
food systems to achieve the goal of healthier diets for all.  
 
The food system is one of the components of the ecological framework for 
understanding the social aspects of nutrition (Kuhnlein & Pelto, 2009). The performance 
of the food system at the global, national, regional, local or community levels partly 
determines the food security situation of a population; therefore, to achieve food 
security, there must exist effective food systems in all populations (Tacoli, Burkhari, & 
Fisher, 2013). Kuhnlem and Pelto (2009) posited that even though food is made 
available through the physical environment (food system) and constitutes the main 
source of human nutrition, its accessibility and consumption are influenced by social and 
cultural factors. Understanding how these socio-cultural factors influence nutrition is 
very important when addressing nutritional issues in a community (Delormier et al., 
2009; Lartey, 2008). In their article “Ethnographic aspects of human nutrition”, 
Kuhnlein and Pelto (2009) outlined four activities involved in the food system pathway: 
food production and acquisition; food processing, preservation, storage and preparation; 
distribution; and, consumption and biological utilization. 
 
Food production: The joint-activity of food production and acquisition involves the 
adaptive mechanisms for the growing of crops and rearing of animals and fish to sustain 
and promote human growth and health. Five strategies have been described as 
mechanisms man has devised in acquiring his food over the years. These strategies are 
 29 
University of Ghana http://ugspace.ug.edu.gh
hunter-gathering, pastoralist, horticulturalist, intensive agriculture and industrial 
agriculture food systems. Even though some of these strategies are fading out and giving 
way to mechanized ways of acquiring food, some rural societies may still exhibit 
characteristics of some of them. It is important to note that these strategies of food 
production influence people’s social organizations and culture. 
 
In Ghana, the traditional labour-intensive way of farming is the norm in rural 
communities (Darfour & Rosentrater, 2016). Most rural households produce foods from 
their farms and store them to last throughout the year, supplementing them with foods 
purchased from the market (Armar-Klemesu et al., 2014). In urban areas, however, 
acquisition of foods is done mostly through buying them from the market (Armar-
Klemesu, 2000). The factors that influence the purchasing decisions of women in Ghana 
include availability of money, what food to eat, number of people in the family, cost of 
food items and time available to prepare the food.  
 
In northern Ghana, small-scale farming (less than 5 acres of land), involving 62% of 
farming households is the main livelihood activity of the people (WFP, 2012). Women, 
who constitute the majority of small-scale farmers, are mostly disadvantaged when it 
comes to access to productive resources such as land, inputs, irrigation facilities, credit 
facilities and the services of extension officers (Agarwal, 2014; WFP, 2012). Accounting 
for this limited access to productive resources by women is the patriarchal nature of the 
society, which perpetuates the control of land in the hands of men (Bambangi & 
Abubakari, 2013; Doss & Morris, 2000). Ministry of Food and Agriculture (MOFA) 
reported that, even though the adoption of improved or innovative ways of production is 
limited in the region owing to limited household resources to afford such services, there 
 30 
University of Ghana http://ugspace.ug.edu.gh
is lack of sensitivity to women’s needs when it comes to the design of agricultural 
innovations and extension services (MOFA, 2010). 
 
Food production heavily depends on the climate. Foods produced including staple crops 
such as cassava, cowpea, maize, rice and yam have very low annual yields and low 
economic returns to farmers (MOFA, 2010). In Ghana, it is generally difficult to 
distinguish between men’s and women’s crops. Both men and women cultivate whatever 
crops they are capable of producing but there are a few crops whose cultivation is 
dominated by either men or women, hence, there exist some gender-based patterns of 
cropping (Doss & Morris, 2001). While some crops have more women involved in 
cultivating them, others are more likely to be grown by men than by women.  
 
Doss (2001) indicates that the crops in Ghana that may be considered men’s crops 
include rice, sorghum, tobacco and coffee. Doss is of the opinion that even though both 
men and women in Ghana are involved in producing for both home consumption and 
sale, men are more involved in cash crop production while women dominate in 
producing for home consumption. Even though agriculture is the main livelihood source 
in northern Ghana, the low level of productivity makes it impossible for the zone to be 
self-sufficient in terms of food production.  
 
Worsening the food insecurity situation in northern Ghana is the limited opportunities for 
off-farm income-generating activities especially in the rural areas, making it difficult for 
the capacity of households to be able to offset the shortfalls in their own farm outputs. 
Livelihood diversification comes with crop and animal production, trading in food crops, 
 31 
University of Ghana http://ugspace.ug.edu.gh
some dry season farming (for communities with rivers near-by) and remittances from 
migrant household members (Aasoglenang & Bonye, 2013). 
 
Food processing, preservation, storage and preparation: The development of manual 
techniques of processing and preservation of food such as drying, salting, fermentation 
and milling is dominated by women in northern Ghana. Some of these processing 
techniques improve their nutritive qualities (Sanjukta & Rai, 2016). Use of plant ash and 
fermentation to dry and process maize results in improved nutritional quality of maize. In 
addition, some food-processing techniques bring about new flavours and textures. In 
Ghana, a number of food items are locally processed before they are prepared into edible 
states. Maize and cassava are typical examples of locally-processed food items. Drying 
of fish and meat is also common. Commercial processing and canning of palm-fruit, 
cassava, plantain and cocoyam for fufu are more recent examples of food processing in 
Ghana. Mechanized techniques such as canning help to extend the shelf-lives of foods. 
 
Preparation of food involves the processing of food into an edible state ready for 
consumption at the household level. This activity is dominated by women and influenced 
by factors such as the type of meal and when it is eaten, the household members’ 
preferences, money available, family members’ physiological and health needs, size of 
the family and prizes of foods. The other factors enumerated by Armar-Klemesu (2000) 
included where to acquire the foods and the equipment used in cooking. The most 
commonly-used equipment for preparing and cooking most Ghanaian foods consists of 
firewood and charcoal in rural communities in contrast with gas and electric cookers 
together with charcoal in urban communities.  
 32 
University of Ghana http://ugspace.ug.edu.gh
Food distribution: Food is distributed through outlets such as farm gates, markets, 
supermarkets, other retail shops, schools and homes. Both men and women are involved 
in food distribution. The outlets provide consumers with a wide variety of food items. 
The effectiveness of these outlets of distribution, however, depends on the transport 
system and proximity of food to consumers (Darfour & Rosentrater, 2016).  In places 
where the transport system is efficient and food outlets are near to consumers, they are 
more likely to have easy access to food. The cultural aspect of food distribution involves 
the kinds of food that are believed to be appropriate for different categories of members 
of a household. This manner of food distribution in the household is affected by the 
cultural belief system in the community.  
 
The practice of giving preferential treatment to men when sharing meat at the household 
level and the buffering of young children in times of food shortages at the expense of 
adults are examples of the effects of culture on the food distribution practices of 
populations at the household level (Kuhnlein & Pelto, 2009). Ceremonial and individual 
life-cycle events, which are commemorated with feasting, offer opportunity for foods 
rich in some nutrients to be distributed within the community. Examples of such feasts 
are marriage ceremonies and funerals. The other forms of food distribution include 
charity from the rich to the poor (Hossain, 2012) and food distribution through 
governmental and non-governmental organizations to people in need. 
 
Food consumption: Food consumption at the household level or within families and 
communities is affected by cultural rules (Douglas, 2014; Whitehead, 1984). Cultural 
beliefs determine, for example, what food is either edible or not edible, what constitutes 
a meal or which foods are restricted for spiritual or ceremonial purposes. The household 
 33 
University of Ghana http://ugspace.ug.edu.gh
food consumption pattern in Ghana indicates a trend towards three meals a day for most 
households (FAO, 2009). These foods are either home prepared or purchased ready-to-
eat (RTE) within the community. Ready-to-eat purchased foods have become very 
common in most parts of the country. Their consumption is, however, influenced by the 
availability of money, the need to vary foods consumed on a daily basis, the absence of 
family members from home (either working or in school), increasing numbers of women 
working away from home for long hours and absence of food items at home (Underwood 
& Galal, 2004; Armar-Klemesu, 2000).  
 
Food security issues 
Food insecurity has been identified as an important cause of women’s poor dietary status 
in Bangladesh (Levay et al., 2013). Insufficient access to food directly reduces the 
quantity and quality of food available and accessible to women and other members of 
their households. Food security is the ability of individuals or household to have access 
to adequate quantities and qualities of food needed for an active and healthy life at all 
times (WFP, 2012; Quaye, 2008). It has also been reported that the non-availability of a 
variety of nutritious foods and high costs of healthy foods constitute major barriers to 
women accessing optimal diets (Farahmand et al., 2012).  
 
In Ghana, women contribute to all aspects of food security: food availability, 
accessibility and utilization. Women constitute 47% of the agricultural labour force and 
account for about 70% of the total crop production in the country (Mohammed, Solikua, 
& Mansega, 2013 ). Women are involved in the various aspects of the food system 
including the production and acquisition of food as well as post- harvest activities such 
as processing, preservation, distribution and consumption (Braimah et al., 2014; 
 34 
University of Ghana http://ugspace.ug.edu.gh
Mohammed et al., 2013 ; Boakye-Achampong, Mensah, Aidoo, & Osei-Agyemang, 
2012; Kuhnlein & Pelto, 2009). It is usually women who grow the majority of family 
foods including fruits and vegetables and take the decisions on which foods to buy for 
feeding the family when their farm produce gets used up (United Nations, 1990). In spite 
of their immense contributions to all aspects of food security, women tend to be more 
under-nourished than their husbands and other male relatives (Zereyesus et al., 2014; 
Boakye-Achampong et al., 2012). 
 
On the average, while 5% of the population of Ghana is estimated to be food insecure, 
the prevalence rate in the three northern regions is estimated to be 35%. (Zereyesus et al., 
2014; Boakye-Achampong et al., 2012). Saaka investigated the relationship between 
food insecurity and pregnant women’s nutritional status in northern Ghana and reported 
that as many as 60% of households were worried about not having enough food to eat 
and a similar proportion reported having limited varieties of food owing to lack of 
money to buy food. The study also found 28% of WRA to be under-weight - a 
percentage far higher than the 12% in the 2014 GDHS Report (Saaka et al., 2017). 
 
2.4 Interventions addressing challenges to women’s dietary practices 
In order to achieve positive dietary outcomes, interventions need to consider cultural and 
contextual factors affecting dietary intake. (Kong, Tussing-Humphreys, Odoms-Young, 
Stolly, & Fitzgibbon, 2014; Sliwa et al., 2011). Nutrition education programmes are 
designed to facilitate voluntary adoption of food choices and other food and nutrition-
related behaviour conducive to health and well-being (Dunneram & Jeewon, 2015). 
Nutrition education has been considered a priority and an important pathway for 
improving dietary behaviour of populations through behavioural change communication 
 35 
University of Ghana http://ugspace.ug.edu.gh
strategies. Nutrition education has a strong influence on people’s knowledge, attitudes 
and perceptions (Contenta, 2008). 
 
Effective nutrition education should be based on appropriate theory and evidence 
(Contenta, 2008). Most theories that are used in nutrition education are drawn from the 
field of social psychology and health behaviour models, which concentrate on the 
importance of people’s cognition, motivations, values and perceptions about the world as 
well as take into consideration people’s cultural and physical environments, both of 
which influence their behaviours and practices. Nutrition education can take place at 
varied settings including the community and health facility levels. 
 
The effects of health and nutrition education on women’s post-partum beliefs and 
practices were investigated in China by using a randomized controlled trial (Liu et al., 
2009). The strategies for the intervention included face-to-face contacts between the 
health workers and post-partum women and a health education guidebook. The outcome 
of the intervention showed that participants in the intervention groups showed 
significantly higher levels of improvements in overall dietary behaviour than participants 
in the control groups. The outcomes that were observed in the study included the 
consumption of fruits and vegetables, soya bean and soya products. Significantly more 
women in the intervention groups stopped adhering to traditional food taboos. Other 
effects of the intervention included lower scopes of constipation, fewer leg-cramps or 
joint pains and prolonged lochia rubra among participants in the intervention against the 
control groups. They concluded that, to have a greater impact on the target audience, 
intervention strategies must include television programmes, magazines and the internet.  
 
 36 
University of Ghana http://ugspace.ug.edu.gh
Community-based participatory dietary interventions are among the means by which 
women have been reached with dietary interventions. Community-based participatory 
research promotes the participation of members of communities in controlling and 
influencing decisions and actions that affect their communities (Savin-Baden & Major, 
2013b). In this instance, the research partners respond to community needs and the 
environment including culture rather than impose predetermined ideas of what will work 
best for the researched community (Sloane et al., 2003).  
 
In the past, the use of community participation in implementing health interventions has 
been minimal (Moore et al., 2015). As a result, outcomes of interventions were not very 
encouraging (Kpormegbe and Ahorlu, 2014). Recent recognition of the effectiveness of 
community-based participation in facilitating positive outcomes in interventions led to 
the use of the approach in implementing health-related interventions (Kpormegbe & 
Ahorlu, 2014b). 
 
In a systematic review of literature on healthy diets and nutrition education among 
WRA, Dunneram and Jeewon (2015) emphasized the importance of dietary interventions 
at community levels in encouraging the adoption of healthy dietary practices and 
improving WRA’s health at the family and community levels. The authors reviewed a 
cross section of nutrition and dietary interventions among WRA and came to the 
conclusion that community-based interventions using multiple strategies had proven to 
be useful for improved behaviour modifications among women. Additionally, gathering 
of women to share nutrition information in community-based intervention had proven to 
be effective in improving women’s dietary knowledge and practices. 
  
 37 
University of Ghana http://ugspace.ug.edu.gh
In a similar community-based intervention in Mali, Bonde (2016) introduced an 
intervention consisting of giving livestock to vulnerable households with children under 
two years of age and training recipients on the care of the animals as well as the 
preservation, drying and storage of crops. Education on nutrition and hygiene as well as 
detection of child mal-nourishment was provided. The channels for communication 
included radio broadcasts and food demonstration sessions using community health 
agents. Results of the intervention showed that there was general improvement in the 
scopes of access to food in the intervention community. On children’s dietary diversity, 
the assessment recorded an increase from 3.9 food groups to 4.4 (from 9 groups) at the 
post-intervention stage. Similarly, the proportion of women with low dietary diversity 
scores decreased from 46% at baseline to 26% at post-intervention (Bonde, 2016). 
 
Recourse to sustained education with the view of improving the dietary status of 
pregnant women and lactating mothers has been suggested (Oluwafolahan, Catherine, & 
Olubukunola, 2014; Haileslassie et al., 2013). Haileselassie et al. (2013) went on to 
specify increased intake of food, proper dietary practices and dietary diversification 
during lactation as the main topics for nutrition interventions. Ersino et al. (2017) 
supported the need for prolonged community-based intervention activities involving 
monthly interactive community meetings and home-visits. Ersino et al. opined that such 
interventions could enhance the dietary diversity scores (DDS) in both mothers and their 
children and also reduce the scopes of mal-nutrition among children. The opinions of 
Ersino et al. derived from the outcomes of their six-month randomized controlled 
nutrition education intervention on the intake of pulse and other foods among 200 
mother-child pairs (children less than five years old) in two purposively-selected 
communities in the pulse-growing region of Halaba in south Ethiopia. The channels for 
 38 
University of Ghana http://ugspace.ug.edu.gh
delivering the intervention consisted of interactive monthly community meetings and 
home-visits.  
 
Data was collected on variables such as socio-demographic characteristics, knowledge, 
attitudes and practices related to food, dietary information, DDS and anthropometric 
information at both baseline and endline. Additionally, focus group discussions with 
local farmers were used. Outcomes among participants in only the intervention 
community were more favourable and included improved knowledge, attitudes, practices 
and perceptions of the benefits of pulse nutrition among mothers. Other benefits were 
DDS and consumption of pulse and animal-source foods by both mothers and children 
and mean body-mass index- for-age (BMI for-age). Furthermore, farmers in the FGDs 
were said to have indicated their willingness to increase their production of pulse as well 
as allocate more quantities for home consumption (Ersino, Henry, & Zello, 2017). 
 
2.4.1 Interventions addressing challenges to women’s dietary practices in Ghana 
Interventions addressing challenges to women’s nutrition including women’s dietary 
needs in Ghana include micronutrient supplementation, behaviour change 
communication (BCC), food supplementation with nutrition education in the three 
northern regions, food-based strategies and improving healthcare services for women 
(GOG, 2013; FAO, 2009; Lartey, 2008). Furthermore, the government of Ghana in 
collaboration with the USAID has been implementing Ghana’s Feed the Future Initiative 
with focus on the three northern regions owing to their higher scopes of food insecurity 
and levels of poverty (Zereyesus et al., 2014).  
 
 39 
University of Ghana http://ugspace.ug.edu.gh
The effectiveness of community-based interventions in improving women’s dietary and 
nutritional status has been emphasized in other studies (Otoo and Adam 2016; Tchum et 
al. 2009; Colecraft et al. 2012). In Gomoa-East District of Central Region of Ghana, a 
ten-week randomized controlled trial was conducted to assess the impact of nutrition 
education on HB levels, nutrition knowledge and dietary intake of pregnant women 
(Otoo & Adam, 2016). The content of the intervention included the consumption of iron-
rich foods, vitamin C-rich foods, grains, root and tubers, legumes and red meat by 
women. 
 
The education sessions were carried out on individual basis through bi-weekly home-
visits and bi-weekly phone calls to the participants who were recruited from four ante-
natal clinics.  Participants in the intervention group achieved significant increases in both 
mean knowledge on anaemia and iron-rich foods and HB concentrations over those in 
the control groups. Intakes of red meat, legumes, grains, roots and tubers and vitamin C 
–rich foods were also significantly higher for participants in the intervention group than 
those in the control group. There was, however, no significant difference in the change in 
weight between participants from the two groups.  
  
Tchum et al (2009) conducted a three-month community-based randomized controlled 
trial involving post-partum mothers in Ejisu-Juaben District of Ashanti Region to assess 
the effects of the consumption of African egg-plant leaves on post-partum Ghanaian 
mothers’ vitamin A status. The intervention included daily supplies of 200grams of 
African egg-plant leaves for only participants in the intervention groups for 3 months. 
Results of the intervention indicated a significant improvement in the vitamin A status of 
participants in the intervention group. Tchum et al. suggested that dietary modification 
 40 
University of Ghana http://ugspace.ug.edu.gh
and nutrition education to WRA should include natural and locally-available food 
sources rich in pro-vitamin A. They justified their suggestion with the claim that 
indigenous leafy vegetables could be easily cultivated locally thereby enhancing their 
sustainability and affordability in comparison with periodic oral dosing with vitamin A 
(Tchum et al., 2009).  
 
Even though both interventions focused on improving WRA’s dietary quality, 
community-based randomised control trials were employed in both studies without 
active participation of community members in the design and implementation processes. 
In addition, quantitative methods were used in the studies without recourse to qualitative 
assessment of the process to address implementation bottlenecks and to understand 
participants’ views of the interventions. These limitations notwithstanding, the outcomes 
of both studies were positive: improvements in the quality of WRA’s diets. 
 
2.5 Qualitative Research 
Different researchers hold various views about the nature of reality (ontology) and the 
nature of knowledge (epistemology) but for the qualitative researcher, research is 
inherently subjective (Savin-Baden & Major, 2013a; Creswell, 2009). The subjective 
view about research stems from the ways qualitative research is carried out. First, 
qualitative research occurs in the participants’ natural setting and aims at getting 
researchers to understand social phenomena from the perspective of the researched - that 
is, the “emic” perspective. In this case, the researcher does not aim at analyzing 
relationships between variables in order to determine cause and effect but rather to 
understand the phenomena from the various accounts given by the participants in the 
study. Related to this characteristic is the belief that people may have different 
 41 
University of Ghana http://ugspace.ug.edu.gh
perceptions of reality - thus, attributing meanings and concepts to such perspectives, 
which is what the researcher tries to understand. Second, owing to the emic perspective 
of qualitative research, the study participants become an integral part of the study. Some 
researchers may even engage the study participants as co-researchers to some degree in 
the research process especially in the case of most participatory action researches (Savin-
Baden & Major, 2013a). 
 
The issue of subjectivity is related to the position of the researcher in shaping the nature 
of the research process and vice-versa. In this regard, the researcher decides on what 
should be the inclusion and exclusion criteria in the research process, what kinds of 
questions should be asked and how the results are to be interpreted. The qualitative 
researcher, therefore, believes that research cannot be value-free so there is the need for 
the researchers to state their positions and biases and how these influenced the research 
process (Tong, Sainbury, & Craig, 2007). 
 
Qualitative research assumes the existence of a world that is constantly changing as a 
result of changes in the social, political, economic and cultural influences (Loewenson, 
Laurell, Hogstedt, D'Ambruoso, & Shroff, 2014; Savin-Baden & Major, 2013a). People 
may, hence, have different views of reality at different points in time. Researchers using 
this approach have the option to choose from a range of strategies such as ethnography, 
grounded theory, case studies and phenomenology (Creswell, 2007). A review of these 
methods indicates that even though the different methods have specific foci based on 
their research questions, they all have one common feature – namely, understanding the 
phenomenon under study from the participants’ perspective. Within any of these 
methods, the qualitative researcher could employ naturalistic observations, semi-
 42 
University of Ghana http://ugspace.ug.edu.gh
structured or unstructured interviews, case studies and focus group discussions in 
collecting and analysing data (Loewenson et al., 2014). Considering the foci of the 
current study, which were the cultural and social issues pertaining to women’s dietary 
practices; ethnography was deemed the most suitable method. 
 
2.5.1 Ethnography 
Ethnography is an aspect of anthropology which seeks to understand other cultures 
(Savin-Baden & Major, 2013a). Early anthropologists used ethnography to explore, 
understand and tell indigenous people’s stories. Even though ethnography has gone 
through several developmental phases, it has come to be understood as the study of 
people, their cultures and values and it also attempts to create an understanding of the 
people being studied. What counts as an ethnographic study is the intensive fieldwork in 
order to gain detailed accounts and insight of the group being studied.  
 
Specifically, ethnographic researchers focus on the everyday life of the people being 
studied, spend time with the group being studied, use participant observation as a 
primary tool for gathering data together with in-depth unstructured tool of data collection 
and, finally, present findings from the participants’ points of view. Ethnographic 
research could span months and years in a cultural setting trying to understand the 
people’s ways of life involving an intensive and detailed process of data collection 
(Savin-Baden & Major, 2013b). 
 
Ethnography was an appropriate choice for this study because it sought to understand 
cultural norms, knowledge and perceptions of the study communities and their influences 
on women’s dietary practices by using different tools of data collection. The tools 
 43 
University of Ghana http://ugspace.ug.edu.gh
included community and household observations, key informant interviews, in-depth 
interviews, focus group discussions and qualitative 24-hour recall. 
 
2.5.2 Participatory Action Research (PAR) 
Savin-Baden and Major (2013) defined PAR as a method of intervention, development 
and change that is carried out in communities or among groups in which issues affecting 
them are systematically examined from their point of view. According to Savin-Baden 
and Major, PAR is both a research design and a philosophical world-view, which seeks 
to empower communities to become more aware of their constraints by getting 
researchers to work with the people as opposed to imposing ideas on them. Using PAR 
enables researchers to generate information or knowledge to inform an action 
(intervention). Loewenson et al. (2014) supported the view by Savin-Baden and Major 
(2013) about PAR by defining it as a research design that seeks to understand and 
improve communities by involving the collective action of those affected by the research 
problem to produce desired change as a means of generating new knowledge. In their 
view, researchers interpret participatory action research differently and apply it to 
different research approaches. 
 
2.5.3 Qualitative evaluation of interventions 
While most randomized control trials may not provide information on how an 
intervention may be replicated in a similar context to produce similar results 
(Kpormegbe & Ahorlu, 2014a), qualitative evaluations of interventions are able to fill 
this gap by providing useful information regarding the intervention process, participants’ 
experiences and reactions to the intervention as well as components of the intervention 
 44 
University of Ghana http://ugspace.ug.edu.gh
that either worked or did not work (Bell & Aggleton, 2016; Fonteyna & Bauer-Wub, 
2005; Patton, 2002).  
 
A process evaluation is a form of qualitative assessment of a programme or an 
intervention, which involves reporting on how the programme was implemented either in 
place of or in addition to assessing its outputs and outcomes (Patton, 2002). Process 
evaluation is recommended as a tool for assessing the following three issues pertaining 
to an intervention: first, fidelity and quality of implementation (whether the intervention 
was delivered as intended); second, clarification of causal mechanisms and, third, 
identification of contextual factors that may account for variations in the outcomes 
(Moore et al., 2015). 
 
Patton (2002) argued that some styles or approaches to community development operate 
on the premise that “what we do is as important as how we do it”- that is, actively 
involving people in the development process is an essential outcome in itself and not 
merely a means to achieving some other outcomes. In other words, the process or 
journey rather than just the destination is what matters in undertaking a process 
evaluation of any intervention. In Patton’s view, the level of community participation 
and involvement in carrying out an intervention, for instance, is critical for building a 
better relationship and mutual understanding along the intervention process. “The 
process, in this case, then becomes the outcome since the production of a plan or actual 
intended outcome becomes the means to building the relationship with the community” 
(Patton, 2002; pp. 159). Unfortunately, most interventions emphasize intervention 
outcomes at the expense of the process because they claim that the outcomes are more 
important (Moore et al. 2015). Patton, however, reiterated that the process-outcome 
 45 
University of Ghana http://ugspace.ug.edu.gh
relationship must be the driving force in all evaluations of interventions. Process 
evaluation is, therefore, an essential aspect of designing and evaluating interventions to 
assess their impacts or expected outcomes. 
 
Moore et al. (2015) claimed that there were two ways by which the outcome of an 
intervention could be affected: first, the design of the intervention and, second, how the 
intervention was implemented. The authors added that in some instances, however, an 
intervention could still achieve high impact even when it was not fully delivered. To 
fully understand what actually worked in an intervention, therefore, required a full 
description of the processes of implementing it as against looking at its expected 
outcomes. The authors argued that in process evaluation, a clear description of the 
intervention is key to understanding the outcomes of implementation. They, therefore, 
proposed a framework with three main components for conducting and reporting process 
evaluation studies: the implementation process, mechanism of impact and effect of 
context on outcome. The framework is in line with the Medical Research Council (MRC) 
guidelines for conducting process evaluations (Moore et al., 2015). 
 
Process evaluation describes into detail the implementation of an intervention using 
qualitative inquiry (Patton, 2002). Patton contended that depicting the intervention 
process demands detailed descriptions of the following issues: first, how people engaged 
with one another during the implementation; second, verbatim narratives of different 
participants’ experiences encountered and, third, participants’ perceptions of the 
intervention. Furthermore, process evaluation does not involve assessing only formal or 
planned activities and anticipated outcomes but also informal patterns and unanticipated 
outcomes and how these were mitigated to facilitate change. Reporting these occurrences 
 46 
University of Ghana http://ugspace.ug.edu.gh
permits judgments to be made regarding the extent to which the intervention was 
implemented, the way it was initially planned to operate, which components of the 
intervention worked best and which needed improvements.  
 
Formative evaluation, which aims at improving components of interventions as they are 
being rolled out to improve subsequent components, is encouraged rather than 
summative evaluation in some evaluation researches. Qualitative or process evaluations 
have been approached differently depending on the nature of the intervention (Allison et 
al., 2017; Eaves et al., 2017; Russ et al., 2014; Fonteyna & Bauer-Wub, 2005; Ducharma 
& Trudeau, 2002). Even though these qualitative evaluations were done on the basis of 
fidelity of intervention, participants’ general impressions, benefits and suggestions for 
improvements in future interventions, they did not incorporate comparison groups in 
their studies. Additionally, all except one study did not collect pre-intervention data. 
 
2.5.4 Analysis of qualitative data  
A number of data analysis methods exist for qualitative data inquiry. They include 
grounded theory, discourse analysis, interaction analysis, conversational analysis, 
content analysis, framework analysis and thematic analysis. Six steps have been 
proposed for analyzing qualitative data thematically (Braun & Clarke, 2006). These steps 
are familiarization with the data, generating initial codes, searching for themes, 
reviewing themes, defining and naming themes and, finally, producing the report. 
 
The first step, familiarization with the data, involves exploring the data extensively by 
reading and rereading thoroughly through the transcripts and listening to the recorded 
interviews to reconcile them with the transcripts, thereby ensuring validity of the latter. 
 47 
University of Ghana http://ugspace.ug.edu.gh
Rereading the transcripts, studying field notes and listening to the tapes further enable 
the researcher to take note of words, phrases and expressions frequently emerging from 
the participants’ responses. After being familiar with the stories told by the various 
transcripts, all noted expressions, phrases and words identified in them which appeared 
interesting and meaningful are coded systematically across the entire data set. Verbatim 
responses (or cases) are then collated and assigned to each code. These codes, even 
though numerous, provide some indications of the contexts of the stories told in the 
various transcripts and field notes. 
 
The third step involves searching for themes. Themes are concepts including ideas, 
issues and behaviours emerging from the transcripts. Braun and Clarke (2006) described 
a theme as a subject that “captures something important about the data in relation to the 
research questions and represents some level of patterned responses or meanings within 
the data set” All the coded responses are collated into potential themes and all relevant 
data pertaining to each potential theme are brought together as main themes and sub-
themes. The fourth step comprises a deeper review of the initial themes and sub-themes. 
In this regard, identified themes that appear similar are combined while others are 
considered sub-themes. In the fifth step, each of the identified major themes is given a 
concise and contextual working definition. This exercise enables the story from the data 
to be told by the emerging themes (Braun and Clarke, 2006). The final step involves 
producing the report on the basis of the themes and sub-themes identified in the fifth 
step. 
 
 48 
University of Ghana http://ugspace.ug.edu.gh
Analysis of dietary quality 
In low-income countries, nutrient adequacy is measured using dietary diversity as a 
proxy. This approach is advantageous because it is less costly and less cumbersome 
compared to others methods such as proportionality (more of some food groups and less 
of others) and moderation in consuming certain foods perceived to be risk factors for 
diseases (Workicho et al., 2016). A higher dietary diversity score is associated with 
increased nutrient intake and improved nutritional status (Workicho et al. 2016). Both 
qualitative and quantitative methods exist for analyzing and interpreting dietary data. 
The qualitative methods include interpreting dietary data based on the number of food 
servings compared to requirements by using food pyramids and the use of dietary 
diversity scores. In the FAO (2011) guidelines for measuring household and individual 
dietary diversity, Women’s Dietary Diversity Score (WDDS) has been designed 
specially to assess the nutritive quality and micronutrient adequacy of women’s diets.  
 
Dietary diversity has been operationalized in several ways such as the total of food 
groups consumed across a specific period of time. For many years, a number of 
organizations, including the FAO, have been using results based on a 9-point food-group 
system as indicators for the quality of women’s diets. WDD score is calculated by using 
a count of nine food groups consumed over a 24-hour period. The WDDS has been 
categorized into three. The first category, low-dietary diversity, indicates that a woman 
consumes foods from three or less out of the nine groups. Middle-dietary diversity shows 
that a woman consumes foods from four to five different groups. Last, high-dietary 
diversity means foods from six groups upwards are consumed (Zereyesus et al., 2014; 
FAO, 2011). 
 
 49 
University of Ghana http://ugspace.ug.edu.gh
More recently, however, the need for a simple dietary indicator that can be expressed in 
terms of either meeting or not meeting a minimal diversity score has led to the 
development of a Minimum Dietary Diversity – Women (MDD-W) (Martin-Prevel et al., 
2017). The MDD-W is a dichotomous indicator used to state whether or not a woman 
consumed foods from at least five out of ten groups the previous day. The proportion of 
women achieving this minimum score is used as a proxy indicator for high micronutrient 
adequacy (an important dimension of dietary quality) (Martin-Prevel et al., 2017; FAO, 
2011).  
 
Using ten food groups as against nine used by the FAO (2011), the MDD-W proposes a 
cut-off point of 5 food groups as the minimum number of food groups required to meet 
one’s daily requirement of nutrients so that the consumption of foods belonging to less 
than 5 groups suggests inadequacy of nutrient intake. The MDD-W was more 
appropriate for this study because it is simpler (easier to understand and construct). It 
also gives a quicker indication of the proportion of women meeting their requirements of 
nutrients in a population. A comparative break-down of foods belonging to the nine and 
ten food groups proposed by the FAO (2011) and FAO and FHI 360 (2016) respectively, 
for assessing women’s dietary diversity scores is shown in Table 2.1. 
 50 
University of Ghana http://ugspace.ug.edu.gh
Table 2.1: Comparative schedule of the Nine and Ten food group classifications 
FA0 (2011) FAO and FHI 360 (2016) 
Food Group Individual foods Food groups Individual foods 
1.Starchy Cereals such as corn, maize, 1. Starchy Cereals such as corn, maize and 
staples rice, sorghum and millet and staples rice, sorghum and millet and 
foods made from any of these foods made from any of these 
White roots and tubers such White roots and tubers such as 
as white potato, white yam white potato, white yam and 
and cassava and foods made cassava and foods made from 
from any of these any of these 
2. Dark Dark green leafy vegetables 2. Dark green Dark green leafy vegetables 
green leafy including wild fruits, vitamin leafy including wild fruits, vitamin A- 
vegetables A- rich leaves such as cassava vegetables rich leaves such as cassava leaf, 
leaf, kontomire and amaranth kontomire and amaranth,  
3. Other Vitamin A-rich fruits such as 3. Other Vitamin A-rich fruits such as 
vitamin A- ripe mango, apricot, pawpaw vitamin A- ripe mango, apricot, pawpaw 
rich fruits and dried peach and 100% rich fruits and and dried peach and 100% fruit 
and fruit juices made from these vegetables juices made from these fruits 
vegetables fruits 
4. Other Tomato, onion, egg-plant and 4. Other fruits Wild fruits and 100% fruit juices 
fruits and other locally-available 
vegetables vegetables, wild fruits and 
100% fruit juices 
  5. Other Tomato, onion, egg-plant and 
vegetables other locally-available 
vegetables 
5. Organ Liver, kidney, heart, any 6. Meat, Liver, kidney, heart, any organ 
meat organ meat, blood-based poultry and meat, blood-based foods, 
foods fish chicken, beef, mutton, lamb, 
fish, anchovies 
6. Meat and Beef, pork, lamb, goat, rabbit, 7. Pulses Dried beans, dried pea, lentil, 
fish game, insects or any other (beans, peas 
bird and lentils) 
7. Eggs Eggs from chicken, duck, 8. Egg Eggs from chicken, duck, 
guinea-fowl or any other egg guinea-fowl or any other egg 
8. Legumes Dried beans, dried pea, lentil 9. Nuts and Nut, seeds or foods made from 
and nuts nuts, seeds or foods made seeds these -e.g. peanut butter, 
from these- e.g. peanut butter hummus 
and hummus 
9. Milk and Milk, cheese, yogurt or other 10. Dairy  Milk, cheese, yogurt or other 
milk milk products milk products 
products 
Source: FA0, (2010), Zereyesus et al. (2014) and FAO and FHI 360 (2016) 
 
 
 51 
University of Ghana http://ugspace.ug.edu.gh
The content of the two approved categorizations of food groups are similar; however, 
while the 9-food group approach combined a number of foods in one group (in the case 
of legumes, seeds and nuts and other vegetables and fruits), the ten-food group approach 
separated them into nuts/seeds; bean/pea/lentil; other vegetables and other fruits). In the 
case of animal-source foods, the 9-food groups approach separated organ meat from 
meat and fish, while the ten-food groups approach combined all three animal-source 
proteins in one group. 
 
2.6 Behaviour change communication (BCC) interventions  
One of the widely used promotional strategies in health- and nutrition-related 
interventions is BCC (Tull, 2017). BCC is a process of working with individuals and 
communalities or groups to develop communication activities to enhance the promotion 
of positive or beneficial health-related behaviour patterns and to create a supportive 
environment for sustaining the new behaviour (Tull, 2017). 
 
Human behaviours, including sub-optimal dietary behaviour or practices, are complex 
(USAID, 2017) but play an important role in many of the leading causes of morbidity 
and mortality such as malnutrition (both under-and over-nutrition) and its implications 
for cardiovascular diseases globally (Davis, Cambell, Hildon, Hobbs, & Susan, 2014). 
The complexity of human behaviour is partly the result of social norms guiding human 
actions or inactions, access to resources, self-efficacy, habit and structural weaknesses 
and opportunities (USAID, 2017). Altering such behaviour, no matter how little, could 
have some appreciable impact on the health outcomes of a population (Davis et al., 
2014). Investments in undertaking behaviour change interventions to address some of 
these behavIoural lapses has, unfortunately, been relatively small even though the 
 52 
University of Ghana http://ugspace.ug.edu.gh
evidence shows their in improving the health outcomes of individuals, families and 
communities at large (Davis et al., 2014).  
  
Recent discourse in health- and nutrition-related interventions using behaviour change 
techniques has shown a shift from the use of the concept “behaviour change 
communication” (BCC) to “social behavior change communication” (SBCC) (Tull, 
2017). The evidence suggests that the shift in terminology from BCC to SBCC is 
necessary to reflect current emphasis place on improving health outcomes through 
individual and community behaviour where the social context of behaviour as well as the 
systems and processes influencing behaviours are identified and strengthened. Social-
behaviour change communication (SBCC) intervention is, therefore, understand as a set 
of interventions or strategies made up of elements of inter-personal communication, 
social change and community mobilization activities, mass media and advocacy to 
support individuals, families and communities in adopting and maintaining behaviour 
that has positive health outcomes. In some cases, the concepts of BCC and SBCC are 
used interchangeably depending on the focus of the intervention (USAID, 2017)  
 
The main principles underlying SBCC are that individuals do not act in isolation but in 
the context of families and communities. Peoples’ behaviours are guided by the social 
and physical environments in which they find themselves; therefore, designing 
interventions that address these external factors stand a better chance of making more 
impact than focusing only on the individual (USAID, 2017). A second consideration in 
SBCC is the notion that to adopt and maintain a new behaviour requires information and 
the skills necessary for such behaviour change to occur. Individuals may need access to 
resources, motivation, encouragement, self-efficacy and, perhaps, support to act. In some 
 53 
University of Ghana http://ugspace.ug.edu.gh
cases, a diversion in some of the social norms guiding behaviour may be needed 
(USAID, 2017).  
 
In nutrition-related SBCC, some dietary practices may be habitual and, therefore, the 
need to disrupt certain habits may become inevitable in achieving positive dietary 
behaviour. There is, however, evidence to show that individuals are more likely to adopt 
a new behaviour if it does not deprive them of their group identity (USAID, 2017). Last, 
but not least, SBCC interventions need to be targeted for specific population groups for 
the desired impacts to be realized. This need partly rises because different population 
groups may have different perspectives of an issue and their concerns need to be built 
into the design of the intervention for these differences to be addressed (USAID, 2017). 
 
The effectiveness of social behaviour change interventions in addressing maternal and 
child nutrition- and health-related issues has been reported through a rapid review of 
literature in a number of socially-conservative settings where beliefs and traditions are 
entrenched and influence health behaviour outcomes (Tull, 2017).  The review gives 
examples of the use of SBCC in maternal, infant and young child nutrition in 
predominantly Muslim settings where community-based participatory approaches were 
employed as part of the process of changing behaviour and improving health outcomes 
of children and mothers.  
 
Unfortunately, the body of evidence on the effectiveness of SBCC in improving 
women’s dietary practices is limited even though the indication is that SBCC could be 
successful in improving the uptake of behaviour that is being promoted (Tull, 2017). The 
review concluded among other things that, depending on the social context of a 
 54 
University of Ghana http://ugspace.ug.edu.gh
particular behaviour of interest, the involvement of religious or traditional leaders, 
teachers, female associations or health professions among others are crucial in the 
successful implementation of a SBCC intervention that would bridge the gap between 
knowledge and practice. 
 
A number of hebaviour change theories or models have been propounded by behavoural 
scientists in guiding interventionists aimed at addressing individual and social problems 
affecting health outcomes (Davis et al., 2014; Glanz et al., 2008).  The application of 
behviour change theories is encouraged as part of the process of an intervention design 
and evaluation in addressing public health related issues (Davis et al., 2014). This is 
important for a number of reasons including identifying the antecedents of behaviour and 
the causal determinants of change is important for appropriate targeting of interventions 
and testing the effectiveness of the theory and possibly, the development of more 
effective ones.  
 
In a scoping review, Davis et al., (2014) identified eighty-two different kinds of 
behaviour change theories. While some of these theories are more appropriate for 
changing individual level behaviour, others are more focused on group or inter-personal 
behaviour. The health belief model by Rosenstock (1966); theory of planned 
behaviour/Reasoned action by Ajzen (1985); transtheoretical/stages of change model by 
Prochaska (1983) are examples of theories focusing on individual behavior while the 
social cognitive theory and the social ecological model of behviour are a few examples 
of theories that focus on group behaviour. The most commonly used theories being the 
transtheoretical/stages of change, theory of planned/reasoned action and the social 
cognitive theories (Davis et al., 2014).  
 55 
University of Ghana http://ugspace.ug.edu.gh
2.6.1 Social cognitive theory 
The social cognitive theory (Bandura, 1997) proposes that behaviour is influenced by 
personal, social and environmental factors. The personal factors relate to one’s cognition 
such as knowledge, awareness, attitudes and self-efficacy or agency. The lack of self-
efficacy or agency can be a strong barrier to behavioural change. An appropriate 
communication strategy can, however, help increase the individual’s and community’s 
sense of agency in a number of ways: by helping to teach relevant skills, providing clear 
instructions that make a particular behaviour look achievable and, last, by using 
examples to show how others have been able to accomplish a similar behaviour. 
 
The social factors, on the other hand, relate to social norms or group rules that determine 
what is deemed acceptable behaviour in the community. Social norms include people’s 
beliefs, values and behaviour. For instance, social norms can influence one’s thoughts 
and behaviour in a situation in which one is unsure of how to act and looks up to others 
for social approval on how to behave. Failure to act in accordance with social norms may 
result in social exclusion from the group. 
 
The third factor that influences behaviour, according to the social cognitive theory, is the 
physical environment in which one finds oneself. These physical environmental factors 
include access to resources, economy, technology and the legal system that will enable 
the new behaviour being promoted to be possible (McAlister et al., 2008). For instance, 
in promoting the consumption of certain food items, one should be certain that the 
community in which they are being promoted should have access to them. 
 
 56 
University of Ghana http://ugspace.ug.edu.gh
The social cognitive theory proposes a number of assumptions for an effective behaviour 
change intervention. First, there should be provision of information to increase people’s 
awareness and understanding of the risks involved with the unhealthy behaviour one 
wishes to change and the creation of an atmosphere in which people are convinced they 
have the ability to change such behaviour. Second, there should be development of the 
skills needed to translate the acquired knowledge into positive behaviour - thus, 
increasing people’s level of self-efficacy to practise the positive behaviour. The last 
assumption is the creation of a social support system to facilitate and sustain the newly-
adopted behaviour by adopting participatory approaches through collective efficacy, in 
which people are allowed to play active roles in the change process (McAlister et al., 
2008). Dietary interventions are, therefore, likely to have greater impacts if these 
concepts underpin them. 
 
2.6.2 Community Readiness Model (CRM) 
To make communities more effective partners in intervention activities –that is, to 
facilitate effective community participation –the community readiness model was 
designed by a group of researchers of the Tri- Ethnic Centre for Prevention Research 
(Edwards et al., 2000). Community readiness is the extent to which a community is 
ready and willing to take action on issues affecting its members. Community readiness 
has been defined as “the observable and psychological characteristics of a community 
that influence its ability to initiate change, including, but not limited to, organizational 
resources and the capacity and attitudes of the community” (Stanley, 2014; Sliwa et al., 
2011; Findholt, 2007; Edwards et al., 2000). The concept is synonymous with 
community agency – that is, the ability of a community to take action on a specific issue 
(Giddens, 1993). The Tri-Ethnic Centre for Prevention Research initiated this concept to 
 57 
University of Ghana http://ugspace.ug.edu.gh
plan preventive ways of dealing with community problems since all communities varied 
in their ability to implement and sustain interventions. Using the approach could help to 
place communities at their appropriate stages of readiness for more effective intervention 
programmes.  
 
The CRM is based on four main premises: first, community readiness is issue-specific - 
implying a community could be at different levels or stages of readiness for different 
community issues; second, the stage of readiness in a community regarding any 
community issue can be accurately assessed to determine the local capacity or otherwise 
to carry out an effective intervention; third, it is possible to increase a community’s 
readiness by increasing community knowledge and awareness of the issue and, last, 
determining a community’s stage of readiness for an intervention is important since it is 
a determining factor for moving the community from one stage of development to 
another (Stanley, 2014; Findholt, 2007).  
 
Community readiness is measured on five dimensions: a) community knowledge about 
the issue, b) community knowledge about existing efforts to address the issue, c) 
community climate regarding the issue, d) community leadership position on the issue 
and, finally, e) community resources to address the issue (Kostadinov et al., 2015; 
Stanley, 2014; Sliwa et al., 2011; Edwards et al., 2000). Thus, the model measures 
community knowledge, attitudes, efforts and resources in order to assess the level of 
readiness to effectively carry out an intervention. 
 
The model is based on a number of theories with the most notable being the trans-
theoretical theory of change formulated by Pochaska and Diclemente- also known as the 
 58 
University of Ghana http://ugspace.ug.edu.gh
stages of change theory - and the social action theory formulated by Warren (1978) 
(Edwards et al., 2000; Warren, 1978). The Tri-Ethnic group of researchers incorporated 
the principles of these theories into the community readiness model because, unlike 
individuals, communities are more complex in their processes of change and finally 
came out with 9 stages of change/readiness at which communities may find themselves 
at any point in time. The nine stages of readiness and their definitions are as follows:  
 59 
University of Ghana http://ugspace.ug.edu.gh
Table 2.2: Description of the nine stages of community readiness 
 Stage of readiness Description 
1 No awareness  At this stage of readiness, community members or 
leaders do not generally recognise the issue as a 
problem and may actually see it as part of their 
everyday life. 
2 Denial  Some community members, particularly some leaders, 
may recognize the issue as a problem but majority of 
community members may not and, therefore, they do 
not see the need to address it.  
3 Vague awareness There is a general feeling among some members of 
the community about the issue being a problem that 
needs to be addressed, but they are doing nothing to 
address it. 
4 Preplanning There is a clear recognition of the need to do 
something about the problem but nothing is being 
planned to address it. 
5 Preparation Community planning is underway to address the 
problem and there are modest community efforts to 
address it. 
6 Initiation There are some activities underway to solve the 
problem; leaders may be actively involved but 
community members may not be involved in 
addressing the problem. 
7 Stabilization A programme or two may be in place in the 
community and supported by community leaders to 
address the issue.  
 
8 Confirmation  There are policies and activities to address issue and 
or expansion community leaders and members lend their full 
support to seeing the programme through. Progress 
has been evaluated and lessons learnt incorporated 
into new programmes.  
 
9 Professionalization or There is detailed knowledge of problem; its  
community ownership prevalence, causes and consequences are well 
understood by entire community. Programmes are in 
place for either the entire community or targeted at 
specific groups of people. Community involvement is 
high and leadership completely supportive. There are 
also periodic evaluation processes to guide new 
programmes. Community ownership is high and 
community is held responsible for programmes aimed 
at addressing the problem.  
Source: Adapted from Findholt (2007), Edwards et al. (2000) and Plested et al. (2004) 
 
 60 
University of Ghana http://ugspace.ug.edu.gh
Depending on the stage of readiness in which a community finds itself, different 
strategies are proposed for making an intervention more focused in addressing the issue. 
For instance, for the first 4 stages of community readiness, creation of awareness about 
the nature, causes and consequences of the problem as well as the need for the 
community to do something about the problem are key among the intervention activities. 
 
The proponents of the CRM tested it in different settings with different focuses including 
drug and alcohol abuse, HIV/AIDS, child abuse, obesity and nutrition(Findholt, 2007; 
Edwards et al., 2000). Sliwa et al. (2011) used the model to select communities for an 
obesity prevention programme in eight cities in the USA. The results showed that overall 
readiness scores based on the five dimensions ranged from 2.97 to 5. 36 on the nine - 
point scale. The mean readiness score of 4.28 represented a preplanning stage of 
readiness. Recently, Pradeillis et al. (2016) applied the model in exploring the potential 
of using a faith-based organization as community agents for preventing over-weight and 
obesity among adolescents in South Africa. 
 
So far, the CRM has been extensively used in the USA but few studies in Canada and 
Australia have used it in issues related to drugs and obesity. By the end of December 
2016, only two studies from Africa had reported the use of the model in South Africa 
(Pradeillis et al., 2016) and Liberia (Kennedy, Johnson, & Alberto, 2004) with focus on 
adolescent weight management and HIV/AIDS respectively. Before the current study 
there was no evidence of the model being used in Ghana to help in the designing and 
implementation of nutrition-related interventions. 
 61 
University of Ghana http://ugspace.ug.edu.gh
2.6.3 Principles of adult learning 
The principles of adult learning are relevant in community-based interventions. 
Previously, assumptions on which adult learning were based was called pedagogy - a 
Greek version of “child-learning”. Later, the realization of the ineffectiveness of the 
application of pedagogy to adult learning led to the development of andragogy – a Greek 
version of “man-learning”. Andragogy, which was first used by a German educator 
called Alexandra Kapp in 1833, might be defined as the art and science of helping adults 
to learn or any form of adult learning (Kearsley, 2010). 
 
Malcolm Shepherd Knowles (1984) popularized andragogy and made a set of four 
assumptions about adult learning. Knowles’ first assumption is that, as one becomes an 
adult and matures, one becomes self-dependent and has a sense of self-directedness and 
expects society to perceive one as such. The second assumption is that as a person 
matures, he or she accumulates a rich reservoir of experiences that facilitates learning but 
failure to utilize an adult’s experiences creates a feeling of dejection or being insulted. 
Third, as a person matures, his or her readiness to learn becomes increasingly oriented to 
the developmental tasks of his or her social roles. This assumption implies that the adult 
educator should adopt a problem-solving rather than a subject-matter orientation to 
learning – for example, emphasis should be placed on the functional characteristics of 
topics in regard to participants’ occupations. Knowles’ last assumption is that as a person 
matures, the motivation to learn becomes internal rather than external. The implication of 
this assumption for adult learning is that the adult educator should motivate the learners 
with things that are relevant to them. 
 
 62 
University of Ghana http://ugspace.ug.edu.gh
In addition to his four assumptions, Knowles encouraged adult instructors to consider 
four main principles when leading adults in learning. First, adults should be involved in 
the planning and evaluation of their learning exercises; second, experiential learning 
should form the basis for the learning activities; third, subjects should have immediate 
relevance to and effects on their personal lives and, finally, adult learning should be 
problem centred rather than content centred (Knowles, 1984). 
 
Knowles’ adult-learning principles have been adopted in nutrition education (Lieb, 
1991). The author observed that to be an effective adult educator, one needs to 
understand the manner in which adults learn best. Since interventions target a wide range 
of individuals including adults, there is the need for interventionists to understand that 
adult learning is very different from child learning. The assumptions proposed by 
Knowles should, thus, not be overlooked in nutrition education and communication. 
Creating a comfortable environment for adult learners, valuing their rich experiences, 
encouraging participation and guiding them through the discussions rather than adopting 
the “student-lecturer” kind of relationship will enhance the effectiveness of nutrition 
interventions. To facilitate adult learning, the use of visual, audio and tactile aids has 
been suggested. 
 
2.7 Gaps in the reviewed literature 
Even though a number of socio-cultural barriers have been identified as factors for 
WRS’s poor dietary practices and quality in Ghana, the review indicated few 
community-based participatory interventions addressing these barriers. The review of 
literature showed that there were many nutrition- and dietary-related interventions in 
communities with health facilities as well as in those where access to health facilities 
 63 
University of Ghana http://ugspace.ug.edu.gh
was a great challenge. These interventions, however, concentrated on the immediate 
factors for poor diets such as inadequate knowledge on nutrition among members of 
communities at the expense of the need to address socio-cultural barriers such as 
women’s poor access to household food resources, pro-men preferential food allocation 
system and food taboos and restrictions pertaining to WRA. 
 
The interventions also failed to provide advocacy on the importance of women’s diets to 
the development of their children and the community at large. This study, therefore, 
sought to fill this gap by using a community-based participatory action approach to 
identify and address the socio-cultural barriers to women’s diets in a rural community 
called Yilkpene. The study was guided by the social cognitive theory and the community 
readiness model was used to assess Yilkpene’s level of readiness to implement an 
attitudinal and behavioural change communication intervention for addressing the 
identified socio-cultural barriers to women’s diets. 
 
 
 
 
 
 
 64 
University of Ghana http://ugspace.ug.edu.gh
CHAPTER THREE 
3.0 METHODS 
This chapter presents details of the study location, design and process, methods of data 
collection, sampling of participant categories, data recording, management and analysis 
and themes. Monitoring of the intervention activities, quality control as well as ethical 
considerations are also presented. 
 
3.1. Study location  
Yilkpene and Kpachilo were the two locations for this study. Both communities belong 
to Savelugu-Nanton District, one of the 26 districts of Northern Region. Its 
administrative capital, Savelgu, is located 20 kilometres west of Yilkpene and 22 
kilometres east of Kpachilo. Figure 3.1 shows a map of Savelugu-Nanton District with 
the two study communities. 
 65 
University of Ghana http://ugspace.ug.edu.gh
 
Figure 3.1: Map of Savelugu-Nanton District 
Source: Savelugu-Nanton District Assembly (2017) 
 
Savelugu-Nanton District occupies a total land area of about 1,760 square kilometres and 
has a total population of 1,596,159 (GHS, 2014). The district is predominantly rural with 
60% and 40% of its communities being rural and urban respectively. The main ethnic 
 66 
University of Ghana http://ugspace.ug.edu.gh
groups are Dagomba (88.8%), Gonja, Mamprusi and Ewe. Islam is the predominant 
religion (95.4%). 
 
The district experiences Harmattan, the dry and dusty North-East Trade Wind, from 
December to March. The mean annual rainfall is 600 mm, which is low but considered 
enough for a single farming season (GSS, 2014a). The low rainfall coupled with its 
irregularity has negative effects on crop yields and food security (GSS, 2014b). Less 
than 50% of the population have access to potable drinking water. The majority get their 
drinking water from boreholes, hand-dug wells and dams. Sanitation is poor and there is 
a high scope of open defaecation in the district. 
 
Savelugu-Nanton District is a patrilineal society and authority is traditionally vested in 
male chiefs, religious leaders and clan heads. Female-headed households form 3.1% of 
all households (GHS, 2014). Females typically have less access than males to education, 
health, land, capital and other assets and social amenities (GHS, 2014).  
 
The district has a total of 79 primary schools, 19 junior secondary schools, a special 
school for the deaf and dumb and one tertiary educatonal institution. The level of 
illiteracy is high (78.5% among females and 59.1% among male). School enrolment for 
boys is estimated at 56% as compared to 35% for girls (GSS, 2014a). 
 
Savelugu Municipal Hospital is the main health facility. Respective proportions of 
pregnant women with haemoglobin levels of less than 11gm/dl and still-births are 
estimated at 39% and 1.5%. Malnutrition rate among children below five years of age is 
about 17.3% while infant mortality prevalence is 6.3% (GHS, 2014).  
 67 
University of Ghana http://ugspace.ug.edu.gh
3.1.1 Selected study communities and justification for selection 
One study community was selected from each of two sub-districts. Savelugu Sub-District 
was purposefully selected because there was data from an earlier ethnographic study 
(Armar-Klemesu et al., 2014) on the cultural-ecological context of women’s nutrition. 
This data provided important background information for the current study. Yilkpene 
Community was purposefully selected from Savelgu Sub-District as the intervention 
community because it it is the only rural community out of the three communities 
included in the 2015 study.  
 
On the contrary, the selection of Kpachilo in Nanton Sub-District as the comparison 
community was due to similarity to the intervention community based on characteristics 
to those of the intervention community such as the size of population, dietary practices, 
occupations, education and access to social amenities. However, owing to a large 
number of communities which exhibit characteristics similar to those of the intervention 
community, a random sampling procedure was adopted to choose from the communities. 
This method involved writing the names of the four remaining sub-districts on four 
separate pieces of paper which were folded, put in a container and shuffled. One of the 
folded pieces of paper – that had the name of Nanton Sub-District -was selected as the 
comparison sub-district. The same random sampling procedure was used in the selection 
of Kpachilo as the comparison community. The distance between the two selected 
communities is forty-two kilometres. 
 
3.1.1.1 Yilkpene (Intervention Community) 
Yilkpene is a rural community in Savelugu Sub-District. In 2017, it had a population of 
1,099 including 329 women of reproductive age. The road linking Savelugu to this rural 
 68 
University of Ghana http://ugspace.ug.edu.gh
community is not paved and, therefore, not in a good state for vehicular traffic during the 
rainy season. The main livelihoods of the inhabitants are farming and trading in food 
crops. Women typically dominate trading activities in the community. They buy and sell 
farm produce in neighbouring communities. Other income-generating activities carried 
out in the community are mat and rope-weaving by men and the processing of rice, shea-
butter and groundnut by women. 
 
The community has no health facility but benefits from monthly visits by personnel of 
the Child Welfare Unit of Ghana Health Service. The nearest health centre to the 
community is at Mogla, which is about 10 kilometres away. Pregnant women in this 
community receive monthly ante-natal services at this health centre. Most deliveries are, 
however, done by traditional birth attendants (TBAs). Emergency cases are referred to 
Savelgu Hospital for medical attention. 
 
Yilkpene has no market infrastructure but the community is close enough to two 
prominent markets in Northern Region: Savelugu and Kumbungu. These two markets 
attract traders from elsewhere in northern Ghana. Members of the community buy and 
sell from these markets even though a few local traders sell petty commodities such as 
sugar, milo, salt, tea, bread and sugar-sweetened drinks. Operating only one primary 
school has contributed to the low levels of education in the community. After completing 
primary school, pupils can continue with their education at Kumbungu or Savelgu by 
choosing between walking daily for about two hours or moving out to live with relatives 
in these two communities and returning to Yilkpene only during week-ends and 
holidays. 
 69 
University of Ghana http://ugspace.ug.edu.gh
Yilkpene has since 2013 been connected to the national electricity grid. In addition, it 
takes about 45 minutes to access a bucket of potable water from any of the four 
boreholes in the community owing to the acute water problem that, in turn, is traced to 
the low water-table. Only one non-governmental organization (NGO) - Resiliency in 
Northern Ghana (RING) - was active in the community during the period of data 
collection. The main activity of RING was to encourage women to participate in making 
small-scale savings. Previously, there was Regional Advisory Information and Network 
Systems (RAINS), which provided the community with silos for storing grains.  
 
3.1.1.2 Kpachilo (comparison community) 
Kpachilo is a rural community in Nanton Sub-District. In 2016, the community had a 
population of 828 including 199 women of reproductive age. In addition to farming 
activities, the women engage in petty-trading activities such as food-vending and 
processing of shea-butter, groundnut and rice.  
 
Like Yilkpene, Kpachilo has no health facility, no market place and only a primary 
school. Nanton Health Centre and Savelgu Municipal Hospital serve as its health 
facilities and provide monthly child welfare and ante-natal care services. Home delivery 
by TBAs is the norm but complications and challenging deliveries are referred to 
Savelgu Hospital. In April 2016, the community had four boreholes as its main sources 
of water. The community has access to electricity even though most households are not 
connected to power. 
 
Unlike Yilkpene, there is heavy presence of NGOs at Kpachilo. In May 2016, the NGOs 
operating in the community included Alliance for a Green Africa (AGRA), World Vision 
 70 
University of Ghana http://ugspace.ug.edu.gh
(WV), Resiliency In Northern Ghana (RING), Regional Advisory Information and 
Network Systems (RAINS), Plan Ghana (PG), Innovation for Poverty Action (IPA), 
Complementary Basic Education (CBE) as well as Agricultural Development and Value 
Chain Enhancement (ADVANCE) project. Members of the community reported that 
World Vision used to promote the use and consumption of moringa and soya bean in the 
community through annual food and cooking demonstrations for women. The NGOs 
were also involved in the promotion of formal education. They supported children to 
enrol and remain in school, provided school uniforms, footwear as well as scholarships 
for students in senior high school and organized literacy programmes in Dagbani 
language.  
 
3.1.2 Target population 
The target population for the study at baseline and endline were members of selected 
identifiable groups in the two communities - Kpachilo and Yilkpene. The population 
conprised men and women aged 15 years and above. The target beneficiaries of the 
intervention, however, were members of Yilkpene Community. The intervention targeted 
women, men (including heads of households and husbands), community leaders (chiefs, 
clan heads, religious leaders, assemblymen and women and leaders of women’s groups) 
and community health volunteers. Both married and unmarried adolescent girls aged 15-
19 years of age also participated in the intervention because they were in their 
reproductive age, a critical stage to ensure optimal nutrition for their future reproductive 
roles. Boys in the same age group were, however, excluded because they were unlikely 
to be immediate beneficiaries of an intervention on WRA’s diets. At endline, data was 
collected from community members at Yilkpne and Kpachilo who took part in the 
 71 
University of Ghana http://ugspace.ug.edu.gh
baseline study but for participants at Yilkpene, men, women and adolescent girls who 
participated in the intervention activities there were targeted. 
 
3.2 Study design 
This study was a participatory action research (PAR) initially utilizing a formative 
research design involving a qualitative exploration of community food system, 
community members’ dietary knowledge and attitudes (perceptions and beliefs) as well 
as women’s dietary practices and reasons for these practices. With the results from the 
formative study, participatory rapid appraisal methods were used to design and pilot an 
attitudinal and behavioural change communication intervention with the active 
participation of key community representatives to address socio-cultural barriers to the 
consumption of optimal diets by WRA.  
 
This design was chosen because it was relevant to the study. Research methodologists 
perceive PAR to be both a research design and a philosophical world-view. As a research 
design, PAR seeks to empower communities to become more aware of their constraints 
by getting researchers to work with the people as opposed to imposing ideas on them 
(Savin-Baden and Major, 2013; Loewenson et al., 2014). PAR leans towards the 
advocacy/participatory paradigm, that believes that research must contain an action 
agenda for reform that may bring about social change (Creswell, 2007). This way, the 
approach enables researchers to generate information from the perspectives of 
communities who are affected by the issue under investigation to inform an intended 
intervention. The intervention undertaken involves the collective participation of the 
community with the view of improving the community to produce desired change.  
 
 72 
University of Ghana http://ugspace.ug.edu.gh
The socio-cultural nature of the study, therefore, called for a research design that is 
participatory so that key community members and other stakeholders would make inputs 
into the design and implementation process and become part of the change process 
without compromising their cultural values that identify them as a group. This way, 
people were more likely to adopt new behaviour/practices because they were active 
participants in developing the intervention strategies to drive home change (Savin-Baden 
and Major, 2013). 
The study consisted of five phases as depicted in Figure 3.2. 
 
 
 
 
 
 
 
 
 
 
 73 
University of Ghana http://ugspace.ug.edu.gh
Exploratory qualitative  Assessment of the food system and gender dimensions  
study  Assessment of community dietary knowledge and attitudes 
(perceptions and beliefs)  
   Assessment of women’s dietary practices and drivers   
 
 
Analysis of gaps by  Determination of community readiness to address socio-cultural 
community consensus challenges/barriers  
  Identification of barriers to optimal diets for women  
  
 
Development of  Development and contextualization of messages 
 
intervention  messages  Identification of channels  
together with community  Identification of key influencers  
 
 
 
 Implementation/  Intensive delivery of intervention for 4 months (17 weeks) 
dissemination of  Three (3) additional months of diffusion 
 
intervention  
 
 
 
 Evaluation of intervention   Reassessment of the food systems and gender dimensions 
 Reassessment of community dietary knowledge and attitudes 
  
 Re-assessment of women’s dietary practices and drivers  
  Participants’ perceptions of the intervention 
 
Figure 3.2: Study design  
 
 74 
 
 
 
  
University of Ghana http://ugspace.ug.edu.gh
3.3 Study process 
The study went through a process consisting of five sets of activities (Figure 3.2) after 
the chiefs and elders had granted the study team entry into the communities. The 
assemblymen of both communities led the research team through the community entry 
processes. The community health volunteers from each of the study communities 
facilitated the mobilization of members of the communities for the various interviews 
and group meetings. 
 
3.3.1 Phase 1: Qualitative exploration of the socio-cultural practices  
The exploratory qualitative study was the first phase of the study. In this phase, the local 
food system in the two communities was explored. Women’s participation in the food 
system was explored to understand socio-cultural challenges that impeded their access to 
adequate food. The study focused on the four-stage food system pathway consisting of  
food production and use of farming inputs; food processing, preservation, storage and 
preparation; food distribution and utilization/consumption at the household level. 
Women’s participation in each stage of the food system was explored in detail including 
gender-based issues within each of the activities. Community members’ dietary 
knowledge and attitudes (perceptions and beliefs) regarding healthy diets, food taboos 
and food restrictions were explored to identify gaps emanating from their knowledge, 
misconceptions and belief system.  
 
Dietary knowledge was measured by participants’ ability to distinguish between the 
three food groups as indicated by the Department of Women in Agriculture of the 
Ministry of Food and Agriculture. The three food groups are energy-giving foods such as 
cereals and grains, roots and tubers; body-building foods  such as livestock products, 
 75 
University of Ghana http://ugspace.ug.edu.gh
legumes, nuts and seeds and protective foods such as fruits and vegetables. Knowledge 
was also measured by their sources of information on healthy diets as well as knowledge 
of foods that enhanced WRAs’ health.  
 
Attitudes (perceptions and beliefs) that were explored included what participants thought 
constituted health and healthy diets, their food preferences and reasons for liking and 
dislikng some foods, assessments of women’s dietary adequacy in terms of quality and 
quantity and the effects of women’s poor diets on them. Also, participants’ perceptions 
of food shortage and its effects were explored. Participants’ atttitudes to food such as 
food taboos and restrictions that affected WRA and their levels of adherence to such 
taboos and rstrictions were explored. Finally, participants’ perceptions  of the effects of 
food taboos and restrictions on women’s diets were assessed. 
 
The third issue explored during this phase was women’s dietary practices and factors that 
influenced their choices of food. The study explored women’s dietary practices by 
documenting the meals consumed in the 24 hours preceding the day of data collection, 
frequency of meal consumption, the ingredients contained in each meal, their methods of 
preparation and members of the households who consumed the same foods. The sources 
of other foods consumed within the 24 hours were explored.  
 
The main issue of interest in the first phase of the exploratory study was the 
identification of personal and socio-cultural factors influencing women’s dietary 
practices in the communities. These issues were fed into the design of the intervention 
activities. The outcome indicators that were measured in the intervention were dietary 
knowledge and attitudes (perceptions and beliefs). It was envisaged that once the 
 76 
University of Ghana http://ugspace.ug.edu.gh
intervention was able to address the gaps in knowledge and perceptions, the improved 
knowledge and perceptions should lead to attitudinal change and, ultimately,  improved 
quality of women’s diets. 
 
3.3.2. Phase 2: Analysis of gaps by community consensus  
Following the exploratory study, two separate activities were held in the intervention 
community to reach a consensus on, first, whether the community was ready to address 
socio-cultural challenges to women’s diets and, second, what activities to implement to 
address the challenges. The first set of activities consisted of key informants’ interview 
with six key members of the community and four focus group discussion sessions to 
assess, through the seven steps proposed for conducting a CRA (Stanley et al., 2014) the 
level or stage of readiness at Yilkpene for implementing an intervention addressing the 
identified socio-cultural challenges to women’s diets.  
 
The CRM questions (Stanley, 2014) were adapted to suit the focus of the current study. 
An interview guide was developed along the five CRM dimensions and administered to 
six key informants; but, to ensure that the opinions of a larger number of members of the 
community were represented (Pradeillis et al., 2016), four separate focus group 
discussions sessions, with two sessions each comprising men and women, were held 
using the same interview guide administered to the key informants. The discussion 
sessions had an average of 8 participants. The specific issues that were discussed with 
participants are indicated in the interview guide presented in Appendix C. Using 
descriptive statements on anchored scores, each dimension was scored. The mean for the 
collated scores for each dimension awarded by the six key informants interviewed was 
 77 
University of Ghana http://ugspace.ug.edu.gh
used to produce a final community readiness score. This final score was rounded down 
to a whole number that corresponded to one of the nine stages of community readiness. 
 
The community readiness assessment was followed by a presentation of the baseline 
results to community stakeholders at a two-day meeting. The aim of the meeting was to 
reach a consensus on three issues: first, whether the research findings represented the 
cumulative views of the community regarding women’s diets; second, the specific socio-
cultural challenges and, third, the approach to be used in addressing them. The 
stakeholders comprised the district nutrition officer, district agricultural extension 
officer, a community health nurse and community opinion leaders including the 
assemblyman for the zone, community health volunteers, a women’s leader, the 
community Imam and chief’s elders. In all, 24 members, including the research team, 
participated in the meeting, which was held in the intervention community. The 
identified challenges which were presented to the community for deliberations are 
presented in Table 3.3.  
 
 
 
 
 
 
 
 78 
University of Ghana http://ugspace.ug.edu.gh
Table 3.1. Identified challenges to women’s optimal diets  
Climate and Community-level Household-level Individual-level 
farming challenges challenges challenges challenges 
(natural causes) 
1. Seasonal nature 1. Lack of dams and 1. Household foods 1. Diets lacking 
of crops irrigation facilities not diversified variety (almost 
2. Changes in (Lack of dry-season 2. Women’s limited one-half of women 
rainfall pattern farming) control over household not meeting MDD-
(irregular and short 2. Sale of most food resources W) 
duration) nutritious farm produce 3. Women burdened Inadequate 
3. Poor soil fertility (legumes and animal- with providing other quantities of body-
(soil erosion and source foods) food resources for building foods 
continuous 3. System of land household (legumes and fish) 
cultivation on same ownership consumption in diets 
land) 4. Observance of food 4. Preferential food 2. Low 
4. Dwindling sizes taboos and restrictions allocation practices consumption of 
of farmland 5.Large sizes of (Food-sharing, meat- fruits and raw 
(through rapid households  sharing and parts vegetables 
increases in 6. Lack of jobs and given to women)   
population and credit facilities 5. Limited 
urbanization) 7. Inadequate food- preservation practices 
 vending activities (drying of limited 
8. Poor environmental number of vegetables, 
sanitation little fermentation, 
9. Inadequate direct sun-drying) 
knowledge about 6. Inadequate 
women’s diets and their knowledge of food 
relationship to their preparation methods 
reproductive health (overdoing of 
10. Misperceptions vegetables 
about the health 7. Women’s heavy 
benefits of some foods workload (18 hours 
day- time activities 
and 6 hours night 
sleep) 
 
The data presented in Table 3.3 shows that the challenges to women’s diets in both 
communities came from four inter-related categories of factors: climate and land, 
community-, household- and individual-level challenges; however, because the study 
was focused on addressing socio-cultural challenges, only issues related to the 
community, household and individual-levels were considered for appraisal at the 
community stakeholders’ meeting.  At the end of the appraisal meeting with Yilkpene, 
 79 
University of Ghana http://ugspace.ug.edu.gh
the intervention community, six specific challenges were selected to be addredded in the 
intervention.   
 
The chalenges were: inadequate knowledge of nutritive qualities of different foods and 
importance of women’s diets to their reproductive health; misperceptions about health 
and healthy foods; poor methods of preservation, storage and preparation of food; poor 
methods of preservation, storage and preparation of food; poor methods of preservation, 
storage and preparation of food; Community beliefs and practices contributing to 
women’s poor dietary practices and Women’s lack of sufficient power and control over 
household food resources While the first two challenges were directly linked to socio-
cultural practices, the other four were linked with issues of knowledge and attitudes 
(perceptions and beliefs) related to diets and healthy diets. Even though the last two 
challenges were not linked exclusively with socio-cultural practices, they were 
challenges that the community perceived could have potential impact on their diets and 
health hence, their inclusion in the intervention.  
 
Two intervention activities (solutions) were prioritized by stakeholders to address the six 
selected challenges to WRA’s diets: The first component of the intervention activities 
was nutrition and dietary education with emphasis on four issues: first, importance of 
consuming a variety of foods in their diets; second, importance of adolescent nutrition; 
third, importance of women’s diets and their reproductive health and, fourth, improved 
techniques of food preservation, storage and preparation. The aim of the nutrition and 
dietary education was to enhance self-efficacy among members of the community.  
 
 80 
University of Ghana http://ugspace.ug.edu.gh
The second component of the intervention activities was advocacy on two issues: first, 
importance of women’s access to and control over productive and household food 
resources to enhance household and women’s diets and, second, effects of food taboos 
on women’s dietary quality. Advocacy was meant to sensitize and create collective –
efficacy for social change. 
 
3.3.3. Phase 3: Development of intervention together with community  
Development of messages and contextualization with community: The identified 
challenges to women’s diets were translated into a communication strategy and 
culturally-appropriate messages targeted at addressing them. The activities carried out in 
this phase comprised development of intervention messages together with the 
community and pre-testing of the messages before the actual implementation. The 
schedule of communication activites implemented to address the six challenges are 
presented in Appendix N. A total of 28 messages targeted at the six challenges were 
communicated to the audience. This section, however, presents a summary of the 
challenges, the current state of behaviour, interventions (solutions) implemented and 
their respective key messages in Table 3.2. 
 81 
University of Ghana http://ugspace.ug.edu.gh
Table 3.2: Challenges, current state of behaviour, interventions and their respective 
key messages 
Challenge Current behaviour/practices Intervention Key message 
Challenge 1: Inadequate Women’s diets were found to Nutrition education with Women need a variety of 
knowledge of nutritive be limited in variety. The main emphasis on 1) foods in their diets for good 
qualities of different groups of foods consumed were Importance of consuming health and well-being. 
foods and importance of grains, some leafy vegetables, a variety of foods in their Combining and consuming a 
women’s diets to their anchovies and legumes but the meals, 2) importance of variety of foods in our meals 
reproductive health amounts of anchovies, legumes adolescent nutrition and improve the quality of our 
and green-leafy vegetables 3) women’s diets and diets. 
consumed were small resulting their reproductive health Women’s diets are important 
in poor qualities of their diets. and 4) Training on for their health and 
- Although there was awareness techniques of reproductive health 
of the effects of poor diets on preservation, storage and outcomes. 
women’s health, knowledge of preparation of food 
effects of poor diets on  
women’s reproductive health 
was limited. 
   
Challenge 2: There were some All foods are healthy 
Misperceptions about misperceptions about healthy depending on how you 
health and healthy foods diets, thus, influencing their conbine and consume them. 
preferences for foods. 
   
Challenge 3: Poor Post-harvest food preservation Improving preservation, 
methods of preservation, and storage systems were storage and preparation of 
storage and preparation of inadequate resulting in a lot of food enhances the nutritive 
food infestation of crops by fungi qualities of diets. 
and weevils. Overdoing of 
vegetables and little use of 
fermented foods deprived 
women of their nutritive values. 
   
Challenge 4: Poor Levels of hygiene and Let’s keep our environment 
methods of preservation, sanitation in communities and clean to avoid contracting 
storage and preparation of households were unsatisfactory. diseases. 
food 
    
Challenge 5: Community A number of food taboos and Advocacy on 1) Address food beliefs, taboos 
beliefs and practices food restrictions negatively importance of women’s and discriminatory practices 
contributing to women’s affected women’s diets. These access and control over to help improve women’s 
poor dietary practices food taboos and restrictions productive and household diets. 
were linked to both animal- and food resources to enhance 
plant-source foods. household and women’s 
  diets and 2) effects of  
Challenge 6: Women’s Women had limited control food taboos on women’s Empower women to have 
lack of sufficient power over food resources at the dietary quality control over food resources 
and control over household level even though so that they can improve 
household food resources they were expected to their dietary quality. 
contribute to the provision of 
food to their households. 
Women also had limited access 
to productive inputs such as 
land and labour. 
 
 82 
University of Ghana http://ugspace.ug.edu.gh
Identification of channels for delivering messages: The stakeholders together with the 
research team identified six channels that were used in designing and delivering the 
messages as well as implementing all the activities planned for the intervention. These 
channels were community durbars, group meetings with members of the community, 
home-visits, food demonstration sessions, clean-up exercises and the composition of 
local songs targeted at women’s diets. The scheule of the channels ia presented in Table 
3.3. 
 83 
University of Ghana http://ugspace.ug.edu.gh
Table 3.3: Channels for delivering  intervention 
Channels for Purpose/aim Frequency Target 
communication 
1. Community To share baseline findings, Two durbars: one at the Whole community 
Durbar sensitize community on beginning and the other at the 
intervention activities and seek end of the 17-week activities 
their support for the 
implementation and sustenance 
of intervention  
2. Training To build capacity of Two days of training and pre- Thirteen-member 
implementation team especially testing of messages implementation team: eight 
the community partners for the selected members of the 
delivery of the intervention community, 4 research 
messages assistants and the principal 
student-researcher 
3. Weekly group To improve community 24 group meetings: 12 group Men, WRAs, older women 
meetings members’ knowledge and meetings each for men and and adolescent girls 
attitudes in respect of women’s women 
diets 
4. Home-visits To discuss and encourage One home-visit per All 56 households in 
participation in the group compound/household community 
meetings, encourage 
adoption/practice of BCC 
messages and to address 
misconceptions and other issues 
arising from the group meetings 
5. Food To improve community Two sessions Men and women 
demonstration members’ knowledge on 
sessions methods of food processing and 
preservation, improve 
household cooking practices 
through enriching household 
diets with legumes in particular 
and cooking vegetables for 
shorter periods of time 
6. Composition of To sustain participation and Throughout the Whole community 
songs in the local interest in the intervention  implementation period; 
language two local songs composed 
7. Clean-up To emphasize the importance Four exercises: one exercise Whole community 
campaign of good environmental per month for 4 months 
sanitation and to sustain interest 
in intervention activities 
 
An implementation plan was developed with the communtiy stakeholders to guide the 
implementation process (Appendix L). The plan spelt out the various activities and 
topics that were discussed every week. The meetings were held weekly- that is, on 
Saturday mornings from 10 a.m. to 11 a.m. During each meeting, one topic was 
discussed and promoted for the week. The group was led to review the messages the 
following week before the introduction of a new topic. A total of 138 women and 91 
men participated in the group meetings. A register was developed to keep tract of 
 84 
University of Ghana http://ugspace.ug.edu.gh
attendance (participation in the meetings). Each compound in the community was visited 
once during the period of intervention and at least a woman  from each of the households 
attended the weekly meetings. 
 
Two sessions of food demonstration exercises were held. One session concentrated on 
training participants on the use of local and waste materials to prepare insecticides and 
improve soil fertility, prevention of post-harvest losses and improved ways of preserving 
crops to avoid aflatoxin. Additionally, participants were taken through ways of 
preserving vines of the orange-flesh sweet potato. A total of 138 members of the 
community (89 females and 49 males) participated in this session. The other session 
concentrated on demonstrating the cooking of some common dishes in the community 
through the addition of soya bean flour to enrich their nutritive contents and cooking 
vegetables to minimize nutrient loss. A total of 174 members (102 females and 72 males) 
participated in the cooking demonstration session. The ingredients and methods adopted 
for the cooking demonstrations are presented in Appendix O. 
 
To sustain participation and interest in the intervention, the entertainment group in the 
community composed two songs which were sung amidst drumming and dancing during 
the course of some of the weekly activities. The lyrics of the songs were based on the 
intervention messages and included issues of sanitation, importance of women’s diets 
and food in general. In delivering the intervention activities, four principles of adult 
learning were adopted (Lieb, 1991). These principles were respect, adults’ autonomy and 
self-directedness, experience and knowledge and, last, their readiness and orientation to 
learn. The research team acknowledged the fact that participants had a wealth of 
experiences regarding the focus of the study and this knowledge was utilized during the 
discussions and interactions with them. 
 85 
University of Ghana http://ugspace.ug.edu.gh
Identification of community influencers: The stakeholders selected eight members of 
the community to join the research team to form the implementation team for the 
execution of the intervention. The eight persons selected were the local Assemblyman, 
the assistant Chief Imam, two community health volunteers, a women’s leader, leader of 
the youth sanitation group, the leader of the community entertainment group and the 
chief-crier. These community-appointed members of the team were responsible for 
mobilizing people for the subsequent intervention activities and delivering the weekly 
messages with support from the research team. During the two-day stakeholders’ 
meeting, snacks were made available to the intervention team.  
 
3.3.4. Phase 4: Implementation of intervention 
The intervention was a 17-week community-based attitudinal and behavioural change 
communication intervention, which utilized existing community groups and resources in 
the design and implementation process. This exercise involved the participation of 
identified community stakeholders in the development of context-appropriate 
educational messages on women’s diets and advocacy on women’s empowerment 
(specifically, women’s control of food resources). The messages were targeted at 
identified socio-cultural norms as well as personal factors that constituted barriers to the 
achievement of optimal diets for women in the intervention community. Selected 
community members assisted the research team to delivery the educationation and 
advocacy messages during the intervention activities. The intervention was embedded in 
a pre- and post-intervention design with a comparison group - a quasi-experimental 
design (Fisher & Foreit, 2002). The quasi-experimental design is illustrated  in Figure 
3.3 as follows: 
 
 86 
University of Ghana http://ugspace.ug.edu.gh
 
Two rural communities selected from 
Savelgu and Nanton sub-districts (N =182) 
 
 
 
Intervention community: Yilkpene Comparison community: Kpachilo 
   
 
 
 
 
Baseline data collection using multi Baseline data collection using multi 
 qualitative methods  qualitative methods 
   
 
 
 Community readiness assessment 
 Stakeholders consensus meeting and 
 development of intervention activities 
  
 
 Implementation of Intervention with 
active participation of community 
 members 
  
 
 Endline data collection using same data Endline data collection using same data 
collection tools as at baseline 
 collection tools as at baseline 
 Community perceptions of intervention 
 
Figure 3.3: Illustration of quasi-experimental design for implementing attitudinal 
and BCC intervention 
 87 
University of Ghana http://ugspace.ug.edu.gh
The quasi-experimental design was chosen for three main reasons. First,  the qualitative 
nature of the entire study did not require randomization in the selection of participants 
and, therefore, a design which did not require random assignment of participants was 
most appropriate. Second, the financial resources needed to carry out a true experiment 
such as a randomized controlled trial (RCT) were beyond the researcher’s reach. Last but 
not least, the time needed to complete a truly randomizsed controlled trial was not 
feasible given the researcher’s time frame for completion of the thesis. 
 
The venues for the meetings to deliver the intervention activities were four locations 
where different groups in the community usually held their meetings. The first venue 
was the chief’s palace where the women (“amasachina” group) usually held their 
meetings. The second venue was a school block which served as the venue for both the 
stakeholders’ meeting and training of the members of the implementation team. The 
community durbars and food demonstration sessions were also held on the school 
premises. The space under a big baobab tree served as the venue for the meetings of both 
the men’s group and demonstration lessons on proper ways of preserving food and 
preparing local insecticides. The last set of venues consisted of the homes of individuals 
or compounds involved in the home-visits.  
 
3.3.5 Phase 5: Evaluation of intervention 
Post intervention data collection was carried out between August and November 2017. A 
period of four weeks was devoted to collecting data. Transcription and data management 
and analysis lasted a period of three months. Both implementation process and outcome 
evaluations were reported. The implementation process evaluation aimed to determine 
whether the intervention was delivered as planned. The attitudinal and BCC strategy that 
 88 
University of Ghana http://ugspace.ug.edu.gh
contained the educational messages for the intervention and its implementation plan as 
well as the weekly field notes during the implementation was used to assess the fidelity, 
dosage, target and unforeseen bottlenecks of the intervention and how they were 
addressed. Additionally, participants’ assessment of the intervention was carried out.  
 
3.4 Methods of data collection  
Qualitative data collection methods were employed for this study. A number of 
qualitative approaches to data collection were reviewed to assess their suitability for this 
study. A cocktail of qualitative methods were used for collecting the different but 
relevant sets of information from different categories of study participants to address 
specific objectives. The methods used were key informant interviews, focus group 
discussions and qualitative 24-hour dietary recall with in-depth interviews as well as 
household and community observations.  
 
The purpose for using a mix of qualitative methods in this study was to ensure that the 
relevant issues at stake were explored in detail to understand women’s dietary behaviour 
more holistically. For instance, while the key informant interviews, focus group 
discussions and community observations focused on community-level factors affecting 
women’s dietary behaviour, the qualitative 24-hour recall with in-depth interviews as 
well as the household observations focused on household practices and women’s daily 
experiences related to their dietary practices. Triangulating data from these multiple 
sources gave a better understanding of women’s dietary situation in the study 
communities. Understanding community, household and individual factors facilitated the 
design of a participatory behavioural change communication strategy for addressing 
identified socio-cultural challenges to women’s optimal diets in the study communities.  
 89 
University of Ghana http://ugspace.ug.edu.gh
Presented in Table 3.4 are the data collection methods together with the specific tools 
used in collecting information from participants. The table further indicates the phases of 
the study that utilized specific methods in collecting information. 
Table 3.4 Data collection methods, tools, phases in which used and information 
collected 
Data Data Phases in Data collected 
collection collection which used 
method tool 
Key- Key- 1, 2 and 3 Socio-demographic characteristics; food system 
informant informant components (food production, processing and 
interview interview preservation, distribution and utilization/consumption); 
guide control of food resources at the household level; food 
availability and access; community readiness for an 
intervention addressing challenges to women’s diets 
and community assessment of the intervention  
Focus group Focus group 1, 2 and 3 Participants’ socio-demographic characteristics; 
discussion  discussion community knowledge, and attitudes (perceptions and 
guide  beliefs) regarding food and health, awareness of food 
taboos and restrictions and levels of adherence; effects 
of food taboos on women’s diets; perceptions of 
healthy foods for WRA; sources of dietary and health 
information; community food preferences; community 
readiness for an intervention addressing challenges to 
women’s diets and community assessment of the 
intervention  
24-hour 24-hour 1 and 5 Detailed description of all foods consumed at 
dietary recall dietary recall breakfast, lunch and supper as well as water and 
guide beverages consumed 24 hours preceding the interview; 
methods of preparing these foods, in- and out-of-home 
eating habits 
In-depth In-depth 1 and 5 Sources of household food, decision-making and 
interview Interview consumption arrangements at the household level 
guide 
Community Community Throughout Members’ daily routine activities regarding their social 
observation observation study  and economic activities; sources of main staples in the 
check list communities; market days, outdooring ceremonies as 
well as a funeral ceremony  
Household Household 1 and 5 Composition of households, hygiene practices, 
observation observation women’s daily chores and their time-use; household 
check list food preparation and sharing; meal times and 
frequency of meals consumed by WRA  
 
Key-informant interview participants: The key informants comprised two categories. 
The first comprised key informants within the community, namely community health 
 90 
University of Ghana http://ugspace.ug.edu.gh
volunteers, chief’s elders, chief imams, male and female farmers, women’s leaders and 
traditional birth attendants (TBAs). The second category comprised the district 
agricultural and nutrition officers, a community health nurse and local representatives of 
NGOs who work in collaboration with the community. Some key informants were 
interviewed in their homes while others were interviewed at their workplaces.  
 
Focus group discussion participants: Separate focus group discussion sessions were 
held with men, women of reproductive age, young, unmarried adolescent girls aged 15 to 
19 years and elderly women (aged 50 and above). The group of young adolescent girls 
who were recruited for the FGDs included both in-school and out-of-school girls. Each 
session comprised an average of 9 participants.  
 
Observation targets: Direct observations were done at both the community and 
household levels in respect of women’s dietary practices. The community observation 
lasted the whole period of the study (from April 2016 to November 2017). The 
household observations lasted 12 hours- from 6a.m. to 6p.m. In all, 10 households- that 
is, 5 households each in the intervention and comparison communities -were observed. 
 
Qualitative 24-hour dietary recall: The qualitative 24-hour dietary recall involved a 
recall of the foods consumed by women aged 19- 49 years in the immediate 24 hours 
preceding the day of data collection. The aim of the recall was to determine WRA’s 
achievement of the minimum dietary requirement for women (MDD-W)- that is, the 
average dietary quality or nutrient intake). Twenty women each from the intervention 
and comparison communities were involved in the recall - making a total of 40 women. 
These same 40 women were followed up during the post-harvest period (that is, 
 91 
University of Ghana http://ugspace.ug.edu.gh
November), six months after the baseline study was done in May – June, to collect the 
qualitative 24-hour recall data to assess the effect of seasonality on the quality and 
frequencies of foods consumed (FAO, 2011). Each woman’s dietary information was 
collected four times: it was collected at baseline in May-June 2016 - when it was 
perceived there was food shortage – and, again, in November 2016 - when food was 
perceived to be in abundance. The recall was repeated at endline in May-June and 
November 2017. 
 
In-depth interview: In addition to the information collected through the 24-hour dietary 
recall from the 40 WRA, information on sources of acquisition of food for the 
households as well as the decision-making and consumption arrangements was collected 
through in-depth interviews. This information was collected soon after the 24-hour recall 
sessions for each of the 40 women in only May-June at both baseline and endline.  
 
The data collection tools that were used at baseline and endline included a key-informant 
interview (KII) guide, a focus group discussion (FGD) guide and household and 
community observation check lists. The other tools were qualitative 24-hour recall tool 
and an in-depth interview guide. In addition, a community readiness assessment guide 
was adapted from the CRM and used to assess the intervention community’s readiness 
for an intervention. This tool was administered after the initial baseline study. At endline, 
a community assessment guide was, again, included for participants in the intervention 
community to assess the intervention. These guides were administered to both key 
informants and focus group discussion participants (Appendix C). The use of interview 
guides and checklists to guide the data collection process was relevant in ensuring that 
topics of interest to the issues being discussed were not overlooked. Additionally, it 
 92 
University of Ghana http://ugspace.ug.edu.gh
ensured that in respect of each of the topics, as much detail as possible was gathered with 
a lot of probing and, in some cases, prompting for further explanations. All the data 
collection tools are presented in Appendices A to H. 
 
3.4.1 Sampling of participant categories 
At baseline, the study used purposive sampling owing to the need to provide information 
relevant to the issues at stake. There was, therefore, the need to use community 
volunteers to assist the research team to identify the appropriate potential informants.  
 
At endline, recruitment of participants for the various categories of interviews was, 
again, purposively done. Regarding the 24-hour recall, household observations and in-
depth interviews, the same participants who were recruited at baseline were used for the 
assessment phase. The key-informant participant category comprised, as much as 
possible, either participants who were interviewed at baseline or their representatives. 
The focus group discussion participant category in the intervention community, however, 
comprised 3 categories of participants: those who participated in less than 6 group 
meetings; those who participated in either 6 or 7 group meetings and those who 
participated in 8 up to 12 group meetings. The essence of grouping the FGD participants 
on the basis of exposure or participation to group meetings was to assess the correlation 
between this variable (scope of exposure/participation) and levels of knowledge on 
issues discussed at the group meetings. The FGD sessions were organized for men and 
women separately. Table 3.4. gives a break-down of the numbers of participants 
interviewed by methods.  
 93 
University of Ghana http://ugspace.ug.edu.gh
Table 3.5:  Methods, participant categories, number of interviews, samples of participants in baseline, endline and community readiness 
assessment (CRA) phases of the study 
Participant categories  Number of participants in baseline phase  Number of Number of participants in endline phase  TOTAL 
participants in CRA 
 Interv. Comp. Sub-Total Interv. Interv. Comp. Sub-  
 comm comm comm Comm comm Total 
 
KEY INFORMANTS          
Community health volunteer 1 1 2 1 1 1 2 5 
Chief’s elder 1 1 2  1 1 2 4 
Chief Imam - 1 1  - 1 1 2 
District agricultural extension officers 1 - 1 1 1  1 3 
District nutrition officer 2 - 2 1 1 - 1 4 
Community health worker 1 - 1 1 1 - 1 3 
NGO representative in community 1 - 1 1 1  1 3 
Women’s leader 1 1 2  1 1 2 4 
Male farmer 1 - 1  1 - 1 2 
TBA - 1 1  - - - 1 
Assemblyman 1 1 2 1 1 - 1 4 
Sub-total 10 6 16 6 9 4 13 35 
FGD SESSIONS         
Men 2 2 4 2  1 1 7 
WRA 2 2 4 2  1 1 7 
Elderly women 1  1     1 
Adolescent girls 1 1 2   1 1 3 
Attendance at 8 or more meetings     2  2 2 
Attendance at 6 and 7 meetings     2  2 2 
Attendance at less than 6 meetings     2  2 2 
Total number of Groups 6 5 11 4 6 3 9 24 
Total number of participants 48 39 87 33 52 20 72  
Average number of participants in a group 8 8  8 9 7   
Qualitative 24-hour recall with in- depth interviews-WRA 20 20 40*  20 20 40*  
Observation of women’s time-use 5 5 10*  5 5 10*  
Community observations         
Total (all methods) 78 65 143 39 81 44 125  
 94 
University of Ghana http://ugspace.ug.edu.gh
It is shown in Table 3.7 that key participant categories in the communities that were 
involved in the baseline, community readiness assessment and endline phases of the 
study included WRA and, particularly, pregnant and lactating women. In addition, male 
and female heads of households, elderly members of the community, chief’s elders, chief 
imams, farmers, local government officers (district agricultural extension and nutrition 
officers) and NGOs working with the women in the communities were included. 
 
One hundred and forty-three (143) participants constituted the sample for the baseline 
phase in addition to 39 participants for the community readiness assessment- making a 
total of 182 participants for the study. At endline, 125 of the 143 participants at baseline 
participated. The intervention community had 78 and 81 participants at baseline and 
endline phases respectively and the comparison community had 65 and 44 participants at 
baseline and endline respectively. Table 3.6 presents the total number of interviews 
conducted at baseline and endline. 
Table 3.6: Number of interviews by methods at baseline, CRA and endline 
       Baseline CRA               Endline Total 
Yilkpene Kpachilo Yilkpene Yilkpene Kpachilo 
KII 10 6 6 9 4 35 
FGD 6 5 4 6 3 24 
24-hour recall+ IDI 20 20 _ 20 20 80 
Total No. of 36 31 10 35 27 139 
interviews/transcripts 
 
The composition of the participants on the basis of the methods used in gathering the 
data shows that there was an aggregate of 139 interviews. In terms of types, these were 
 95 
University of Ghana http://ugspace.ug.edu.gh
35 KIIs, 24 focus group discussion sessions and 80 24-hour recall with in-depth 
interviews. 
 
3.5 Data recording, management and analysis 
A total of 139 interviews were conducted with most of them in Dagbani and only three in 
English. The interviews were audio-recorded with participants’ consent, transcribed 
verbatim into English and imported into Nvivo for data management. A note-taker took 
notes during the interview. At the end of each day, the notes were coded to identify 
emerging issues that required follow-up.  
 
“Interim” analysis was done alongside data collection to reflect on emerging issues, to 
refine the questions and to summarize what was said during the interviews (Pope, 
Ziebland, & Mays, 2000). After data collection, thematic analysis was used to explore 
more intensively how community, household and personal factors affected the quality of 
women’s diets (Braun & Clarke, 2006). The thematic analysis consisted of six steps - 
namely, familiarization with the data collected; generation of initial codes; searching for 
themes; reviewing themes; defining and naming themes and producing the results. 
 
The steps were followed by reading and rereading the transcripts, listening to the 
recorded interviews and taking note of frequent expressions from participants; for 
instance, “food that makes one drink a lot of water is healthy” and “food that makes you 
sweat a lot after eating is healthy” were coded as perceptions of healthy foods. All noted 
expressions were coded systematically across the entire data set. Verbatim responses 
were collated and assigned to each code. After this, all the coded responses were collated 
into initial themes and sub-themes. Each major theme was given a working definition 
 96 
University of Ghana http://ugspace.ug.edu.gh
and imported in Nvivo so that the various responses from the transcripts that related to 
each theme were pooled together. Finally, the major themes and sub-themes formed the 
basis for reporting the results. Data recording, management and analysis was similar in 
all the phases of the study. The themes and sub-themes were used to analyze the endline 
data by using the same thematic analysis framework.  
 
3.6 Emerging themes and sub-themes 
Themes and sub-themes that emerged from the data recording, management and analysis 
are presented in Table 3.7.  
 
 
 
 
 
 
 
 
 
 97 
University of Ghana http://ugspace.ug.edu.gh
Table 3.7: Themes and their sub-themes  
Objectives Themes Sub-themes 
1. To describe the The food system and gender- Gender-driven access to productive resources 
food system and based issues Gender differences in types and use of foods 
women’s interaction produced 
with it in Savelugu- Women’s role in processing, preservation and 
Nanton District storage of food 
Gender roles in food distribution and household 
allocation  
Gender dynamics in household food provision, 
preparation and sharing 
Food shortage, coping strategies and effects on 
women’s diets 
 
2.To assess Community dietary Knowledge on nutritive qualities of foods and 
communities’  dietary knowledge and attitudes sources of information 
knowledge and Perceptions of health, healthy foods and 
aattitudes at Yilkpene community food preferences 
and Kpachilo in  
Savelgu-Nanton Perceptions of nutritional adequacy of women’s 
District diets and effect on women’s reproductive health 
 
Communitiy members’ beliefs about food  
 
3.To describe Women’s dietary practices  Women’s eating pattern and dietary diversity 
women’s dietary   
practices    
  Drivers of women’s dietary choices 
4. To assess the Drivers of women’s dietary 
drivers of women’s choices 
food choices 
   
5. To assess Level of community readiness Community knowledge of women’s sub-optimal 
Yilkpene’s readiness for socio-cultural intervention diets 
to modify socio- on women’s sub-optimal diets Coommunity knowledge and efforts addressing 
cultural barriers to I Yilkpene challenges to women’s diets 
women’s optimal Community climate regarding women’s sub-
diets optimal diets 
Community leadership in support of efforts 
addressing women’s diets 
Community resources to address challenges to 
women’s diets 
 Description of implementation Exposure to and coverage of intervention activities 
of intervention 
  Modifications and adaptation of initial plan 
6. To test the post- Change in dietary knowledge -Improved dietary knowledge 
intervention change in and attitude   
community dietary -Changes in some dietary misconceptions 
knowlwdge and  
atttitude (perceptions -Modifications of plant-based food restrictions 
and beliefs) at  
Yilkpene -Minor changes in women’s dietary practices 
 
 Community perceptions of the Intervention messages recalled by participants 
intervention Messages practised by participants 
Messages not practised by participants 
Benefits of intervention to community 
Challenges related to the intervention activities 
 98 
University of Ghana http://ugspace.ug.edu.gh
Analysis of dietary information: Data on all foods consumed by each of the women who 
participated in the 24-hour dietary recall was used to assess the quality of their diets. 
Women who consumed diets from 5 or more food groups were classified as meeting the 
minimum dietary diversity for women (MDD-W) but those who consumed foods from 
less than 5 groups were classified as not meeting the minimum dietary requirements 
(Martin-Prevel et al., 2017; FAO & FHI360, 2016). 
 
Analysis of intervention results: Analysis of the intervention results was carried out by 
comparing outcomes in the intervention community with the outcomes in the comparison 
community. The comparison was done at a single point following the intervention (the 
single  difference impact estimate) (White & Sabarwal, 2014).  
 
3.7 Monitoring the process  
During the 17 weeks of implementation, different strategies were used to monitor the 
study. First, there was a record of participation using a register of participants’ names 
and households. Second, home-visit records helped to assess the individualss’ 
participation in the group meetings. If the hosts or hostesses had participated in group 
meetings, they were asked to recount some of the messages they had heard and whether 
they had ever tried to practise any of them. The visiting team corrected misconceptions 
or misunderstandings that they detected. The team also reinforced relevant specific 
messages. Furthermore, husbands, heads of households and elderly women were 
encouraged to support women to practise the recommended behaviours since the former 
(men and elderly women) can influence the choices the latter make regarding their food 
and nutritional needs (Ahrari et al., 2006). Additionally, participants were asked to give 
their impressions of the intervention. While the BCC activities were going on in the 
 99 
University of Ghana http://ugspace.ug.edu.gh
intervention community, a monthly report of the activities in the comparison community 
was recorded. 
 
Implementation activities ended after 17 weeks of intensive BCC in the community 
covering the period 25th December, 2016 - 23rd April, 2017. A period of 3 months (May, 
June and July 2017) was allowed for the community to assimilate the messages and for 
the research team to assess if there would be any observable application of the messages 
in the women’s dietary behaviour. 
 
3.8 Quality assurance in the study (trustworthiness) 
A number of steps were taken to ensure the quality and trustworthiness of the study 
(Lincoln & Guba, 1985). Trustworthiness implies the accuracy and consistency of the 
study and the extent to which its results can be trusted based on four criteria proposed by 
Guba (1981). The first criterion is  credibility –that is, accurate recording of issues 
investigated and producing results that are convincing and believable. The second 
criterion is transferability- which implies generalizing results to similar contexts or 
findings being applicable to other contexts with similar characteristics.  The third 
criterion is dependability -implying replicating the study in similar contexts and arriving 
at similar results. The last criterion is conformability which suggests that the researcher 
is as neutral as possible during the research process with very minimal investigator 
biases (Shenton, 2004). These criteria were achieved at different stages of the research 
process by using various strategies.  
 
 100 
University of Ghana http://ugspace.ug.edu.gh
The study adopted appropriete and well-recognised qualitative research methods such as 
focus group discussions, key informant interviews and observations which 
complemented one another in gathering rich data. In using these methods, the support of 
research assistants was needed. At the pre-data collection stage, four research assistants 
were recruited and trained to collect data during the baseline phase and implementation 
of the intervention. The research assistants were all holders of master’s degrees in 
nutrition. They were trained for a period of five days on the objectives of the study, 
techniques of qualitative interviewing, note-taking, data collection instruments and 
verbatim transcription of  recorded interviews into English. The data collection 
instruments were also pre-tested in a different but similar community with the assistants 
to determine their accuracy and consistency in gathering the required information. 
Applying “probes” and “prompts” as and when necessary allowed participants to give 
detailed descriptions of issues as far as they could.  
 
The investigator’s twenty-one-month engagement with the communities (from March 
2016 to December 2017) enabled participants to develop high level of trust and 
confidence in her so they gave their candid views on issues. At the end of the baseline 
study, the information gathered was shared with the whole community at a durbar to 
confirm its credibility. To enhance the credibility of the quasi-experimental design, the 
two communities were selected on the basis of their non-proximity to minimize 
information contamination.  
 
The nutrition education counselling cards, which were based on the three food groups 
developed by Women in Food and Agriculture (WIFA) of the Ministry of Food and 
Agriculture (MOFA), were adapted by the implementation team to develop the nutrition 
 101 
University of Ghana http://ugspace.ug.edu.gh
messages. Also, both communities were monitored during the implementation of the 
intervention to take note of other on-going activities whose impact might influence its 
outcome – that is, to avert intervention contamination. In the comparison community, for 
instance, a number of nutrition–related interventions by World Vision were on-going 
while Ghana Health Service had similar activities running in both communities. This 
information contamination was used to explain the outcome of the intervention.  
 
To ensure transferability and dependability of the study results, detailed descriptions of 
the two communities and the methods  and procedure for gathering and analysing the 
data have been provided earlier in this chapter. The detailed description ensures that the 
findings of this study are generalizable only to communities with similar characteristics 
and that similar results may be obtained if the study is replicated in similar communities 
using similar methods and procedures. Additionally, justifications for choosing the 
methods used in the study and an audit trail for the data analysis are presented to 
minimize investigator biases. Conformability was achieved, first, through the use of the 
same data collection tools at baseline and endline and, second, through the use of 
different data collectors at endline so as to minimize effects of investigator bias as much 
as possible.  
 
3.9 Ethical considerations 
Noguchi Memorial Institute for Medical Research (NMIMR), University of Ghana, 
Legon, granted ethical clearance for the study (Reference No. (CPN): 077/15-16). 
Permission to carry out the study was obtained from Savelgu-Nanton District Assembly 
and the chiefs of the two selected communities where data was collected. Their support 
for the study was solicited after being informed about the purpose and procedures.  
 102 
University of Ghana http://ugspace.ug.edu.gh
Ethical issues addressed during data collection included obtaining informed consent from 
participants and informing them of possible risks and discomfort their participation may 
cause, their right to leave the study and possible benefits to them and the whole 
community. To obtain  informed consent from participants, they were provided with 
general information about the purpose of the research and what they were expected to 
contribute to ensure its success was spelt out to them. The majority of participants gave 
their consent by thump-printing on the agreement section of the consent form. Consent 
for the adolescent girls who were below 18 years of age was obtained from their parents 
or guardians.  
 
The issue of possible risks and discomfort was addressed by assuring participants that 
the study would not involve any invasive procedures in collecting the data so that any 
unforeseen risks it might incur as a result of their participation would be minimal. The 
only anticipated discomfort was associated with the time spent to discuss the issues 
relevant to the study. In this regard, participants were encouraged to spare about one 
hour of their time to discuss the issues. The burden of the study, participants were 
assured, would be borne by the researcher. 
 
The issues of participants’ participation, confidentiality, safety and right to leave the 
study were addressed in three ways. First, at each stage of the research process, 
participants were encouraged to participate fully in all aspects of the study by giving 
their candid opinions on issues being discussed and asking questions when necessary to 
keep the discussion interesting and participatory. They were further encouraged to share 
their personal experiences as they related to issues being discussed. Second, participants 
were assured that their records regarding their participation in the study would be strictly 
 103 
University of Ghana http://ugspace.ug.edu.gh
protected. In this respect, participants’ true names were replaced with dummy ones in the 
transcripts and also they did not appear in the reports. Third, participants were told they 
could pull out of the study at any point or stage of the study if they so wished. 
 
Participants were told at the start of the interviews and group discussions that their 
involvement in the study would not attract any monetary compensation for information 
volunteered; however, each of the participants at baseline and endline was given a bar of 
soap valued at  ten Ghana cedis (GHc 10.00) as a token of appreciation of their time 
spent in participating in the study. Similarly, each of the eight community volunteers 
who were members of the implementation team was given a monthly incentive of one 
hundred Ghana cedis (c100.00) for a period of 4 months.  
 
The incentive was meant to encourage the volunteers to suspend all other activities on 
the dates set for the intervention activities so as to mobilize the community and to lead in 
the dissemination of the intervention messages. The study communities were assured that 
the study was going to benefit them immediately by improving their knowledge on diets 
and health, which would enhance the health statuses of members of their households in 
general and the women in particular as well as improving the well-being of the 
communities at large.  
 
Even though the comparison community did not benefit directly from the intervention 
activities, a half day community meeting was organized to share the baseline results with 
members of the community. The dessimination meeting helped to improve their 
awareness of the importance of consuming variety of foods in their meals and the 
importance of diets for WRA. It is hoped that a similar intervention would be organized 
for the comparison community if some funding is secured.  
 104 
University of Ghana http://ugspace.ug.edu.gh
CHAPTER FOUR 
4.0 RESULTS OF THE STUDY 
This chapter presents the results of the study. The chapter is organized into four main 
sections namely, the participants’ characteristics, pre-intervention (baseline) results, 
implementation of the intervention and post-intervention results. The pre-intervention 
results describe the situation in the two study communities and are organized into four 
sub-sections: the food system and women’s role in it; community dietary knowledge and 
attitudes, women’s dietary practices and drivers of their dietary choices and community 
readiness to address challenges to women’s diets. The process data from the 
implementation section describes how the community engaged with the intervention 
activities, coverage of intervention and modifications and adaptations with respect to the 
original intervention design. The post-intervention section describes changes in dietary 
knowledge, attitudes and women’s dietary practices. The section also describes 
community members’ assessment of the intervention. 
 105 
University of Ghana http://ugspace.ug.edu.gh
4.1 Study participants’ socio-demographic characteristics  
Table 4.1: Participants’ socio-demographic characteristics 
Yilkpene (119) Kpachilo (63) Total (182) 
Characteristics Sub-
Male Female Sub-total Male Female Male Female Total 
total 
n(%) n(%) n(%) n(%) n(%) n(%) n(%) N(%) 
n(%) 
1.  Sex 43(36.1) 76(63.9) 119(65.4) 16(25.4) 47(74.6) 63(34.6) 59(32.4) 123(67.6) 182.(100) 
2.  Marital status                   
Single 11(25.6) 0(0) 11(9.2) 31(8.8) 8(17) 11(17.5) 14(23.7) 8(6.5) 22(12.1) 
Married 32(74.4) 76(100) 108(90.8) 13(81.2) 38(80.9) 51(81) 45(76.3) 114(92.7) 159(87.4) 
Widow 0(0) 0(0) 0(0) 0(0) 1(2.1) 1(1.6) 0(0) 1(0.8) 1(0.5) 
                   
3.  Women’s 
                  
groups 
Pregnant   13(17.1)     10(21.3)     23(18.7)   
Lactating   22(28.9)     16(34.0)     38(30.9)   
NPNL   20(26.3)     10(21.3)     30(24.4)   
Adolescent girls   9(11.8)     8(17.0)     17(13.8)   
Elderly women 
  12(15.8)     3(6.4)     15(12.2)   
(50+) 
                    
4.  Education                   
None 13(30.2) 72 (94.7) 85(71.4) 1(6.3) 44(93.6) 45(71.4) 14(23.7) 116(94.3) 130(71.4) 
Primary 7(16.3) 1(1.3) 8(6.7) 8(50) 2(4.3) 10(15.9) 15(25.4) 3(2.4) 18(9.9) 
Secondary 8(18.6) 3 (3.9) 11(9.2) 2(12.5) 1(2.1) 3(4.8) 10(16.9) 4(3.3) 14(7.7) 
Arabic 15(34.9) 0 15(12.6) 5(31.3) 0 5(7.9) 20(33.9) 0 20(11.0) 
                    
5.  Occupation                   
None 0 9(11.8) 9(7.6) 0(0) 8(17.0) 8(12.7) 0(0) 17(13.8) 17(9.3) 
Farming 34(79.1) 64(84.2) 98(82.4) 16(100.0) 39(83.0) 55(87.3) 50(84.7) 103(83.7) 153(84.1) 
Formal-sector 
9(20.9) 3(3.9) 12(10.1) 0(0) 0 0(0) 9(15.3) 3(2.4) 12(6.6) 
employment 
                    
6.    Monthly 
                  
income (Ghc) 
Mean 121,07 41,29 81,18 135,83 48 91.90 112,33 42,47 80,23 
Range 40--320 16--128 16--320 100--245 28--100 28--245 40--320 16--128 16--320 
                    
7.    Sizes of 
                  
households 
16--15     45(37.8)     38(60.3)     83(45.6) 
16--40     71(59.7)     22(34.9)     93(51.1) 
Missing cases     3(2.5)     3(4.8)     6(3.3) 
 
 106 
University of Ghana http://ugspace.ug.edu.gh
The total sample of participants was 182. Females dominated the sample in both 
communities but the female –male gap was wider at Kpachilo (49.2% - that is, 74.6% - 
25.4%) than at Yilkpene (27.8% - that is, 63.9% – 36.1%). The proportion of married 
women was higher at Yilkpene (90.8%) than at Kpachilo (81%). The minority of women 
in the two communities were elderly (50+ - 12.2%). Similar proportions (71.4%) of 
participants in the two communities had no formal education. Higher proportions of 
participants at Yilkpene, however, had Arabic education (12.6%). Farming was the 
dominant occupation in either community (84.1%).  The majority of participants in either 
community earned less than GH500.00 monthly (92%). The sizes of households were 
generally large ranging from 16 to 40 persons per household. The proportion of 
households at Yilkpene with members ranging from 16 to 40 was 59.7% compared to 
34.9% at Kpachilo. The missing cases (3.3%) were participants whose socio-
demographic characteristics were not complete. 
 
4.2 Pre-intervention (baseline) results 
Baseline data was collected in the two study communities as part of the pre-intervention 
activities. This section presents results on the themes that emerged from the data. These 
themes include the food system and gender-based issues, community dietary knowledge 
and attitudes, women’s dietary practices and drivers of their dietary choices and 
Yilkpene’s level of readiness to address socio-cultural challenges to women’s diets. 
 
4.2.1 The local food system and gender-based issues 
The food system in the two study communities is built around an agrarian economy that 
involves food production, processing, preservation and storage, distribution and 
utilization/consumption. Men, women and children perform diverse roles within the 
 107 
University of Ghana http://ugspace.ug.edu.gh
different components of the system. While some roles are performed based on gender, 
both men and women are involved in others.  
 
4.2.1.1 Gender-driven access to productive resources 
Women are unable to produce large quantities of food because they have limited access 
to and ability to produce inputs/resources for production. The following narrative 
expresses how women feel about their inability to produce enough food owing to their 
limited ability to access food resources such as land: 
“But if women had large tracks of lands like the men, we wouldn’t be struggling 
to put food on the table… But how can we produce large quantities of food? We 
don’t have the resources to do so, so we keep to smaller land and manage with 
it” (Pre intervention Female, KII K09). 
 
Farming inputs including land, labour, seeds, farming implements and chemical 
fertilizers are important determinants of food production; however, access to these inputs 
is gender driven. Key informants indicated that household land-holdings, which are 
generally small, ranged between 10 and 15 acres in size and are usually shared only 
among male members of the household. Women, therefore, have less access to land in 
this highly-patrilineal society. Women are not considered part of their fathers’ 
households hence, allowing them inheritance to their fathers’ or family land would 
deprive the fathers’ household of land once the women got married into other 
households. Female FGD participants explained women’s lack of access to farmland in 
particular as emanating from their traditional inheritance system: 
“It’s because if a woman inherits the land from her father, it’ll not benefit his 
household any more because once the woman gets married, she’ll take the 
produce from the farm to her husband’s house to feed its members. (Pre-
intervention FGD Yilkpene, elderly women).  
 
 108 
University of Ghana http://ugspace.ug.edu.gh
A married woman could, however, have access to a portion of her husband’s farmland as 
stated by a community leader at Kpachilo as follows: 
“If your wife expresses interest in farming, it’s you yjr husband’s responsibility 
to look for land for her to farm and if she gets much yield it’s for you and your 
wife” (Pre-intervention FGD community leaders K27)  
 
Alternatively, a woman can borrow land from neighbouring communities if she has the 
financial resources to carry out farming activities as expressed in the following 
narratives: 
A part from going to other neighbouring communities, women who have the 
financial resources to undertake farming activities may also borrow land from 
other families in their communities if those families wouldn’t be farming on their 
land for a particular farming season” (Pre-intervention KII Y22) 
 
Women’s limited access to farmland contributes to their inability to grow a variety of 
food crops in appreciable quantities to meet their food needs and those of their 
households. Productivity among women is also limited by their inability to procure 
labour or receive compensation for labour. Traditionally, women are expected to 
contribute labour for their husbands’ farms but the reverse is not practised.  
“As for your wife, it’s a must for her to come to the farm and help. If she doesn’t 
come, who else can come? The only reason your wife can’t come to help you 
harvest is ill health. In that case, she wouldn’t be at fault and her portion will be 
served. But if she refuses to come for any other reason, well, it has never 
happened before. She must be on the farm” (Pre intervention, Men FGD, K 31) 
 
Additionally, men can hire labour that is paid for through the sale of their farm produce. 
Because women do not have productive capital, they are usually constrained, and, thus, 
unable to hire labour. Women assist in planting crops (except yam), cook for farm 
labourers, harvest the produce and carry it home but other farming activities such as 
preparation of land for planting and making of beds and mounds for planting of yam and 
 109 
University of Ghana http://ugspace.ug.edu.gh
cassava are the preserves of men because they are perceived to be beyond women’s 
physical strength as narrated by one key informant:  
“But, then, it’s usually yam and other root crops which are difficult for the 
women to cultivate because they require physical strength, which the women 
don’t have; and so, it’s the men who usually produce them. Even with that, if a 
woman can pay for others to work on the farm for her, then she can produce the 
crops that demand physical strength” (Pre-intervention KII Y22) 
 
When women are hired by other households for farm work such as to harvest crops, they 
are paid in kind compared to men who are paid in cash:  
“For families without matured sons, they pay for others to work on their farms 
but the payment is usually made to men. Women who help on the farm are given 
portions of the produce as their rewards” (Pre-intervention KII Female K25) 
  
Apart from preserving seeds to be used in subsequent planting seasons, buying and 
occasional assistance from non-governmental organizations are common in the 
communities. Farming equipment and agro-chemicals are usually accessed from the 
market. Women may be constrained from purchasing these inputs by their limited 
financial situation.  
 
4.2.1.2 Gender differences in types and use of foods produced  
Women’s productivity is limited by their traditional role of cultivating crops with low 
market values. The two communities share similar practices in that they mainly produce 
food for home use and for sale. Crop production is the main livelihood; livestock 
farming is a supplementary activity. Even though tradition does not spell out 
categorically what kinds of crops men and women should produce, it was observed and 
reported that gender plays a significant role in the types of crops produced and their 
socio-economic importance in the two communities. This claim is supported by the 
response from a male key informant when he was asked whether there were specific 
crops cultivated by men and women in the communities. The informant indicated that 
 110 
University of Ghana http://ugspace.ug.edu.gh
even though both genders were at liberty to produce any crop of their choice, the ability 
to do so depended on one’s access to productive resources such as land and finance: 
“There’re no specific crops produced by men or women in this community. As 
long as you’ve the land and you’re strong enough to produce a particular crop, 
then you just do so. If a man or woman has land, depending on what the land is 
fertile for, then he or she produces it. But then it is usually yam and other root 
crops, which are difficult for the women to cultivate because they require 
physical strength, which the women don’t have; and so mostly, it’s the men who 
produce them. Even with that if a woman can pay for others to work on the farm 
for her, then she can produce the crops that demand physical strength” (Pre-
intervention KII K22). 
 
Table 4.2 shows the types of foods produced in the two communities and those men and 
women dominate in producing. 
Table 4. 2: Types of foods produced 
Food Predominantly for home- Predominantly for sale 
production consumption 
1) Crop Men Women Men Women 
production 
a) Commonly- Maize, guinea- Maize and  Rice  Rice, 
produced crops corn  groundnut  groundnut, cow-and Soya 
pigeon-pea;  bean,  
okro, tomato, okro 
pepper, garden-egg 
water-melon 
b) Less Millet  Millet  
commonly- melon-seed (neri bambara bean,  
produced crops and agushie) yam 
yam; sweet- cashew nut 
potato; cassava, 
cotton 
c) Crops  Ayoyo, bra,  Shea-nut 
harvested from kuka,   
the wilds baobab fruit, 
dawadawa fruit, 
shea-fruit, 
mango, ebony 
fruit, 
“sinsaba”(black 
berry) 
2) Livestock   Cattle, sheep, goat,  
production fowl, guinea-fowl, 
rabbit and egg 
 
 111 
University of Ghana http://ugspace.ug.edu.gh
The foods produced in the study communities included cereals/grains, roots/tubers, 
legumes/seeds/nuts, green-leafy vegetables, other vegetables, fruits, meat and egg. 
Maize, guinea-corn and groundnut constituted the main staple crops. Men controlled 
cultivation of staple crops for home consumption but when they produced nutrient-rich 
legumes in large quantities, they were often sold for money. A female focus group 
discussion participant supported the claim that men often sold out nutrient-dense foods, 
leaving very little or none at all for home-use with the following narrative: 
“Even if our husbands harvest some of these crops (beans, Bambara, pigeon pea etc.), 
they sell them at the market. If you’ve the money, you can buy otherwise you miss those 
foods that add variety to your food. In a few cases, the men may leave some for us at 
home but that’s small and won’t be enough to last long” ( Pre-intervention FGD women 
K 34). 
 
The assertion by the focus group discussion participant at Kpachilo was confirmed by 
one of her counterparts at Yilkpene as follows: 
“The men don’t just harvest their groundnut and give you your portion as they do 
with the maize. No way. You have to work to earn it so the harder you work, the 
more groundnuts you accumulate to last long. As for the men, their portions go to 
the market”(Pre-intervention FGD women, Y34) 
 
Additionally, men dominated in the cultivation of crops with high market values such as 
tomato, pepper, rice and cashew-nut. The main market crops cultivated by women were 
rice, soya-bean and okro. Cotton and melon seeds used to be common ingredients in 
soups but they were currently less cultivated owing to unfavourable weather conditions. 
Production of yam and millet, which used to be important cash crops in the communities, 
had also dwindled for the same reasons as with cotton and melon seeds. A male focus 
group discussion participant indicated that such nutritious foods were scarcely available 
in the communities: 
 112 
University of Ghana http://ugspace.ug.edu.gh
“So these days, cotton seed (kanton) isn’t in great supply. Very few people 
cultivate it...Yes, and neli (melon seeds), yam and millet too aren’t being 
cultivated in large quantities because the rains are poor and cultivating them has 
become a problem”(Pre-intervention FGD men, Y35). 
 
Women mainly harvested nutrient-rich vegetables and fruits from the wild; however, 
these were seasonal crops and, were thus, not available throughout the year unlike the 
staple crops. The communities were not interested in cultivating foods collected from the 
wild. One key informant explained during the community readiness assessment that they 
were not cultivating the foods obtained from the wild because it was not important to put 
their efforts into producing crops that grew naturally: 
“One doesn’t look for what one already has but rather puts efforts into finding 
what one doesn’t have”(CRM 14).   
 
Although men and women were involved in livestock production, men dominated the 
control and utilization of livestock resources. A male key informant defended this 
practice during the exploratory phase of the study as follows: 
“In this community… the men have total control over everything concerning the 
livestock... She can’t do that (sell her livestock) without the man’s consent... The 
way women are, if you allow them too much authority today, tomorrow she’ll ride 
on your back…” (Pre-intervention KII Y24). 
 
Key informants indicated that even though livestock was reared primarily to obtain 
savings, animals might be used as food during festive occasions and to welcome special 
guests. This assertion was made bluntly by a male focus group participant at Kpachilo:  
“Those animals you see around are investment for us like the bank accounts you 
people have. Madam, you don’t just withdraw money until the need arises” (Pre-
intervention FGD Men, K28). 
 
 113 
University of Ghana http://ugspace.ug.edu.gh
The above assertion was confirmed by a male key informant at Yilkpene .He identified 
the value of livestock to the community as their form of investment and explained that 
they consume them only during festive occasions: 
“To be frank, we hardly kill any livestock for personal consumption. We usually 
kill livestock during occasions such as a naming ceremony, marriage or funeral. 
Other times that we can kill any livestock for preparation of food are when we 
get important visitors. Besides these occasions, we rear livestock so that when 
we’re in need of money, we can sell them or when we run out of foodstuffs, we 
can sell these animals to buy needed foodstuffs” (Pre-intervention KII Y23).  
 
4.2.1.3 Women’s role in processing, preservation and storage of food  
Although women were involved in all aspects of post-harvest activities including food 
preservation, they had little economic value for them. Women’s role in husking maize, 
rice, millet, guinea-corn and also in shelling legumes (beans, groundnut) was narrated by 
a female key informant at Yilkpene:  
“The harvesting and the dehusking are mostly done by men and women. 
Whenever the crops are ready for harvest, we go and harvest them and gather 
them under a shade on the farm so that we can sit there to remove the husk. Then 
we carry it to the house where we look for a shelling machine to do the shelling -
that is, if it’s in large quantities. Otherwise, we do it manually. We then dry the 
produce in the sun” (Pre-intervention KII Y22). 
 
Women also the primary group for processing of shea-nut, dawadawa and groundnut, 
boiling and milling of rice as well as fermentation of grains. Both men and women 
participated in solar drying of grains, legumes and some root crops (cassava in 
particular). The drying of vegetables such as bra, kuka, pepper and okro was dominated 
by women as explained by a key informant at Yilkpene:  
“As for vegetables, when it’s their season like the rainy season, the women 
harvest them only on days they need to use them to cook. Vegetables decay fast 
and so they can’t be stored for long. We dry some of them like okro, bra and 
kuka. This activity is usually done well by women” (Pre-intervention KII Y22) 
 
 114 
University of Ghana http://ugspace.ug.edu.gh
Traditionally, men had the responsibility of storing household grains and their other farm 
produce in food barns and silos (if they own them).  On the other hand, women mostly 
stored their produce in sacks but they did not offer adequate protection against insect 
pests which attacked the stored food. Women were, thus, more likely to suffer post-
harvest losses. A female key informant explained the predicaments of women during 
post-harvest storage of their produce:  
“After husking, we dry the maize and then shell it. The women bag their produce 
and store it in rooms but most men store their produce in silos. What you should 
know is that, once the maize is being put into silos, there’s no need to add 
chemicals to prevent weevils from spoiling the grains because the silos are air-
tight. However, with the bags, you have to spray the produce with chemicals to 
prevent weevils and other insects from destroying them but the women are unable 
to afford such chemicals”(Pre-intervention KII Y22). 
 
4.2.1.4 Gender-based roles in food distribution and intra-household allocation  
Women exerted a lot of energy in the distribution and sale of farm produce. Foods 
produced in the two communities for sale was usually sold either at the farm-gates to 
traders or carried to Nanton, Savelugu, Kunbungu and Tamale markets for sale. Both 
men and women were responsible for determining the prices of their produce and selling 
it. This role of both men and women was explained by a male key informant at Yilkpene:  
“But the produce belongs to me. Why should someone sell it for me? I sell it 
myself. Anyone who plants and harvests sell the produce himself or herself. 
Sometimes, people (traders) from elsewhere come to the community to buy it but 
at other times, I send it to Savelugu Market by means of a motor king to sell” 
(Pre-intervention KII Y22). 
 
Women further assisted in transporting the produce home or to the market by carrying 
the loads on their heads. Men, however, would be involved in transporting farm produce 
to the market if they had access to mechanized/motorized transportation such as bicycles 
and “motor-king”. 
 115 
University of Ghana http://ugspace.ug.edu.gh
 
Traditional roles of resource management favoured men. Men traditionally controlled the 
allocation of stored grains that women used for preparation of food for the household. 
Male focus group discussion participants at Kpachilo confirmed this practice: 
“They (women) have to wait for their husbands or men to give them portions of 
grains for the management of the home”(Pre-intervention FGD men K34). 
 
Allocation of grains for the household’s use was usually done once each week. In 
polygamous households, the allocation was done on the basis of which wife had the turn 
to prepare meals. Women were not traditionally allowed to enter food barns to take 
grains for preparation of meals. A male focus group participant at Yilkpene defended the 
“no entry for women” on the basis of mismanagement of grains by the women: 
“The freedom for women to go into the grain barns to fetch grains by themselves 
is impossible. Women aren’t allowed to go into the grain bans or silos to take 
grains for the home. They’ve to wait for their husbands or men to give them 
portions for the management of the home. If allowed, they’ll mismanage the 
grains and, in no time, nothing will be left to survive us for the rest of the year” 
(Pre-intervention FGD men, Y35). 
 
This tradition is enforced as a taboo. The reason for this taboo was the perception by 
men that women are not good stewards of stored grains, was confirmed by a male focus 
group discussion participant at Kpachilo:  
 “They (women) will mismanage the grains and in no time nothing will be left to 
survive us for the rest of the year”(Pre-intervention FGD, men, K30). 
 
The men’s stance was more or less supported by a female focus group discussion 
participant at Yilkpene: 
“It’s believed that if a woman is seen going into the barns to fetch grains, her 
children will turn up to be thieves”(Pre-intervention FGD Females Y30). 
 
 116 
University of Ghana http://ugspace.ug.edu.gh
While some women agree with the perception that women are poor managers of food 
resources and hence the need for men to control household grain, not all women are 
happy with this tradition on grounds that the freedom to have direct access to grains 
makes it possible for them to feed the households better. Another female focus group 
discussion participant at Yilkpene said the following in support of men controlling 
household food resources: 
“Because some women will mismanage the supply, even if the men say we can go 
in for the grains, we won’t go. They should just give us our portions to go”(Pre-
intervention FGD Females Y31). 
 
A female participant at Kpachilo supported the “no entry for women” tradition on 
grounds of making men responsible for replenishing grain barns: 
It’s to the benefit of us all.  This prevents us women from misusing the grains until it was 
time to farm again for new foodstuffs. But if women are allowed to manage the grain 
barns and the foodstuffs are finished and the man isn’t able to provide any, then the 
woman will be the very person to provide the grains. That’s what we don’t want. That’s 
why we’re according the men that respect to fetch the grains for us to cook” (Pre-
intervention FGD women k27). 
 
A female focus group discussion participant at Yilkpene, however, held a contrary view 
about gran allocation by men: 
“We wish we could be allowed to get all our food needs to be able to feed our 
households well but we are limited”(Pre-intervention FGD Women, Y31). 
 
In other words, grain allocations by men, is perceived by some women as inadequate for 
the needs of their households. Some men, however, had an issue with that view. To 
them, the allocation of grains is usually planned with the women, who decide on how 
much grain will suffice for a specific period. A male focus group discussion participant 
at Yilkpenen explained how the planning is done:  
“You know a man doesn’t decide by himself how much grain will suffice the 
house. He discusses it with the woman and she tells him how much is needed for 
 117 
University of Ghana http://ugspace.ug.edu.gh
a certain period. So once the man knows this quantity, he plans his harvest and 
knows how long it’ll last. If it won’t be enough for the year, he makes plans about 
how to cope until the next farming season. So if a man gives out grains to last till 
a certain period but the woman comes in earlier than that time to say it’s 
finished, she must be answerable for that. She can’t even come and ask for 
more”(Pre-intervention FGD Males Y35). 
 
Some women, however, indicated that their requests for additional supplies of grains are 
usually honoured by their husbands or heads of households when they run out of grain 
before the stipulated time.  A female key informant at Yilkpene explained the tradition of 
making supplementary allocations: 
“If the grains given get finished, you go and tell him again and he’ll dish out 
additional grains. But, how you manage to grind the grains, get ingredients, 
firewood, etc is no longer his problem”(Pre-intervention KII Y26). 
 
Giving of farm produce, usually grains, as gifts after harvest is a common practice that 
minimizes the effects of food shortages on vulnerable households. In these communities, 
widows, the elderly, chief imams and ultra-poor households receive provisions of food 
from households whose harvests are good in a season. According to a key informant at 
Kpachilo: 
“If you bring the food home after harvesting, you have to give some to the chief of the 
community even if it’s just a bowl of the grains that you can give to him. If you have a 
bumper harvest, say 10 bag of rice, you can decide to give one bag to your Imam. You go 
and leave it in the mosque and the Imam will decide what to do with it. So we do give 
some of our food out as gifts. Sometimes too, you could also give to the aged in the 
community” (Pre-intervention KII, K22). 
 
Food resources are also distributed freely during festivals, funerals and marriage 
ceremonies. Even though men take the decision to give out the foods, women usually 
carry them to the beneficiaries in their homes. 
 
 118 
University of Ghana http://ugspace.ug.edu.gh
4.2.1.5 Gender dynamics in household food provision, preparation and sharing 
In spite of their inability to access and control household food resources, women are 
expected by tradition to provide food to complement men’s provisions for the 
households. Preparation of food for the household is the sole responsibility of women. 
Men are traditionally responsible for providing staple (grains) components of household 
meals. Women are expected to provide the ingredients needed for preparing the soup. 
These ingredients include anchovies (small herrings), salt, onion and bouillon cube.  One 
female key informant explained the responsibilities of men and women regarding 
provision of food for households: 
“It’s not also his responsibility to give you money to buy ingredients to prepare 
soup. His responsibility is just to provide the grains for you. So we use our own 
money to prepare the soup. No, the piece of land they apportion to us to farm is 
what they expect of us to get the funds from for the ingredients, when we plant 
groundnut, we also plant okro so that when they both mature we can sell and 
consume some at home” (Pre-intervention KII Y26). 
  
A hierarchical system of power among women centralizes decision-making. At the 
household level, older wives are responsible for the preparation of major meals such as 
lunch and dinner while younger wives are responsible for preparing breakfast. A key 
informant gave an account of decision-making among women in a polygamous union 
regarding food preparation for the household: 
“It’s the first wife and her co-wives who are responsible for cooking. If they are 3 
or 4 wives, the first two or three will be responsible for preparing the food and 
they will do so alternatively but the youngest wife (that is, the last wife) will be 
responsible for preparing koko in the morning.  So when you’re giving the senior 
wives their grains, you’ll give the youngest wife too her portion” (Pre-intervention 
KII K22). 
 
The decision to consume a particular food is a joint decision arrived at between men and 
women. The female participants in the focus group discussion sessions indicated that the 
decision to eat a particular food in a household depends on the kinds of foodstuffs 
 119 
University of Ghana http://ugspace.ug.edu.gh
provided by their men but women determine the soup and other sauces that accompany 
the main meal. An in-depth interviewee at Kpachilo narrated the tradition in her 
household: 
‘My husband provides us with maize but I decide about the kind of food to cook 
based on the ingredients he provides. After harvest, he would store the maize and 
the guinea-corn for our consumption and only sell them if new yields are in and 
we haven’t run out of stock. However, when it comes to the ingredients for soup, 
it’s left to us women to provide them” (Pre-intervention IDI, K3b). 
 
Traditional practices about preparation and sharing of food limit women’s dietary 
quality. In addition to preparing meals, women serve it out to members of the household. 
There is a specific tradition with food service. Women serve their portions first, followed 
by those for the heads of households, elderly women, young men, young women and, 
last, children. Female focus group discussion participants at Yilkpene further explained 
that although women are served first after cooking, they do not eat first:  
“The fact that we serve our bowls before other household members does not 
imply we get the lion’s shares or we’re the first to eat. No! We only put a little in 
our bowls simply because we cooked the food and want to be sure the food is 
good enough to be consumed by other members of the household. Then after 
every one else is served, whatever is left is added to our bowl. If nothing is left, 
then we resort to other people’s leftovers. Sometimes, we even have to add ours 
to the children’s if they don’t have enough to eat. As for we women, if we don’t 
get anything to eat, it is ok. We can always manage to survive. But all other 
members of the households must get something to eat. Even sometimes if we get 
unexpected visitors, our bowls of food are sacrificed for them” (Pre-intervention 
FGD Women, Y34). 
 
Usually, women are the last to eat after everybody else is satisfied and would not need 
additional food as indicated by the female FGD participants above. Furthermore, if 
visitors arrive without prior notice at mealtime, the woman’s food is often offered to 
them - thus, leaving her with little or no food at all. One female participant at Yilkpene, 
however, explained that a woman never goes on an empty stomach as she always 
 120 
University of Ghana http://ugspace.ug.edu.gh
manages to get some leftover from other members of the households after they have all 
eaten to their satisfaction:  
“But this rarely happens. At least, we always manage to get leftovers from other 
members of our households to eat. We never starve in the house. We might not 
get the best portions to eat but we manage to get by with the little leftovers. More 
often then not, we get enough into our bowls” (Pre-intervention FGD Women, 
Y34). 
 
Sharing of meat is done by the male head of the household or elderly men at both the 
household and community levels. Similarly, when there is a social occasion and meat 
(whether raw or cooked) is to be shared, elderly men usually perform this task. A female 
participant narrated the process of sharing meat among members of a household:  
“You see the thigh and breasts of the fowl, they go to the men. Then the neck and 
wings go to the young men. The back and waist go to the women and the legs and 
head go to the children. The men can explain why that’s done. The bony part is 
given to us women because we need stronger bones to carry the next generation 
of children. So by eating the bones, we make our bones stronger to carry healthy 
pregnancy and be delivered safely” (Pre intervention FGD women, Y34). 
 
Participants reported patterns of social interactions during mealtime determined by 
gender and age. Wives usually eat together. Other groups also eat together including 
older women, young women or girls, young men and children. Male heads of households 
eat alone. If there is only one elderly woman, she also tends to eat alone.  
 
4.2.1.6 Food shortage and coping strategies  
Women’s access to optimal diets is affected by seasonal food shortages. Participants 
reported that staple foods (particularly cereals and legumes) are usually available 
throughout the year. On the other hand, seasonal foods like vegetables (including green-
leafy vegetables) and fruits are only available for brief periods in the year unless they are 
processed into dry products. The availability of food depends on adequacy of rains for 
 121 
University of Ghana http://ugspace.ug.edu.gh
production and storage from the previous farming season as explained by some key 
informants: 
“As I told you, except for a year that rains aren’t enough for us, food is always 
available in the community but, then, it’s most abundant during the harvest 
period” (Pre-intervention KII, K26).  
 
 
Another key informant at Kpachilo had earlier stated the importance of preservation as a 
copying strategy: 
“The fact is that for all the foods, we’ve specific times for harvesting them and 
we harvest most of them at the same time but how we preserve them is what 
makes them last long. We harvest them and keep and eat them gradually until the 
next farming season but even if stocks get depleted, we can still get some from the 
market to buy so we have them all year round” (Pre-intervention KII K22). 
 
Food is most abundant during the harvest period. Key informants reported that during the 
periods of September to February, all crops, with the exception of green-leafy 
vegetables, are abundant at the household level; however, the period between the months 
of April and August is a lean period characterized by food shortages. This period is when 
food crops have been planted and are awaiting harvest. During this period, even though 
fresh leafy vegetables are available, the main staples (maize, guinea-corn and groundnut) 
are typically in short supply. A key informant explained the period of food shortage in 
the following narrative: 
“Yes, we call it the “fufali saha”  (“hunger time). This period is when you use all the 
food you have to farm and then midway, the food in the house is finished and the crops 
too aren’t even ready for harvesting. So that time, things are really difficult for most 
households” (Pre-intervention KII Y26). 
 
Both Yilkpene and Kpachilo employ similar coping strategies including decreased 
consumption, increased buying of staples and borrowing grains from neighbours. During 
periods of shortage, men reduce quantities of grains allocated for cooking for the 
 122 
University of Ghana http://ugspace.ug.edu.gh
household. As a result, households reduce daily frequencies of meals or sizes of portions 
of meals served to members of the households, especially women. The following 
quotation explains the coping strategy adopted at the household level to manage food 
shortage: 
“For instance, if you’ve three wives and during the time of harvest, when food is 
in abundance, you can give each of them four bowls of maize to cook for a week, 
that will amount to twelve bowls of maize in a week. When there’s scarcity of 
food, you can reduce the weekly allocation from four to three bowls of maize 
each so that the weekly allocation of twelve bowls of maize will be reduced to 
nine. The women will, in turn, reduce the sizes of portions of the food after it’s 
prepared to accommodate the reduction in weekly allocations” (Pre-intervention 
KII Y23).  
 
The subsequent key informant at Yilkpene confirmed women’s sacrifice of their own 
portions of food: 
“When it happens like this, the woman begins to ration the food. For instance, if 
she used to serve herself a full bowl of TZ, during the rationing period, she can 
reduce it to a half-bowl” (Pre-intervention KII Y24). 
 
During the lean season, when some food ingredients are scarce, women cope by 
enhancing palatability of meals with bouillon seasoning. Although men dislike use of 
bouillon seasoning in meals, they are unable to restrict women from using it. Also, 
during periods of low availability of food, men may sell livestock, groundnut, pigeon- 
and cow-peas, and soya bean and use the income to purchase grains. In some cases, 
household property and income from casual labour may provide resources for purchasing 
grains. Women complement men’s income by selling shea-nut, firewood and personal 
effects (clothing). Women may also migrate to southern Ghana to work as potters during 
the off-farming season to earn money to supplement household incomes. Men may also 
borrow grains as a coping strategy. Additionally, resources from remittances and support 
from non-governmental organizations (NGOs) provide support during periods of 
 123 
University of Ghana http://ugspace.ug.edu.gh
insufficient food resources. A key informant gave an account of some of the copying 
strategies households and the community adopt during food shortage: 
“We sell the animals we raise to purchase the food … in our family during 
scarcity we sell some properties to buy food for the house and decide who among 
the young men would go to the south to work as a farm-labourer. Others in the 
community sometimes borrow from their neighbours. For some of us, we do petty 
trading to support the households during these times” (Pre-intervention KII, male, 
Y21). 
 
4.2.2 Community dietary knowledge, attitudes and sources of dietary information 
4.2.2.1 Knowledge of nutritive qualities of foods and sources of dietary information  
Knowledge about the nutritive benefits of foods was sub-optimal among both male and 
female participants in both communities. Although participants believed they produce 
health-giving foods, there was no consensus on the right classification of food on the 
bases of the three food groups (body-building, energy-giving and protective foods) 
among focus group discussion participants. The following narrative is an example of 
how some participants in the study communities classified the various food crops 
produced in the communities.  
  “Foods like maize, millet, rice, yam cassava, fufu give us strength, a lot of 
blood and good health. They make you drink more water when you eat 
such foods and that makes us healthy… The fruits also add blood” (Pre 
intervention FGD women K29). 
 
Some key informants, however, identified legumes, meat/fish, egg and vegetables as rich 
sources of blood-forming nutrients (iron) and smooth skin. Cereals, roots and tubers 
were associated with filling the stomach and giving energy. Vegetables (tomato and 
fruits) were associated with preventing constipation and diseases when eaten in 
moderation. The women in the in-depth interviews reported awareness of beneficial 
 124 
University of Ghana http://ugspace.ug.edu.gh
health components in the foods as well as the medicines (iron-folate supplements) they 
receive during ante-natal and post-natal care visits: 
“What I know is the medicine that they give pregnant women during ante-natal 
services when they go monthly for weighing. The medicines give them blood and 
then it makes them healthy. Some people also say that the medicines make the 
babies grow very big” (Pre-intervention IDI Woman, K35). 
 
The main sources of information on women’s health and diets are health facilities, 
households and community meetings. In addition, dietary and health information was 
obtained from elderly persons, family members, friends and teachers. Female 
participants indicated that during ante-natal and post-natal visits to the hospital, the 
health-care attendants educate them about their health including what they need to eat to 
enhance their health statuses. Female participants indicated they are advised by health 
workers to eat a lot of plant-based foods to enhance their health: 
“Ok, when we go to the hospital we’re always told to eat a lot of vegetables like 
ayoyo, alefu, and bra. We’re also told to eat a lot of  amani and dwadawa and 
prepare the food under hygienic conditions so that when the children eat they’ll 
be healthy as well as we the women. We also get to know from people like you 
during community meetings. We learn a lot from our discussions with you” (Pre-
intervention FGD women, Y30).  
 
Female adolescent participants confirmed the health facilities as one of the sources of 
dietary information in the communities: 
“Our mothers talk about them when they come back from the hospitals. They talk 
about how the doctor wants them to eat some kinds of foods in order to keep both 
themselves and their babies healthy” (Pre-intervention FGD adolescent girls 
K31). 
 
4.2.2.3 Perceptions of health, healthy diets and community food preferences 
Perceptions of health and healthy diets  
 125 
University of Ghana http://ugspace.ug.edu.gh
Participants often discussed diet in relation to their health. Many participants 
demonstrated misconceptions about what constitute good health. Some perceived health 
as the absence of disease or signs of ill-health, having good appetite for food and water 
and the ability to sweat. In respect of the perception of a healthy diet as one that induces 
sweating, a male participant’s comment was as follows: 
“If after eating a meal, one experiences profuse sweating, then one will know one 
has gained health from consuming that particular food. Therefore, one can say 
the food is healthy. And as you know, sweating after eating a meal is a sign of 
good health. So foods that make you sweat after consuming them are healthy 
foods” (Pre intervention FGD Men, Y35). 
 
Second, health was linked with physical energy to carry out daily activities. According to 
a female focus group participant at Kpachilo:  
“When a person is healthy you’ll realize he or she has the strength and gets 
active and performs her daily chores well. But with sickness, you’ll realize the 
person isn’t always active”(Pre intervention FGD women, K30). 
 
Third, the female participants linked health with peace. Women get worried when they 
are unable to provide food needs of their children and other members of the household as 
indicated by some of the women in the FGD sessions: 
“If you’ve your peace of mind and nothing worries you, then you can also say 
you’re healthy. Once you aren’t worried, you can eat well and carry out your 
daily chores satisfactorily; but if you’re worried about even how to feed your 
family or you’ve financial worries, then you’ve no peace and can’t function well. 
In that case, you can’t be said to be healthy”(Pre-intervention FGD, Women Y34). 
 
Finally, health was associated with physical appearance. A plump-looking body was 
considered an indication of good health. On the contrary, being skinny was linked with 
poor health: 
“If a person is healthy you’ll see that the person is fat but when a person is very 
lean, it means the person isn’t healthy” (Pre intervention FGD women, K29). 
 126 
University of Ghana http://ugspace.ug.edu.gh
Participants identified health-giving foods as having four main characteristics: meal 
composition of food; perceived benefits of food; perception of hygienic nature of food 
and its environment and whether the food is recommended by a health professional. 
Ingredients such as “dawadawa”, fish, meat, egg and condiments were considered 
important ingredients in a meal. Participants in the 24-hour recall in-depth interviews 
indicated the importance of dawadawa and animal source food in keeping healthy: 
“If food is prepared with dawadawa, fish, eggs and meat, then it’s healthy”(Pre-
intervention IDI Y09).  
 
Foods perceived to give blood, promote weight-gain, promote healthy, smooth skin and 
support copious production of breast-milk are linked with good health. A participant 
substantiated the claim with reference to the content of a particular food as a 
characteristic of its healthiness: 
All these raw materials that I’ve mentioned (dawadawa, fish. Eggs and meat) 
give blood so if they’re in a particular food and you eat it, you’ll get a lot of 
blood and so be healthy”(Pre-intervention IDI Y04). 
 
Foods, which, when eaten, induce drinking of a lot of water, profuse sweating and 
production of breast-milk are also considered healthy. Such foods are thought to prevent 
sicknesses as explained by some focus group participants:  
After eating such a meal, you drink a lot of water. That’s what I call a healthy 
meal” (Pre intervention FGD men, K31).  
 
The ability of hot foods to induce profuse sweating was mentioned by participants as the 
reason for considering them as healthy: 
“Well, the hotness of the food makes it healthy and that’s why lactating women 
are advised to eat hot foods to produce more breast milk. Foods that make you 
sweat after eating them are healthy”(Pre-intervention FGD Women Y34). 
 
 127 
University of Ghana http://ugspace.ug.edu.gh
Both men and women considered food hygiene as an important indicator of healthy food. 
They indicated that food should be prepared in a clean environment by using clean 
utensils and materials as well as observing good personal hygiene. A female participant 
indicated that keeping the cooking environment in a sanitary condition enhances the 
healthiness of food: 
“What makes you know your food is healthy is when you ensure the place you 
prepare your food is clean by sweeping the place and also wash all your dishes. 
After cooking and serving, you cover the food right away to prevent flies from 
settling on it” (Pre-intervention FGD women K29).  
 
Fourth, some participants perceived foods recommended by a health professional as 
healthy:  
“If you want to know whether a particular food is healthy, then it must have been 
recommended at the health facility for someone to eat it because that food will 
enhance his or her health…..You may also fall ill and some foods may be 
recommended to you by the doctors to eat in specific quantities and frequencies” 
(Pre-intervention FGD male, Y35). 
 
Misconceptions about the healthiness of some food items were also commonly 
identified. Some female participants reported that meals containing groundnut products 
are not healthy:  
“TZ with dry okro soup or kuuka is a healthy food but the dry okro soup should 
be cooked without groundnut. When you add groundnut paste to dry okro, it 
makes it unhealthy because it causes diarrhoea”(Pre-intervention IDI, Y10). 
 
A male participant endorsed the perception of groundnut as an unhealthy food: 
“Groundnut was never known in our culture as healthy for women who are 
pregnant or lactating…eating it didn’t facilitate proper development of the child 
… that’s why our children aren’t growing well these days”(Pre intervention 
FGD men, Y35). 
 
 
 
 128 
University of Ghana http://ugspace.ug.edu.gh
Community food preferences  
Although foods produced in the communities were perceived to be healthy, community 
members at both Yilkpene and Kpachilo had preferences for certain foods (based on 
their perceptions of the foods). Table 4.3 shows a free-listing of food preferences across 
different categories of community members.  
Table 4.3: Foods preferred and less preferred by communities at baseline 
Categories of participants Preferred foods 
Men, women and adolescent Dishes prepared with tubers such as yam and cocoyam  
girls Plantain 
Livestock products  
Vegetables and fruit including “kontomire”, banana, oranges, 
pear and apple 
Men and women:  Dishes prepared with cereals such as maize, guinea-corn and 
 rice  
-Dishes prepared with cow- and pigeon-peas and bambara bean  
Women and adolescent girls Dishes prepared with rice and cow-pea 
 Sugar-sweetened beverages such as Lipton tea; cocoa-drink, 
milo and soft drinks 
Women Soya bean 
 Less preferred foods 
 
Men and adolescent girls Soya bean 
 Orange-flesh sweet potato 
Adolescent girls Pigeon-pea 
 Nabinchingli (cocktail of different cereals and legumes) 
Cassava 
Men Bouillon cubes (magi) 
 
Table 4.3 indicates that men and women prefer meals made from cereals (maize, guinea-
corn, rice and millet) and legumes (cow-and pigeon-peas and bambara beans). In 
addition, the women prefer soya-bean and beverages such as tea and milo while the 
adolescent girls prefer foods such as tubers, rice, cow-pea and sugar sweetened 
beverages. The lists of preferred foods for the two study communities were identical at 
baseline.  
 
 
 129 
University of Ghana http://ugspace.ug.edu.gh
Reasons for food preferences: 
Participants gave a number of reasons for their preferences. Meals prepared from maize, 
millet and guinea-corn are preferred because of their availability and perceived energy-
density – that is, giving a full feeling over a long period after consumption. Women 
prefer soya bean because it can be used to prepare a variety of foods. Tea and chocolate 
drink are preferred by women because they are perceived to enhance production of 
breast-milk. Some men and adolescent girls, on the contrary, dislike soya bean because 
of its perceived association with stroke, stomach pain, diarrhoea, nausea and high blood 
pressure. Men dislike bouillon cubes because of their perceived association with male 
sexual dysfunction. 
 
4.2.2.5 Perceptions on nutritional adequacy of women’s diets 
Some women in both communities perceived the foods they consume as adequate in 
terms of their quantity and quality.  They indicated that the foods they consume meet 
their dietary needs because they are filling and healthy. Satisfaction with TZ was 
attributed to its status as the main local food: 
For us in the village, our main food is TZ; so when you wake up and get your TZ 
to eat, then you’re satisfied” (Pre-intervention IDI K35). 
 
The lack of meat in women’s diets was not perceived to have any negative effect on their 
nutritional adequacy as indicated by a female in-depth interviewee at Kpachilo:   
“Though we don’t add meat, we always add a lot of anchovies. We also cook 
vegetable soup to go with the TZ so our diets are of good quality in terms of 
nutrients” (Pre-intervention IDI K12).  
 
Satisfaction with the nutritional adequacy of locally-produced foods was another reason 
for one woman’s positive perception of their diets: 
 130 
University of Ghana http://ugspace.ug.edu.gh
I think that so far as we’re able to keep healthy by eating the foods we produce, 
I’ll say our diets give us all the nutrients we need to stay healthy”(Pre-intervention 
IDI Y34). 
 
On the contrary, other women and some men perceived the quality of women’s diets to 
be inadequate. They indicated that women and other members of their households 
consumed foods from a limited variety in their meals. One woman painted a picture of 
despondency in respect of the inadequacy of meat and fish in their meals: 
“I think the nutrient content could be improved with meat and more fish but, as it 
is, we can manage” (Pre-intervention IDI, K15b). 
 
 
The need to make the best out of a seemingly helpless situation was said to be 
responsible for the monotony in women’s diets: 
“The only thing is that these days foods like beans, yam and groundnut don’t 
grow well because of poor rains. So we don’t have enough of variety in our 
homes to cook with all the time. What we tend to eat more is TZ because we 
cultivate more of maize and manage it throughout the season” (Pre-intervention 
IDI, Y34). 
 
The perception held by some participants that most women’s diets were poor in terms of 
quality was emphasized by all key informants and most FGD participants during the 
community readiness assessment at Yilkpene. Participants opined that, even though they 
perceived women’s and other household members’ diets to be sub-optimal, there was 
very little they could do about the situation.  
 
From the above interaction with participants, the problem of women’s poor diets do not 
seem to be apparent to most participants in the two communities. First, participants 
linked women’s poor dietary quality with the periodic shortage of food (food insecurity) 
and not the socio-cultural aspects of the food system and community practices. They 
explained that food shortage led to the consumption of monotonous and unpalatable 
 131 
University of Ghana http://ugspace.ug.edu.gh
meals and the tendency to consume foods merely to satisfy hunger and not because of 
their nutritive benefits.  As a result, women’s diets tended to be poor in quality.  
 
Second, most participants linked the effects of women’s poor diets to their physical and 
psychological health as well as high cost of health-care but not to their reproductive 
health outcomes. Participants perceived that when women do not get quality food to eat, 
they tend to lose weight and their beauty. Women may also become ill and lack the 
strength to carry out their daily chores including the care of other members of the 
household, particularly children. The difficulties result in women becoming worried as 
found in the following narrative by a female key informant:  
“It worries us because if the children need food and there’s none, definitely, as a 
woman, you’ll think a lot about it and sometimes even cry secretly but what can 
we do? We just manage with the little we have” (Pre-intervention KII Y21).  
 
Another key informant emphasized the negative effect of women’s poor diets on peace 
in their households: 
 “It’s just the peace because without peace of mind at home, everything becomes 
bad (...). Yes, when a woman has no ingredients to prepare food, it disturbs her 
and she keeps on worrying” (Pre-intervention KII, K21). 
 
Participants perceived that women’s poor diets could lead to high cost of health-care as a 
result of ill health. Only a few key informants from Yilkpene perceived women’s poor 
dietary status as having negative effects on their reproductive health outcomes. 
 
4.2.2.7 Community members’ beliefs about foods (taboos and restrictions) 
Some gender dynamics were evident in beliefs about food, which to a large extent, 
impeded women’s access to optimal diets in the communities. Two main types of food 
restrictions were identified from participants’ accounts. While some restrictions were 
 132 
University of Ghana http://ugspace.ug.edu.gh
general and affected the entire community, particularly women, others were specific to 
women’s physiological statuses- that is, during pregnancy and lactation. Table 4.4 
presents a list of food taboos and restrictions affecting Kpachilo and Yilkpene at 
baseline. 
Table 4. 4: Food taboos and restrictions at baseline at Kpachilo and Yilkpene  
Target of food Restricted food Reason for adherence 
restriction 
Entire community Meat of pig, dog, frog Religious requirement-Islamic taboos 
All women of Python, dear, donkey, mud- Superstitious belief that children born 
reproductive age fish, rat, crocodile, alligator, of women who consume the flesh of 
cat such animals will give birth to 
children with resemblance in 
character to the animals 
Pregnant women Wild animals such as Baby may resemble animal and 
grasscutter, wild fowl actually exhibit animal behaviour 
(“yonoo”) and guinea-fowl 
(yokpang”) 
Cassava –based foods such as Reduced body fluid and blood 
“gari, and “kontonte” 
Salty foods Make legs swell (oedema of the feet) 
Mango Stomach pains in woman, itchy and 
reddish anus in unborn child 
Sugary and other sweet foods Difficult delivery and convulsion 
when baby is born 
Egg Children born are destined to become 
thieves 
Honey Causes stomachache and premature 
delivery 
Soya bean and pigeon-pea Cause stroke in pregnancy 
(“adua”) 
Oily and cold foods Cause nausea 
Lactating women Green leafy vegetables and Prevent healing of the womb and 
slimy foods such as okra, male child’s circumcised penis 
ayoyo and kuka 
Oily and chewable foods such Not allow healing of the womb, cause 
as rice and stew, jollof rice, stomach pains and do not facilitate 
waakye, adua, nyonbeka (mix production of breast-milk  
of grains and green leafy 
vegetables) and “nabinchingli” 
(mixture of cereals and beans); 
Sugary and sweet foods Cause nausea and stomach pains 
Disturb baby’s stomach after 
breastfeeding 
Dawadawa, mango and shea Itchy and reddish anus and passing of 
fruits greenish stools by babies 
Raw groundnut Enlargement of stomach wounds 
Adolescent girls Fish head and palm kernel Causes candidiasis  
 
 133 
University of Ghana http://ugspace.ug.edu.gh
Basically, there were no differences in the range of taboos and restrictions at baseline in 
the two communities. The consumption of the flesh of the pig, dog and frog was 
considered a taboo in the communities - thereby affecting all community members 
regardless of age or sex. The Islamic faith considers the consumption of pork a taboo 
because the pig is regarded as a sacred animal among Moslems. The dog, on the 
contrary, is considered a totem among the Dagomba tribe while the flesh of the frog is 
perceived to be poisonous. 
 
The consumption of the flesh of other animals such as the python, deer, cat, rat, monkey, 
crocodile, alligator and certain varieties of mud-fish (“zamasugu”) is considered a taboo 
for all women of reproductive age. All participants believe that consuming such flesh 
tends to negatively affect the behaviour of any children born subsequent to the 
consumption by their mothers: Female FGD participants in Kpachilo enumerated a 
number of animal source foods that were taboos in the community: 
“There are, however, animals like the crocodile and snake the flesh of which 
women don’t eat in this community but which men eat. There’re also the “woa” 
and “yogu” (alligator)- the flesh of which pregnant women are forbidden to eat 
because they’ve holes in the ears so when a woman is fond of eating them and 
she gives birth to a baby, it’ll have a similar hole in the body. We also have 
“zamazogu” (mud-fish) which lives in water and that one too, a pregnant woman 
is not supposed to eat its flesh because if she eats it and gives birth to a baby, it’ll 
always be puffing like the fish” (Pre-intervention FGD, women, K30). 
 
Restrictions on consumption of some animal- and plant-source foods were confined to 
pregnant and lactating women. Protection of both mother and the unborn child from 
harm was the main reason for the restrictions. Community-level beliefs in food taboos 
and other food restrictions was found to be high at both Kpachilo and Yilkpene at 
baseline because most of them were perceived to be part of their traditions: 
 134 
University of Ghana http://ugspace.ug.edu.gh
“Yes, we still adhere to these taboos even today (…). We adhere to them because 
they’re part of our Dagomba culture and we don’t want our children to exhibit 
such animal tendencies”(Pre-intervention FGD women, Y22). 
 
Similarly, participants in Kpachilo confirmed the adherence to taboos in their community 
as well:  
“Yes, because they’re part of our belief system. If you don’t abide by them and 
you encounter any problem, you have yourself to blame” (Pre-intervention FGD 
women K30). 
 
It was further revealed that participants were unwilling to relax some of the food taboos 
affecting women of reproductive age. The reason for the unwillingness was based on 
superstition - the belief that the taboos were meant to prevent the development of 
aggressive animal tendencies in their children (superstition): 
“No, we won’t allow the women to stop adhering to such taboos (...). There’s 
very little one can do about a tradition that’s been handed down to us……. In any 
case, why should we encourage the development of animal behaviour in our 
children?(...). Madam, there are certain cultural practices that protect us from 
harm and some of them are these taboos women are supposed to adhere to. If our 
women refuse to adhere to them, then you can imagine the kind of children we’ll 
be giving birth to” (Pre-intervention FGD males, Y35). 
 
Subsequent discussion with a key informant confirmed that, even though some 
individuals in the communities did not adhere to some of the taboos, other community 
members did so. Adherence to food taboos and food restrictions among women, 
especially during pregnancy and lactation, was perceived by a few male key informants 
at Yilkpene to limit the variety of foods available to women. For instance, restrictions 
against the consumption of pigeon-pea, wild animals or bush meat, chewable foods, 
some fruits and especially green-leafy vegetables by pregnant and lactating women were 
perceived to limit the variety of food items available to them:  
 135 
University of Ghana http://ugspace.ug.edu.gh
“It means she’ll have only one kind of food to eat and beside this, it can happen 
that she’ll be fed up with that soup and that means her food intake will decrease 
and that can harm her health. If you eat a particular soup in the morning, 
afternoon and evening, then, the next day, you’re given the same soup again, you 
won’t be able to consume the food as much as you did the previous day and that 
alone is harmful to women’s health. Apart from the woman becoming fed up with 
the food, she won’t get the benefits the consumption of vegetables can bring to 
her health because she doesn’t consume them” (Pre-intervention KII, Y41). 
 
Most participants including all the females at Yilkpene and all participants at Kpachilo, 
however, did not think that food taboos and restrictions had any negative impact on 
women’s diets. Their main reason was that most of the animal-source taboos did not 
form part of their everyday diets and, therefore, restricting their consumption among 
WRAs did not pose a threat to women’s diets: 
“You see, let’s talk about the snake; it can happen that for about five years or 
more, in this community no one would have killed a python. So it isn’t that we 
even have those animals in the community these days. They’re no longer there 
(…). As we hunt them they also run for their lives and so they’re no more there. 
We don’t even have them anymore to eat and for that matter, there’s no 
effect”(Pre-intervention FGD men, Y31).  
 
4.2.3 Women’s dietary practices  
4.2.3.1 Women’s eating pattern and dietary diversity 
In the two communities, women’s diets are essentially not different from those of the 
entire household. The diets consist mainly of grain-based foods. Tuo-Zaafi (TZ) and a 
variety of porridges usually made from maize and guinea-corn are the main staple foods. 
TZ is usually accompanied with green-leafy vegetable soups in which anchovies are a 
common feature. Groundnut is also consumed quite frequently. Legume-based foods are 
consumed when in season but meat and fruits are less frequently consumed.  
 
Some women reported consuming beverages and a variety of porridges in the early 
morning followed by mid-morning, afternoon and evening meals. The early-morning 
 136 
University of Ghana http://ugspace.ug.edu.gh
meal usually consists of Lipton tea with sugar and sometimes with milk, chocolate or 
milo drink, a variety of porridges made from cereals (“koko sali”, “zimbuli”, 
“kokonyina” and “kanwa koko”) and sometimes consumed with roasted groundnut or 
wheat bread. The mid-morning meal, traditionally referred to as the “nagban suli” 
(meaning breaking of a fast), usually includes dishes prepared from rice, cow- and 
pigeon-peas and yam. Sometimes left-over TZ from the previous night may be 
consumed. The afternoon and evening meals are mainly TZ with a variety of vegetable 
soups. Table 4.5 presents the frequency of meals consumed by the 40 women at 
Yilkpene and Kpachilo in May-June 2016 (perceived period of food shortage) and 
November 2016 (perceived period of food availability). 
Table 4. 5: Frequency of meals at baseline  
Freq. of meals Yilkpene Kpachilo 
 May/June 2016 November2016 May/June 2016 November 
2016 
Less than three 0 0 2 (10%) 0 
Three 3 (15%) 8 (40%) 6 (30%) 9 (45%) 
Four or more 17 (85%) 12 (60%) 12 (60%) 11 (55%) 
No. 20 20 20 (100%) 20 
Total frequency 81 73 72 75 
Mean frequency 4.01 3.65 3.6 3.75 
 
Majority of women at Yilkpene and Kpachilo consumed four or more meals during the 
two periods of data collection at baseline. Eighty-five percent of women at Yilkpene 
consumed four or more meals during the period of May/June 2016 while 60% at 
Kpachilo did so in the same period. The proportion of women consuming four or more 
meals in November 2016, when food was perceived to be more available, was lower in 
both communities (Yilkpene-60% and Kpachilo-55%) compared to the period May-June 
2016. The mean frequency of meals consumed at Yilkpene in May-June was higher 
(4.01) than that at Kpachilo (3.6) for the same period. 
 137 
University of Ghana http://ugspace.ug.edu.gh
Although purchase of ready-to-eat foods (RTE) is uncommon, some women reported 
eating RTE foods while at the market or attending social functions. The foods most 
likely to be purchased were “kenkey” with pepper sauce, “waakye” (rice and beans), 
bread and fried pastries. At social functions such as funerals and naming ceremonies, 
Tuo-Zaafi and soups may be served with meat.  
 
Presented in Table 4.6 is a breakdown of the Minimum Dietary Diversity of women 
(MDD-W) at Yilkpene and Kpachilo in May-June and November 2016 at baseline.  
 
Table 4. 6: Minimum Dietary Diversity of Women (MDD-W) at Yilkpene and 
Kpachilo in May-June and November 2016 at baseline 
MDD-W Yilkpene Kpachilo 
    
May/June November  May/June November (N=20)  
(N=20) (N=20) (N=20) 2016 
2016 2016 2016 
Yes 11 (55%) 13 (65%) 11 (55%) 16 (80%) 
No 9 (45%) 7 (35%) 9 (45%) 4 (20%) 
Average 4.45 5.1 4.6 5.1 
DDS 
Note: MDD- W –Minimum Dietary Diversity of Women (Foods belonging to five and more groups 
consumed).  
Yes- Number or percentage of women scoring five or more food groups (or meeting the minimum dietary 
diversity score.  
No-Number or percentage of women scoring less than 5 food groups (or not meeting the minimum dietary 
diversity score).  
Average DDS - Mean Dietary Diversity Score of women surveyed 
 
About half (55%) of women reported achieving the minimum dietary diversity score of 
five good groups or more in the period May-June 2016. In the baseline period of 
November, however, a higher proportion of women in both communities met the MDD-
W (Yilkpene -65% and Kpachilo - 80%). The gap between the scores for the two periods 
was, however, higher at Kpachilo (25%) than Yilkpene (10%). The mean dietary 
diversity score (DDS) for Kpachilo was higher than that for Yilkpene in respect of the 
 138 
University of Ghana http://ugspace.ug.edu.gh
baseline period May-June 2016 (4.6 as compared to 4.45 respectively) though both 
communities had a mean DDS of 5.1 in respect of the baseline period of November 
2016. 
 
4.2.4 Drivers of women’s dietary choices 
Seven main drivers of women’s dietary choices emerged from the various discussions 
with community members. The identified drivers are household food security, tastes of 
food, women’s physiological status (e.g. pregnancy), perceived healthiness of food, food 
restrictions, sources of information on diets and large sizes of households.  
 
Figure 4.1 presents the seven emerging themes on what drives women’s dietary choices 
at Yilkpene and Kpachilo in northern Ghana. The seven drivers of women’s dietary 
choices were observed and concluded from the participants’ accounts. 
 
Food security in households 
 
 
 
Food taboos Women’s Perceived Large sizes Sources of Tastes of 
 and physiologi- healthiness of informatio foods 
restrictions cal status  of foods  households  n on food 
 
 
 Drivers of women’s 
food choices  
 
Figure 4.1: Drivers of women’s dietary choices 
 
Food security: Food security at the household level was identified as one of the major 
drivers of women’s food choices through availability and cost of food, food preferences, 
 139 
University of Ghana http://ugspace.ug.edu.gh
food safety and offer of gifts of food in the communities. Availability of food was 
affected by seasonality, climate change and availability of farming inputs. In the opinion 
of female focus group discussion participants at Kpachilo, their diets tend to be 
dominated by maize-based products such as TZ because that is the crop produced mostly 
in the community: 
“The only thing is that these days foods like beans groundnuts and yam don’t 
grow well because of poor rains. So we don’t have enough of these in our homes 
to be cooking all the time. What we tend to eat more is the TZ because we 
cultivate maize more and even if the rains are poor, we still manage to get 
enough to last till the next farming season”(Pre intervention FGD, women K34). 
 
Foods that were relatively affordable were more likely to be purchased and consumed. 
Examples of such foods were anchovies (fish) which were consumed daily because they 
were affordable animal-source food, tea, sugar and tinned tomato in small sachets. 
Incomes of households relative to prices of foods determine their scopes of 
affordability/accessibility. The issue of cost was raised in relation to the consumption of 
anchovies compared to meat by female participants: 
“Usually we don’t cook with meat. It’s mainly yurayura (fish) we buy and pound. 
Fish is much cheaper than meat. Can you imagine making all those expenses to 
cook a meal and in addition buying meat to add to the food? No madam, it’s too 
much for us women. We can manage to make the food taste nice and eatable 
without meat. It’s too expensive for us” (Pre intervention FGD women Y34). 
 
Women reported that the food preferences of other members of their households 
influenced their choices of food. Some women reported that their husbands did not like 
soups prepared with some varieties of green-leafy vegetables such as cassava and bean 
leaves as well as soya bean products - thus, limiting their consumption of such foods in 
their households. In respect of men’s food preferences, a male FGD participant 
 140 
University of Ghana http://ugspace.ug.edu.gh
confirmed women’s assertion about men’s aversion for green-leafy vegetables and the 
reason why less time was devoted to cultivating them: 
“Madam, one reason is that green-leafy vegetables aren’t our favourite soups. 
But as for okro, we can eat it on a daily basis and never get fed up with it. That’s 
why we cultivate okro in large quantities and care less about green-leafy 
vegetables” (Pre intervention FGD, Men Y31). 
 
Participants were very concerned about issues of safety regarding their farm produce 
particularly the use of agro-chemicals in the production of fruits and vegetables. For 
instance, as a result of the use of chemicals to facilitate the quick ripening of tomato in 
particular, most women reported resorting to using processed tomato preserved in 
sachets in place of the fresh ones even when the latter was in season. Similarly, women 
were concerned about infestation of their farm produce by insects (weevils) that they 
considered unsafe for consumption. In respect of using agro-chemical on tomato crops, a 
female participant explained her preference for processed and package tomato from the 
market: 
“…they spray the chemicals on okro and tomato and because of that, most of us 
fall sick. It’s now very difficult to cook with them. So you see, as we sit now, we 
use tinned tomato to cook our foods because fresh tomato this time contains 
chemicals” (Pre intervention. FGD women Y32). 
 
The practice of making food available through the generosity of food donors within the 
community was a common practice. Households that were better off in terms of the 
availability of food donated food to vulnerable households and individuals:  
“If you bring the food home after harvesting it, you have to give some to the chief 
of the community, your Imam and, sometimes too you, could also give to the aged 
in the community” (Pre-intervention KII male, K22). 
  
Taste of food: The second set of drivers of women’s dietary choices was the tastes of 
food. Anchovies and bouillon cubes were commonly consumed because they were 
 141 
University of Ghana http://ugspace.ug.edu.gh
perceived to enhance the tastes of meals. A female key informant explained why the use 
of anchovies and bouillon cubes was common in the communities: 
“We also add fish to our food to make it taste good but we don’t have any river 
here. We use magi although they say it isn’t good but it makes our food tasty and 
so we still use it” (Pre-intervention KII, Female k25). 
 
 
The consumption of beverages and sugar by the adolescent girls was influenced by their 
good tastes. 
“We like tea and milo because they taste nice and they help women to produce a 
lot of breast-milk for the children to suck. So lactating women too like them” 
(Pre-intervention. FGD with adolescent girls, K31). 
 
Women’s physiological status: The third driver was women’s physiological status. A 
number of lactating women reported consuming beverages such as tea and chocolate 
drink with sugar and sometimes milk because they were believed to enhance production 
of breast-milk. Similarly, pregnant women indicated their aversions to certain foods 
which were perceived to cause nausea and vomiting and their cravings for other foods 
(the consumption of pica). 
 
Perceived healthiness of foods: The fourth driver was the health-conferring ability of 
foods. Foods prepared from maize, millet, guinea-corn, fish, meat and vegetables were 
perceived as foods that were healthy and, therefore, consumed more frequently. The 
foods were perceived to give energy, strength, blood, induced drinking of a lot of water 
and profuse sweating and enhanced health in general. In respect of dawadawa, which 
forms part of their everyday meals, an FGD participant enumerated its perceived 
healthiness: 
“Well if there’s no dawadawa in the soup, we don’t even feel like eating the food 
because the taste won’t be nice and it’s also not considered healthy enough… So 
we never run short of dawadawa. Dawadawa has some health benefits such as 
 142 
University of Ghana http://ugspace.ug.edu.gh
giving blood. Some time ago, you could just make dawadawa solution and eat 
your TZ with it without adding anything and that was healthy”(Pre intervention 
FGD women, Y34). 
 
Food taboos and restrictions: The fifth driver was adherence to food taboos and 
restrictions, which emanate from tradition or religion of a people. Restricting women and 
other categories of people from the consumption of some animal- and plant-source foods 
limited the availability of a variety of foods in their diets. Restricting lactating women 
from consuming slimy, green-leafy vegetables, mango and other wild fruits, for instance, 
deprived them of their protective properties. One key informant stressed the influence of 
food taboos, whether traditional or religious, in determining and actually limiting the 
availability of animal-source foods for members of the community. 
“Added to these factors is the role of religion. Our setting here is more of Islamic 
influence, whereby certain foods aren’t consumed among us just because of 
religious reasons. For instance, the snail isn’t eaten in this part of the country 
and neither are some kinds of bush meat our men go to hunt. Hunting is very 
common here but more often than not, the catch is sold for cash partly because 
some of the animals they hunt aren’t consumed by our people because they’re 
taboos. Otherwise, the catch from the bush would’ve been a good source of 
proteins since the animals we rear are meant for cash and not for home 
consumption”(Pre intervention KII). 
 
Sources of information on food: Sixth, the kinds of information received about diets 
from women’s sources of dietary information influenced their choices of foods. Women 
reported being encouraged by health professionals to consume some categories of foods 
such as eggs, beans and vegetables to enhance their health statuses: 
“Ok, when we go to the hospital, we’re always told to eat a lot of vegetables like 
ayoyo, alefu, and bra. We’re also told to eat a lot of  amani and dwadawa and 
prepare the foods under hygienic conditions so that when the children eat, they 
and the women will be healthy” (Pre-intervention FGD women, k30).  
 
Large sizes of households: Finally, some key informants indicated that large sizes of 
households in the communities were barriers to access to food owing to their low 
 143 
University of Ghana http://ugspace.ug.edu.gh
incomes and communal system of eating. A ley informant reiterated how the large sizes 
of households in the communities limit their ability to consume adequate nutritious 
foods: 
“I think it’s more of survival than getting quality food to eat. Again, the extended 
family system isn’t helpful here. How much meat can a household afford to feed 
its members who are over 30 people for a single meal? It’s impossible. If the 
sizes of the households were smaller, between 4 and 5 or so, then once a while, 
they may be able to buy meat to cook. But with huge members like these, how can 
they afford it especially when the man provides only the grains and the woman is 
expected to provide the ingredients for soup - when she has no meaningful source 
of income” (Pre intervention KII). 
 
4.2.5 Community readiness to address challenges to women’s diets 
Having understood women’s dietary practices and the context that influenced them, it 
was necessary to assess Yilkpene, the intervention community’s, readiness for an 
intervention targeted at socio-cultural barriers to women’s diets as a pre-condition for its 
effectiveness (Stanley, 2014; Findholt, 2007; Edwards et al., 2000).  The readiness was 
assessed by six key informants and participants in four focus group discussion sessions 
in the community and measured on five dimensions (Stanley 2014). 
 
The distribution of community readiness scores awarded by the six key informants on 
the basis of the 5 dimensions of the community readiness model (CRM) is presented in 
Table 4.7. 
 144 
University of Ghana http://ugspace.ug.edu.gh
Table 4. 7: Community readiness scores obtained from key informants 
Dimensions Key informants (N=6) Total Mean Stage of readiness 
01 02 03 04 05 06 score score 
1. Knowledge of 2 6 8 5 5 3 29 4.83 4- Preplanning 
the issue of 
women’s diets 
2. Knowledge of 2 5 4 5 5 5 26 4.30 4-Preplanning 
local efforts 
addressing 
challenges to 
women’s diets 
3. Community 2 6 1 3 1 2 15 2.50 2 -Denial/resistance 
climate 
4. Leadership 1 4 8 1 1 2 17 2.83 2- Denial/resistance 
5. Resources 3 3 5 1 1 5 18 3.0 3- Vague awareness 
Total score 10 24 26 15 13 17 105 3.5 (105/(6x5))= 
(105/30)=3.5 
Mean score 2.0 4.8 5.2 3.0 2.6 3.4 21 3.5 (21/6)=3.5 
Overall stage of         3 – Vague awareness 
readiness 
 
The overall mean readiness score for Yilkpene was 3.5, which corresponds to the third 
lowest of the nine stages of community readiness - the stage of vague awareness. At this 
stage, community members had begun to recognize the existence of the problem of 
women’s diets but they lacked the motivation to do anything about it. 
 
4.2.5.1 Knowledge of women’s sub-optimal diets 
The readiness score for knowledge of the issue of women’s poor dietary quality was low 
but above the overall mean score (4.83 compared with 3.5). The score of 4.83 suggests 
that community members including some key informants had limited knowledge about 
the issue of women’s poor diets.  
 
4.2.5.2 Community knowledge of efforts addressing challenges to women’s diets 
The readiness score for knowledge of community efforts (that is, knowledge of the 
programmes and policies in place to address the issue of women’s poor dietary qualities) 
 145 
University of Ghana http://ugspace.ug.edu.gh
was low (4.3). Some community members reported efforts to address women’s poor 
diets but knew little about them. The media, particularly radio, was the only local effort 
reported by participants which aimed at promoting activities related to diets in 
pregnancy: 
“Zaa Radio usually tells us about good eating habits for pregnant women and for 
the community” (CRM, Y03, women). 
 
Other activities of Ghana Health Service, Resilience in Northern Ghana (RING) project 
and the media were mentioned as efforts at addressing the challenges pertaining to the 
knowledge aspects of women’s diets. Ante-natal and child welfare services, promotion 
of good sanitation practices and use of insecticide-treated nets also provided knowledge 
services:  
“The health workers do come around and they talk to us on what a pregnant 
woman should eat, do and not do. They said a pregnant woman shouldn’t 
overwork herself and that we should try and get rest during our working 
sessions”(CRM Y03, women). 
 
RING assisted the community with farming inputs such as seeds, cutlasses and boots as 
well as sacks and silos for storing grains. Additionally, each woman with farmland was 
assisted to plough an acre. Participants were, however, of the view that coverage of such 
activities was limited for they were targeted mostly at women with children and pregnant 
women. A woman felt dissatisfied with the limited coverage of Ghana Health Service’s 
activities in respect of their focus on only women with young children: 
“Honestly, when the hospital workers come, they gather pregnant women and 
those with little children to talk to them; they usually don’t include some of us” 
(CRM, Y03, women). 
 
Paradoxically, a male key informant defended the concentration of activities on women 
with children and pregnant women: 
 146 
University of Ghana http://ugspace.ug.edu.gh
“But when they come here, their focus is always on the women with children   
because it’s the women who take care of us all”(CRM, Y04, men). 
 
Participants’ impressions about the various efforts in the community aimed at improving 
women’s diets were positive. They explained that the “adakabila” (a savings scheme) 
programme, for instance, was helpful in improving women’s financial situation. All the 
other programmes in the community including those of Ghana Health Service were 
considered useful but had limited coverage. Members of the community, therefore, 
suggested that future intervention activities should target men and other members of the 
community as well: 
“They should involve the husbands too and even the children especially the 
adolescent boys and girls because they have the brains to do something” (CRM, 
Y11, KII). 
 
4.2.5.3 Community climate regarding women’s sub-optimal diets 
The score for community climate – that is, the prevailing attitude of the residents of 
Yilkpene towards women’s poor diets – was the lowest among the five scores (2.5) and, 
thus, fell far below the overall mean of 3.5. The score suggested that residents of 
Yilkpene believed that the issue of women’s poor diets was either not a concern in their 
community or an issue that could not be addressed. In contrast, focus group discussion 
participants wished something could be done about the situation. Two female focus 
group discussion participants echoed the dismal assessment of women’s diets, wishing 
something could be done to improve their helpless situation: 
“It’s a heartache to us and we wish something could be done about it (CRM 
FGD Y03, women). 
 
Another female key informant supported the view that there was nothing they could do 
about their current situation regarding their diets:   
 147 
University of Ghana http://ugspace.ug.edu.gh
“Yes, we know but we’re at our wits’ end” (CRM, FGD Y05, women).  
 
Further probing by the research team to fully understand what they meant by “something 
could be done about it” led one male participant to refer to the poor status of preservation 
of farm produce - an important determinant of women’s poor dietary quality:  
“We’re supposed to preserve our harvests well so that there won’t be insect 
infestation” (CRM FGD Y14, men). 
 
Some of the activities they wished to be implemented included provision of income-
generating activities, loans, agricultural extension services and a dam to produce water to 
irrigate their crops during the dry season. The dam would also address their acute 
domestic water situation. 
 
4.2.5.4 Community leadership  
The readiness score for leadership was low (2.83) so the members of the community 
were not spurred to carry out activities related to women’s diets. There was consensus 
among male and female focus group discussion participants that community leaders were 
supportive of interventions aimed at addressing some issues related to women’s diets. A 
few key informants, however, felt otherwise and indicated that community leaders 
(chiefs and elders) were not doing much to support efforts in the community: 
“They (the community elders) are happy about our (GHS) efforts but the support 
is zero because at most durbars organized by us, we don’t see them. Some of the 
community leaders don’t come for the durbars” (CRM Y10, KII). 
 
Only one key informant was quite satisfied with leadership support and scored it highly 
(8) to agree with the focus group discussion participants’ views. FGD participants who 
maintained that community leadership were in support of programmes targeted at 
 148 
University of Ghana http://ugspace.ug.edu.gh
challenges to women’s diets indicated that leadership showed their support by 
welcoming interventionists into the community, mobilising the community for outreach 
activities, encouraging community members to participate in activities and, occasionally, 
meeting with community health volunteers to assess progress of intervention activities. 
Female FGD participants recounted the support shown by their community leaders in the 
following narratives: 
“They always mobilize us for such activities. Sometimes even if we’re on the 
farms, they send for us to come and that tells us they’ve interest in the activites. 
They also go round and write our names if there’s the need for that” (CRM FGD, 
05, women). 
 
Another female participant added: 
 “The chief and other community leaders show they’re in support by welcoming 
the people warmly and asking the community to co-operate with them. If the 
people even want people to train, the community leaders do assist them in 
selecting those people. If they aren’t in support of whatever activities are going 
on in the community, then they won’t even mind you” (CRM FGD 02, women). 
 
4.2.5.5 Community resources 
The readiness score for resources was low (3.0) indicating there were limited resources 
to address women’s poor diets. Key informants were unanimous about the inability of 
Yilkpene to support activities to address barriers to women’s diets since they lacked 
financial resources for such activities.  A key informant indicated that the community as 
a whole lacked financial resources dedicated to addressing women’s dietary challenges:  
“We’ve no financial resources to devote to address this issue except our time and 
prayers. We’ll always be there to listen to what they bring to us” (CRM KII 08).  
 
Both focus group participants and key informants indicated that the community was able 
to sustain the “adakabila” project for the past two years because of individual women’s 
contributions into the pool. Participants stated that they could make similar contributions 
 149 
University of Ghana http://ugspace.ug.edu.gh
towards similar projects to enhance women’s diets if called upon to do so. One key 
informant indicated community members’ willingness to make communal financial 
contributions towards activities aimed at addressing challenges to women’s diets. He 
cited an instance in which the community made some financial commitment towards 
building a pen for keeping livestock before receiving assistance from an NGO: 
“When the RING people were coming here, they selected 38 households and each 
received three sheep. Before you take the sheep, you have to build their pens and 
for you to do that means that you have to spend money” (CRM KII Y11,). 
 
Even though the key informants indicated insufficiency of personnel to carry out health 
programmes in the community, data from the focus group discussion sessions revealed 
that the community had people who could be trained to deliver the required services. 
They mentioned community volunteers and other persons who were available to engage 
in activities to address issues related to women’s diets. They indicated that the 
community volunteers had received a lot of training and were in a position to carry out 
the intervention activities in the community. Participants mentioned “Amasachina” 
Group (women’s group), “Ataya” (Male youth group which also doubled as the 
sanitation group) and the local drama group as local groups that could be used to further 
women’s dietary activities.  
 
The Infant and Young Child Feeding Mother Support Group, traditional birth attendants, 
School Health Teachers and Agricultural Extension Services Division of MOFA were 
also mentioned as potential groups to be used to further women’s dietary activities. 
Ghana Health Service’s monthly outreach programme was the most notable action 
mentioned. Participants stated that outreach programmes were organized once every 
month for pregnant women and women with children less than 5 years old in the 
 150 
University of Ghana http://ugspace.ug.edu.gh
community. A female participant emphasized the outreach activities of Ghana Health 
Service: 
“Usually, when the hospital workers come, they gather pregnant women and 
those with little children to talk to; they usually don’t include some of us”(CRM 
FGD 03, women). 
 
 
Finally, the community mentioned land and nutritious food crops as resources available 
to enhance efforts aimed at addressing issues related to women’s diets: 
“We don’t have anything but we can give out our lands so that, they’ll help us 
cultivate our vegetables. We also have foodstuffs like soya bean, livestock, maize, 
yam and millet” (CRM FGD 04, men). 
 
Results of the community readiness assessment showed that the overall readiness of 
Yilkpene for an intervention aimed at addressing socio-cultural barriers to women’s diets 
was low - an indication that, even though some community members were aware of 
women’s dietary challenges, very little community efforts were in place to address the 
situation. In spite of this limitation, Yilkpene was endowed with human resources to 
implement an intervention to address the situation. Additionally, existing structures such 
as the monthly health and nutrition outreach activities of Ghana Health Service in the 
community, the community health volunteers and community venues for community 
meetings were some of the resources that could be utilized to carry out sustainable 
interventions in the community. 
 
 
4.2.6 Description of implementation of intervention 
An intervention was implemented at Yilkpene to address the challenges to women’s diets 
identified earlier in the baseline phase of the study. This section presents results of the 
community exposure to the intervention (its coverage) and the modification and 
adaptations of the original plan of the intervention. 
 151 
University of Ghana http://ugspace.ug.edu.gh
4.2.6.1 Exposure to intervention activities  
The data presented in Table 4.8 gives a break-down of the levels of dosage or exposure 
to group meetings by community members (i.e., frequency of participation in the group 
discussions). 
Table 4. 8: Frequency of participation in group meetings 
        (a)                       Groups               Total 
Frequency of Men      Women  
participation in (b) (c) (d) (e) (f) (g) 
group meetings  No.  Aggregate No.  Aggregate Attendance at Aggregate 
attendance attendance group meetings  attendance  
(axb) (axd) 
12 7 84 18 216 25 (10.9%) 300 
11 4 44 6 66 10 (4.3%) 110 
10 0 0 3 30 3 (1.3%) 30 
9 3 27 8 72 11 (4.8%) 99 
8 7 56 29 232 36 (15.7%) 288 
7 15 105 19 133 34 (14.8%) 238 
6 20 120 16 96 36 (15.7%) 216 
5 20 100 25 125 45 (19.6%) 225 
4 9 36 10 40 19 (8.2%) 76 
3 6 18 4 12 10 (4.3%) 30 
Total 91  590 138 1, 022 229 1, 612 
Mean Attendance - 6.5 - 7.4 - 7 
 
Table 4.8 shows that a total of 229 persons (138 women and 91 men) attended the 
weekly group meetings. Eighteen (13%) of the women out of 138 attended all 12 
sessions of the group meetings compared to 7 (7.7%) out of 91 men. The least rate of 
attendance at group meetings was 3. Majority of participants attended between 5 and 8 
meetings. The respective mean attendances or dosages were 7.4 for women and 6.5 for 
men. All the participants had a mean attendance/dosage of 7. The aggregate attendance 
consisted of the product of a group’s frequency of participation in group meetings and 
the size of the group. 
 152 
University of Ghana http://ugspace.ug.edu.gh
The intervention was intended for all adult members of the community comprising men, 
women and adolescents girls. Table 4.9 gives a description of adult population 
participation at group meetings. 
Table 4. 9: Coverage of group meetings during implementation 
Population Population Total number of Proportion at 
category group meetings group meetings 
Adult males 273 91 33.3% 
Adult females 329 138 41.9% 
(including 
adolescent girls) 
Total adults 602 229 38% 
 
The adult population of Yilkpene at the time of the intervention was 602 (273 men and 
329 women including adolescent girls). Less than half (38%) of the target population 
consisting of 33% men and 42% women was reached. While the first durbar recorded 
265 participants (101 men, 151 women and 13 adolescent girls), the second recorded 174 
adults (102 women and 72 men). The first food demonstration session (training on post-
harvest losses) recorded an attendance of 203 (121 women and 82 men) but the cooking 
demonstration a total of 176 (115 women and 61 men). The adolescent girls were, 
however, not reached as desired (only 13) because most of them had migrated to the 
southern part of the country to work as porters. Each compound was visited at least once 
during the intervention and at least one member of each of 55 out of the 56 compounds 
or households was reached with the messages. 
 
4.2.6.2 Modification and adaptation of initial plan of implementation 
The initial plan of implementation and the actual plan that was executed are presented in 
Appendices L and M. In the initial plan, the stakeholders’ meeting and the training 
sessions were to be held for two days each while the other activities were to be held 
 153 
University of Ghana http://ugspace.ug.edu.gh
weekly but owing to community requests, some modifications were made to it 
(Appendix L) resulting in an extension of the duration of intervention activities from 12 
to 17 weeks (Appendix M). For instance, minor modifications were made to the food 
demonstration sessions so that one session concentrated on post-harvest and food 
preservation methods but the other on cooking demonstration of local dishes fortified 
with soya-bean flour.  
 
This change was made specifically at the request of the men who felt that most of the 
intervention activities and messages concentrated too much on women. Again, twelve 
group meetings were held instead of the planned five. This major modification came 
about because community members made a request for some of the messages to be 
repeated and also because of reported bottlenecks in respect of men’s aversion to 
consuming some green-leafy vegetables and difficulty in processing fermented grains as 
well as in accessing a variety of fresh vegetables all year round. 
 
4.3 Post-intervention results 
4.3.1 Post intervention changes in community dietary knowledge and attitudes at 
Yilkpene 
At endline, both communities did not change their practices in respect of the food system 
but Yilkpene was sensitized (raised consciousness) about the gender imbalances in 
respect of women’s limited control over productive and household food resources, 
appropriate techniques of preservation and storage of food and preferential food 
allocation arrangements in the community.  With respect to the traditional arrangement 
for sharing meat, for instance, one male FGD participant indicated his awareness of the 
 154 
University of Ghana http://ugspace.ug.edu.gh
need to share food according to need but, at the same time, expressed his reservations 
about the new practice: 
“Normally, the children are given the legs and the heads and women are given 
the back of the animal while the men and elderly take the thigh and shoulders; we 
now know that the children and women should be given the soft parts but we 
can’t do that because if you do that for the children, they’ll grow up to be 
wayward….Yes, the child may go about killing people’s fowls or stealing them. 
As a child grows, he’s supposed to go through some training and all this is part 
of it” (Post-intervention FGD men, Y31). 
 
4.3.1.1 Improved dietary knowledge  
Some practical changes were, however, observed in respect of the dietary knowledge and 
attitudes in both communities. Dietary knowledge regarding the nutritive qualities of 
locally-produced foods generally improved in both communities at endline. At Endline at 
Yilkpene, there was a general consensus among focus group discussion participants on 
the right classification of locally-produced foods into the three food groups: energy-
giving, body-building and protective foods. Typical examples were maize, guinea-corn, 
yam and rice as energy-giving; bamabara beans, pigeon- and cow-peas as body-building 
and leafy vegetables such as alefu, ayoyo bra and kuka as protective and body-building. 
The improved level of knowledge in respect of the right classification of foods by men 
depicted in the following quotation: 
“ For instance, there are some foods that build the body, some give energy and 
strength and others protect our bodies against diseases (…) TZ and fufu give us 
energy and strength and the fruits protect our bodies and destroy certain diseases 
in them”(Post-intervention FGD men, Y 29). 
 
Similarly, the following quotation exemplifies women’s improved level of knowledge on 
the right classification of foods: 
“Maize gives us strength to carry out our daily activities (…) and examples of 
foods that give blood are moringa leaves, ayoyo and the other leafy soups (...) 
fish builds our bodies and keeps us fit and the fruits protect our bodies against 
diseases”( Post-intervention FGD, women Y30) 
 155 
University of Ghana http://ugspace.ug.edu.gh
At Kpachilo, however, misclassification of locally produced foods into the food groups 
persisted among focus group discussion participants. For instance, meals prepared from 
maize, bambara bean, cow- and pigeon- peas were associated with giving of blood 
(body-building) and energy; rice and yam with giving energy and groundnut, moringa, 
sorghum and millet with building the body as some FGD participants recounted. Female 
participants, during the post-intervention focus group discussions, misclassified some 
food items: 
“Maize, cassava and Bambara beans give us energy…bambara and cow-peas 
make us healthy… beans give us blood, fufu is very healthy and gives us energy. 
We also produce guinea-corn and millet and they build our bodies too and 
moringa leaves prevent malaria” (Post-intervention FGD women, K26). 
 
Endline participants at Kpachilo gave examples of foods that were needed to enhance 
WRA’s health as egg, bean, fruit and vegetables. They also mentioned meals prepared 
from cow-pea, maize and yam as well as beverages, palm products, fruits and animal-
source foods as foods needed specifically by pregnant women. In addition to all the 
foods listed by participants at Kpachilo as needed by pregnant women, Yilkpene added 
groundnut paste, hot foods (such as TZ with bra), water and foods prepared with 
dawadawa. They further indicated that foods needed to enhance adolescent girls’ health 
include cow-pea, leafy vegetables, fish, egg and meat as indicated by a female 
participant in an FGD: 
“They should eat food like beans, alefu and plenty fish and meat. They should eat 
leafy vegetables” (Post-intervention intervention FGD Women, Y33) 
 
At endline at Kpachilo and Yilkpene, health facilities were rated the main source of 
information on health and nutrition. Additional sources mentioned were community 
health workers (volunteers) and health workers engaged in outreach programmes, 
television and radio:  
 156 
University of Ghana http://ugspace.ug.edu.gh
“We get information on nutrition on radio and television, health centres and 
hospitals and the outreach health services as well as from sanitation officers” 
(Post-intervention intervention FGD women Y33). 
 
Kpachilo mentioned non- governmental organizations such as RAINS, World Vision –
Ghana and ADRA-Ghana as the other important sources. Face-book and the inter-net 
were both sources for adolescent female students at Kpachilo. At endline at Yilkpene, 
the only additional source of information on women’s diets was the intervention team. 
 
4.3.1.2 Changes in dietary perceptions at Yilkpene 
At endline, the three additional perceptions of good health mentioned by participants at 
Yilkpene were all identified by female participants. The first perception was sound 
nocturnal sleep. 
“If you’re able to sleep well at night without any pains and you wake up feeling 
good, then we can say you’re healthy” (Post-intervention FGD women, Y28 
Category 2). 
 
 
The second additional perception was not being bed-ridden as narrated by another focus 
group participant: 
“I’ll also say that a healthy person is someone who isn’t bed-ridden” (Post-
intervention FGD women Y30, Category 1).  
 
The third additional perception was the absence of dizziness: 
“If one’s feeling dizzy, then one isn’t healthy” (Post-intervention FGD women, 
Y33 Category 3). 
 
Participants at Yilkpene identified additional attributes of healthy foods than in both the 
baseline study and at endline at Kpachilo. Attributes related to the contents of a meal 
included palm oil and tomato with very little salt and pepper. Orange-flesh sweet potato, 
soya and bambara beans, groundnut and cow-pea without weevils and holes in them and 
 157 
University of Ghana http://ugspace.ug.edu.gh
vegetables not being overdone were additional attributes mentioned. Attractive 
colour/look of the food (red-coloured soup being the favourite) and aroma of the soup 
were other attributes Yilkpene community members considered healthy. 
 
Participants in both communities added that because of the use of chemicals in spraying 
crops such as tomato and okro, cooking with such vegetables was not healthy:  
“Healthy foods are those they don’t spray with chemicals but in our community, 
this is what they do; they spray the chemicals on vegetables such as okro and 
tomato and because of that, most of us fall sick”(Post-intervention FGD women, 
Y28 Category 2).   
 
Furthermore, the use of bouillon cubes in preparing meals was perceived by members of 
both communities as unhealthy. In the endline study at both Yilkpene and Kpachilo, first, 
all the foods community members reported as their preferences were the same as those 
reported at baseline. Second, although the reasons mentioned at baseline for liking 
specific foods were retained, Yilkpene reported additional reasons. The reasons include 
the health/nutritive benefits of the foods such as the foods inducing sweating, giving 
energy/strength and building the body, protection from certain diseases and curative 
potency. Improvement in male sexual potency was an additional reason mentioned 
exclusively by male participants. 
 
All the foods cited at baseline as less preferred by some individuals were reaffirmed at 
Kpachilo at endline; however, two food items mentioned in the baseline that were also 
cited at Endline at Yilkpene as less preferred, particularly by some men, were soya bean 
and bouillon cube. Additionally, shea-fruit and the leaves and fruits of a particular 
baobab tree were mentioned as being disliked by some community members at Yilkpene. 
Participants at endline at both Kpachilo and Yilkpene, however, stated that they disliked 
 158 
University of Ghana http://ugspace.ug.edu.gh
food crops treated with chemicals that facilitated their quick ripening and made 
particular reference to tomato. The endline results in respect of food preferences are 
illustrated in Figure 4.2. 
 Yilkpene       Kpachilo 
 
 
 
Shea fruit, and fruit Meals 
 
and leaves from a prepared 
Orange-flesh 
particular baobab from soya 
 sweet potato, 
tree bean, 
meals prepared 
 bouillon from cassava 
cubes and 
and pigeon-pea 
 chemically- 
treated foods 
 
 
 
 
                  
Figure 4.2: Foods less preferred by Yilkpene and Kpachilo at endline 
 
At endline, there was consensus among all categories of participants at Yilkpene that 
women’s diets were inadequate in terms of quality and variety in their meals and linked 
the poor dietary situation to food insecurity. Male participants lamented that even though 
they were aware of the inadequacy of household and women’s diets, their ability to 
improve the situation was limited:  
“We’re aware that the women’s food is poor in quality but that’s all we have. 
The food we have is TZ and that’s all (…). We’ve all agreed that our diet and not 
just that of women isn’t the best so we’re trying our best but we’re limited as to 
what to do. We plant our crops and the rains aren’t that good so it’s not easy. 
 159 
University of Ghana http://ugspace.ug.edu.gh
We’re doing our best just to improve our women’s diets” (Post-intervention FGD 
men Y31; Category 2). 
 
At Kpachilo, however, divergence still persisted regarding the quality of women’s diets.  
While some participants agreed that women’s dietary situation was poor, others believed 
it was adequate in terms of quality as expressed at baseline. 
 
Participants at Yilkpene identified the harmful effects of women’s poor dietary status on 
women and unborn babies to include passing of bloody stools by women, anaemia in 
pregnancy (and the associated delayed delivery) and need for blood transfusion for 
mothers after delivery. Other harmful effects reported were birth of very small (under-
weight) babies poor health status and insufficient nutrients for the children. One male 
participant’s opinion was as follows:  
“Because the women’s diet is poor, they tend to eat falinkasa (pica), which makes 
their stools bloody and makes them unhealthy (…) Poor food for the pregnant 
woman can lead to anaemia, which isn’t good because it can delay birth of the 
baby. Sometimes, the children are born very small or very sick” (Post-intervention 
FGD men Y31; Category 2). 
 
A female key informant made a more elaborate assessment of the harmful effects of 
women’s poor dietary status on women and their unborn babies:   
“It’s important for women’s diets to contain a lot of different foods that we produce 
because they work a lot without rest and all this requires energy. The foods that we 
produce are very healthy and have specific nutrients the body requires so if the 
woman’s diets don’t contain all these foods, which is the case most of the time, then 
she won’t be healthy and as she carries pregnancies and breastfeeds as well, her 
unborn and lactating children will suffer from deficiencies of nutrients”. (Post-
intervention KII woman, Y23). 
 
Potential harmful effects of women’s poor dietary status on the community were also 
identified by endline FGD participants at Yilkpene. These effects include population 
 160 
University of Ghana http://ugspace.ug.edu.gh
reduction through maternal and child mortality, expenses on health care for unhealthy 
mothers at the expense of naming ceremonies and reduced labour force for farming 
because some members of a household have to sacrifice working on the farm so as to 
take care of an unhealthy mother: 
“Also, if a women is delivered and she isn’t well as a result of poor diets, you’ll 
have to spend some money to treat her, which would’ve been used to do 
something else or do the naming ceremony. As for the effects on the community, 
they are many. We’re in the harvesting season; so if a woman isn’t well, it affects 
the entire household because someone will have to take care of the sick woman. 
Also, the little money that you have is used to treat the woman”(Post-intervention 
FGD men Y31; Category 1). 
 
The male participants further cited another potential harmful potential effect of women’s 
poor dietary quality on their community as neighbouring communities showing contempt 
towards the community for their inability to care adequately for their women: 
“The other communities around us won’t have respect for us and even 
sometimes, if we go to those communities to marry, they might refuse us ”(Post-
intervention FGD men Y31; Category 1). 
 
4.3.1.3 Modifications to plant-based food restrictions at Yilkpene 
At Endline, there were no differences in perceptions about general community food 
taboos in both communities. Two additional taboo foods identified at Kpachilo were 
snail and “zanaza TZ”. “Zanaza TZ” is prepared specially for funerals but it is a taboo 
food for both males and females whose parents are alive. At Yilkpene, gorilla and 
monkey meat were identified as additional Islamic taboo foods.  
 
At Kpachilo, all food restrictions in respect of plant-based foods reported at baseline for 
pregnant and lactating women were mentioned again at endline. At Yilkpene, only egg, 
wild fowl, guinea-fowl and dawadawa fruit were mentioned.  Participants at Yilkpene 
did not report practising some of the restrictions particularly, slimy leafy vegetables for 
 161 
University of Ghana http://ugspace.ug.edu.gh
pregnant and lactating women because there was no evidence of the perceived negative 
effects of the restricted foods. Additionally, the medical interventions put in place to heal 
their wombs after child-birth at the health facility, coupled with the messages recalled 
during the intervention, account for not practising the restrictions for lactating women. A 
female participant explained why some of the plant-based food restrictions were not 
being adhered to: 
“This is because even if you tell a pregnant or lactating mother not to eat a 
particular food she won’t listen to you because when she eats it nothing happens 
to her. I also think it’s because all these beliefs were adhered to back then 
because our mothers used not to go to the hospital when they were pregnant like 
we do today. Once you’re pregnant and you start attending ANC you’re given 
medications until you give birth and even after that you’re given medications to 
take for some time. These medications can heal all the sores in your stomach 
after delivery making it possible for you to eat anything without encountering any 
challenges”(Post-intervention intervention FGD women Y30; Category 1). 
 
Endline participants in both communities attributed adherence to taboos related to 
animal-source foods targeted specifically at WRA to two main reasons. First, respect for 
tradition and, second, superstition - that is, fear of giving birth to a baby resembling the 
specified creatures:  
“As for the python and cat, women don’t still consume them because that’s been 
forbidden since the time of our great-grant-parents (...) and since I was born, I 
haven’t eaten the flesh of a python or cat because I don’t want to give birth to 
either” (Post-intervention FGD women, Y28; Category 2). 
 
At endline, participants at Kpachilo perceived food taboos as having no negative effect 
on women’s diets. At Yilkpene, however, the consensus among participants was that 
food restrictions affecting pregnant and lactating women did affect the quality of their 
diets by limiting their choices of a variety of leafy-vegetable in their diets. Female FGD 
participants in particular were of the view that most animals-source foods women were 
restricted from consuming were currently rare in their diets and hence, did not 
 162 
University of Ghana http://ugspace.ug.edu.gh
compromise the quality of their diets; however, restricting lactating women in particular 
from consuming some categories of vegetables may have some effect on their dietary 
quality:  
“Not these kinds of animals like the python. They aren’t even common in the 
communities and so we don’t add them to our everyday meals. So avoiding them 
doesn’t have any effect on the quality of our diets and the bush meat too. But 
speaking of plant-based foods like vegetables, yes, the lactating mothers may be 
denied their benefits for just a short period”(Post-intervention FGD women, 
Y28; Category 2). 
 
4.3.1.4 Changes in women’s dietary practices 
At endline, the context, pattern, types and frequencies of meals consumed by women did 
not change in both communities. Women’s diets continued to be dominated by grain-
based foods (mainly TZ and porridges) and accompanied by some leafy vegetable soups 
with anchovies. Groundnut was also common in their diets. Beverages were, again, 
mentioned as commonly consumed by some women but meat and fruits were not 
consumed frequently. The pattern of meals was as reported earlier: beverages and a 
variety of porridges in the early morning, followed by “nagban suli” during mid-
morning, then afternoon and evening meals. Changes were, however, observed in the 
frequencies of their meals between the two seasons (that is, May-June and November) 
and proportion of women achieving minimum dietary diversity (MDD-W). The 
frequency of meals is presented in Table 4.10. 
 
 163 
University of Ghana http://ugspace.ug.edu.gh
Table 4. 10: Frequency of meals at endline (May-June and November, 2017) 
Freq. of Yilkpene Kpachilo 
meals Baseline   End-line Baseline End-line 
May/Jun Nov. May/June Nov. May/June Nov. May/June Nov. 
e (N=20) (N=20) (N=20) (N=20) (N=20) (N=20) (N=20) (N=20) 
Less than 0 0 0 0 2 (10%) 0 0 0 
three 
Three 3 (15%) 8 (40%) 3 (15%) 9 (45%) 6 (30%) 9 (45%) 8 (40%)  13 (65%) 
Four or 17 (85%) 12 (60%) 17 (85%) 11 (55%) 12 (60%) 11 (55%) 12 (60%) 7 (35%) 
more 
Mean 4.01 3.65 4.1 3.7 3.6 3.75 3.7 3.35 
 
The observation in both communities was that more women consumed four or more 
meals in the periods of May-June, when there is shortage of food, than in November, 
when food is perceived to be in abundance for both baseline and endline phases. At 
endline at Yilkpene, there was no change in the proportion of women (85%) who 
consumed four or more meals in May-June. Kpachilo too did not experience any change 
in the proportion of women consuming four or more meals during the same period 
(60%). Not much difference was observed in the mean number of meals consumed at 
baseline and endline in both communities. The mean number of meals consumed at 
baseline at Yilkpene in May-June was 4.01 but it was 4.1 at endline for the same period. 
At Kpachilo, the mean number of meals at baseline was 3.6 and 3.7 at endline for the 
same period. 
 
Presented in Table 4.11 is the comparative data on the Minimum Dietary Diversity of 
women at Yilkpene and Kpachilo at baseline and endlines. 
 164 
University of Ghana http://ugspace.ug.edu.gh
Table 4. 11: Minimum Dietary Diversity of women at baseline and endline 
MDD-W Yilkpene Kpachilo 
Baseline   End-line Baseline End-line 
May/June November May/June November May/June November May/June November 
(N=20) (N=20) (N=20) (N=20) (N=20) (N=20) (N=20) (N=20) 
Yes 11 (55%) 13 (65%) 15 (75%) 15 (75%) 11 (55%) 16 (80%) 18 (90%) 10 (50%) 
No 9 (45%) 7 (35%) 5 (25%) 5 (25%) 9 (45%) 4 (20%) 2 (10%) 10 (50%) 
Average 4.45 5.1 5.2 4.95 4.6 5.1 5.45 4.5 
DDS 
Note: MDD- W –Minimum Dietary Diversity of Women (foods from five and more groups consumed).  
Yes- Number or percentage of women scoring five or more food groups (or meeting the minimum dietary 
diversity score. 
 No -Number or percentage of women scoring less than 5 food groups (or not meeting the minimum 
dietary diversity score).  
Average DDS – Mean Dietary Diversity Score of women surveyed 
 
Data presented in Table 4.11 shows a general improvement in the achievement of the 
minimum dietary diversity of women. The data indicate that at endline, three-quarters 
(75%) of the women at Yilkpene achieved the MDD-W in May-June compared to the 
slightly more that one-half (55%) of women who achieved it at baseline. Similarly, nine-
tenths (90%) of women at Kpachilo achieved the MDD-W at endline in the same period 
compared to its baseline proportion of slightly more that one-half (55%). On the 
contrary, while similar improvements in women’s minimum dietary diversity at endline 
at Yilkpene were reported in November (75%) compared to baseline (65%) Kpachilo 
reported a decline from four-fifths (80%) at baseline to one-half (50%) at endline for the 
same period.  
 
 
 165 
University of Ghana http://ugspace.ug.edu.gh
Consumption of micronutrient rich foods in Yilkpene: Apart from groundnut, iron- and 
vitamin A-rich foods were not frequently consumed in both communities as evidenced in 
the 24-hour dietary recall. The frequency of consumption of groundnut was generally 
high in November. Figures 4.3 demonstrate the frequency of consumption of groundnut 
by women in the two communities.  
 
Figure 4.3: Consumption of groundnut 
 
 
The consumption of legumes such as cow- and pigeon-peas and soya-bean was generally 
low at both baseline and endline in both communities. Even though consumption of 
soya-bean in particular was promoted during the intervention at Yilkpene, none of the 20 
women interviewed in the 24-hour recall indicated consuming any. Figure 4.4 shows the 
consumption of cow- and pigeon-pea as well as soya-bean at base and endlines in the 
two communities. 
 166 
University of Ghana http://ugspace.ug.edu.gh
 
Figure 4.4: Consumption of legumes 
 
 
Anchovies were the commonest animal-source food in the dishes of all households at 
both baseline and endlines in both communities. On the contrary, meat, other species of 
fish, egg and cow milk were less frequently consumed as shown in Figure 4.5. 
 
Figure 4.5: Consumption of animal-source proteins 
 
 
The consumption of green-leafy vegetables was common even though it is affected by 
seasonality. Consumption tended to be generally high in May-June in both communities 
 167 
University of Ghana http://ugspace.ug.edu.gh
but low in November. Dry green vegetables such as kuka (baobab leaves) and okro were 
consumed throughout the year. Figure 4.6 presents the frequencies of consumption of 
fresh green-leafy vegetables. 
 
Figure 4.6: Consumption of fresh green-leafy vegetables 
 
Similarly, the consumption of fruits – that is, wild fruits such as shea-fruits, dawadawa 
and ebony (“gaya”) and those externally sourced such as orange and banana - was 
reported to be low.  At endline at Kpachilo, there were only two reported cases of 
consumption of banana and orange and 13 cases of consumption of shea-fruit compared 
with only three cases of consumption of ebony at baseline. Yilkpene, on the other hand, 
reported only three cases of consumption of shea-fruit at endline compared with two 
reported cases of consumption of mango and shea-fruit at baseline. 
 
4.3.2 Participants’ perceptions of intervention  
Focus group discussion participants and key informants at Yilkpene evaluated five issues 
related to the intervention. The issues were messages recalled, messages practised, 
 168 
University of Ghana http://ugspace.ug.edu.gh
messages not practised, benefits of intervention and challenges related to the intervention 
activities.  
 
4.3.2.1 Messages recalled and practised at intervention meetings 
Table 4.12 shows a matrix of participant categories, messages recalled and messages 
practised. 
 169 
University of Ghana http://ugspace.ug.edu.gh
Table 4. 12: Matrix of messages recalled and practised  
Category of Messages recalled Messages practised 
participants 
Category 1(intense participation: 8 - 12 times at group meetings) 
Men -Importance of pregnant women’s diets to their  Sanitation: keeping compound and 
reproductive health outcomes surroundings clean 
  
-Importance of proper procession, preservation and Incorporating soya-bean in 
preparation of food- storage of beans and grains household meals  
 
Women -Good environmental sanitation and hygienic cooking Proper disposal of human waste by 
practices burying it in the ground 
-Nutritive benefits of vegetables and fruits Consuming cassava leaves  
-Consuming a variety of foods in meals (groundnut Drying moringa leaves  
added to koko, preparation of soya with yama  Cooking TZ with bra water  
-Dry-season home-gardening   
-Adolescent girls’ diets: Allowing them to eat more 
food especially animal-source foods and legumes 
-Food preservation and preparation practices 
(fermenting maize flour for TZ)  
-Drying and storage of green vegetables in rainy season 
Category 2( moderately-intense participation: 6-7 times at group meetings) 
Men  -Nutritive values of soya, grain, beans, vegetables  Using soya to prepare foods 
-Fermentation of grains 
-Adolescent girls’ diets are important 
-Environmental sanitation  
-Sharing food according to need (meat for children)  
Women -Consuming a variety of foods (preparation of soup Adding soya flour to groundnut 
with bra, eating groundnuts with koko, preparation of paste to prepare bra soup  
koko with soya, adding fish and dawadawa to Adding soya to koko  
vegetables) Kitchen hygiene and domestic 
-Fermentation of grains  sanitation  
-Environmental sanitation, personal hygiene Washing cooking pots  
-Nutritive values of fruits 
Category 3 (low-intensity participation: 1 -5 times at group meetings) 
Men -Nutritive values of beans, rice, leafy vegetables, fruits  Domestic sanitation  
-Environmental sanitation  Consumption of egg and meat  
-Preparation of foods with soya bean Soaking maize for grinding in next 
-Fermentation of grains community  
 
 
 
Women -Benefits of soya, beans, leafy vegetables, Soaking maize for grinding in next 
-Combining a variety of foods in meals (soya bean in community  
koko, “gable” and TZ)  Preparation of leafy vegetable 
-Using water used to cook bra for TZ  soups/ kuka soup with palm oil  
-Drying bra, moringa, cassava leaves, etc., in the shade Using water used to cook bra for 
not the sun  TZ 
Community -Cultivation of a variety of foods  Sanitation  
key informants -Adding groundnut to koko  Helping wives to care for children 
(leaders) -Nutritive values of banana and other fruits, sweet Healthy diets for children  
potato, vegetables  Eating groundnut with other foods  
-Healthy diets for children  Cultivation of groundnut  
-Environmental sanitation Cultivation of maize 
-WRA’s dietary status important  
Note: Category 1: participation in or exposure to 8 to 12 group meetings  
Category 2: Participation in or exposure to 6 to 7  
 170 
University of Ghana http://ugspace.ug.edu.gh
Category 3: Participation in or exposure to up to 5 group meetings 
 
The data presented in Table 4.11 shows an aggregate of nine (9) messages were recalled 
by participants in Category 1 (participation or exposure to 8 – 12 group meetings) 
compared to 6 and 7 by Categories 2 (exposure to 6-7 group meetings) and 3 (exposure 
to up to 5 meetings) respectively. Women recalled more messages (15) than men (11). 
The messages recalled exclusively by men and women and those common to them are 
presented in Figure 4.7. 
Men     Women 
 
 
 
1. Drying and storing of 
 1. Domestic green leafy vegetables, 
1. Diets for pregnant 
environmental 2. Growing vegetables in 
 women 
sanitation home gardens,  
2. Proper storage of 
 3. Shade drying of 
harvested grains,  2. Nutritive benefits 
of vegetables and vegetables,   3. Importance of 
4. Using stock of 
adolescent girls’ diets,  fruits  
 vegetables to make T.Z, 
4. Meat for children 3. Combining a 5. Fermenting grains, 
 variety of foods such 
6. Adding groundnut and 
as groundnut and 
 soya to koko and TZ  
soya bean in diets 
7. Female personal 
   hygiene  
 
 
 
Figure 4.7: Messages recalled by men and women from intervention 
 
Key Message 4 - “empowering women to have control over food resources”- was the 
only message that was not recalled by any of the participants. Additionally, key Message 
 171 
University of Ghana http://ugspace.ug.edu.gh
3- “modification of unfounded beliefs about food” was recalled only in relation to the 
sharing of food in favour of children. 
 
4.3.2.2 Messages practised by participants 
Categories 1 and 3 participants reported the highest frequency of messages practised 
(n=5 each) but Category 2 reported the least. Key informants had practised six messages. 
Domestic and environmental sanitation (including disposal of human waste) was the key 
message practised by all three categories of participants as well as key informants. Some 
women indicated increased consumption of some foods like leafy vegetables, which they 
were not eating in the past. Additionally, they reported adopting drying of their leafy 
vegetables without using direct sunlight and using the stalk from boiled vegetables to 
enhance the quality of their TZ: 
“Madam, have you seen cassava leaves? We used not to consume them because 
we didn’t know how healthy they were but through those Saturday meetings with 
Madam and her team, we’ve come to understand that they’re very healthy and 
now we consume them (…) Through those meetings, we also learnt that if we 
should dry our moringa and other leaves, we should do so in the shade and not in 
the sun so that they don’t lose their nutritive values (…) Before Madam and her 
team came to enlighten us, we used to throw away our bra stalk but we learnt 
from them that we don’t have to throw the stalk away since it also contains some 
nutrients and that we could stir our TZ with it again and that’s what we do now 
and when we do that, the TZ can even be kept up to 2 days without going bad” 
(Post-intervention FGD Y30 women; Category 1)). 
 
More women than men reported practising messages heard- 8 compared with 4. The 
main messages practised by both men and women were domestic and environmental 
sanitation, fortifying dishes with soya flour and consuming TZ prepared with fermented 
maize flour. Only men reported practising the consumption of egg and meat. The 
messages practised exclusively by women included drying of leafy vegetables, cooking 
with vegetable stalk and observing good environmental sanitation during cooking:  
 172 
University of Ghana http://ugspace.ug.edu.gh
“For me, when I want to prepare bra soup, I add soya flour to groundnut paste 
and even koko I add it to prepare it. I always add flour on the night before 
making the dough and, on the next day, when I prepare it, the koko tastes like 
yellow corn and my family like it a lot” (Post-intervention FGD Y28, women; 
Category 2). 
 
Only key informants had been able to practise supporting their wives to care for their 
children and ensuring that the children consumed healthy diets. The only message 
practised by the community at large was environmental sanitation. The unanimous 
response of the three categories of participants was confirmed by all the key informants. 
 
Both male and female participants in Category 3 identified consumption of fermented 
maize as the message they had not been able to practise owing to objection by the only 
local miller to the grinding of fermented maize:  
“We tried to soak the maize before grinding but the miller will not accept it. Now 
we’re engaged in farming activities else we could send our soaked maize to the 
nearby communities for grinding” (Post-intervention intervention FGD Y33; 
Category 3). 
 
In an exceptional case, a female participant added that she had been unable to practise 
the use of soya in her dishes owing to objection from her landlord to the alleged odour it 
creates: 
“For me, I tried to use soybeans to prepare soup but my landlord doesn’t like it 
so I had to stop” (Post intervention FGD Y33; Category 3). 
 
4.3.2.3 Benefits of intervention to participants 
Female participants indicated that the benefits of the intervention to them were washing 
of drinking pots, disposal of children’s faeces, cleaning of vegetables before cooking, 
consumption of more vegetables, increased consumption of fish and consumption of TZ 
 173 
University of Ghana http://ugspace.ug.edu.gh
prepared with fermented maize. Female participants in Category 1 labelled the benefits 
numerous: 
“As for the benefits, they’re numerous. We can’t quantify them. Imagine you 
don’t know a person and the person skips several communities to come to your 
community just to talk to you about your own health and well-being. Now, we 
wash our drinking pots regularly because we know that it’s healthy for us” (Post-
intervention FGD Y30 women; Category 1). 
 
According to female participants in Category 2, the benefits included enhanced food 
safety as well as consumption of fish in contrast with reduced morbidity: 
“We’ve benefited because now, when we bring the vegetables from the farm, we 
wash them but we wouldn’t do that first as we thought they were clean from the 
farm. I used to use GH¢ 3.00 worth of fish for the whole week but now I’ve 
increased it to GH¢ 5.00 because I want my family to be healthy. This year, there 
hasn’t been a lot of going to hospital and that’s because of Madam (that is to say, 
the intervention)” (Post intervention FGD Y28, women; Category 2). 
 
Category I participants identified 7 specific benefits of the intervention to the community 
at large. Categories 2 and 3 participants identified 3 and 2 benefits respectively. The only 
unanimous benefit identified by the three categories and key informants was 
environmental sanitation. The female participants in Category 1 expressed the 
community’s unreserved appreciation of the communal benefits of the programme and 
specifically referred to the knowledge about the kuka tree obtained through the 
intervention messages: 
 “We know that the kuka tree is one of the few trees around that have no 
chemicals. The farmers have pumped all the other vegetables with chemicals 
which can be poisonous. We got to know from the intervention that the kuka 
leaves are very nutritious. Because of that, we’re so much enthused to prepare 
kuka soup because we know it’s very healthy for us” (Post-intervention FGD 
Y30, women; Category 1). 
 
All the male participants in Category 1 confirmed that the intervention was immensely 
beneficial to the community. They particularly perceived messages relating to the health 
 174 
University of Ghana http://ugspace.ug.edu.gh
of pregnant women and children, exposure to the nutritive benefits of locally-produced 
foods and improvements in dietary practices to be beneficial. They could not but express 
open appreciation of the intervention: 
“We were very pleased and even now we are. If Madam would agree, we’d even 
build a room for her around the school so that she’d always be with us to give us 
more knowledge and counsel. We also wish that the Saturday meetings will 
continue” (Post-intervention FGD Y29, men (8-12). 
  
4.3.2.4 Issues associated with intervention messages  
Only men in Categories 1 and 3 in addition to key informants mentioned issues 
associated with some of the intervention messages they did not appreciate. The sole 
challenge stated by men in Category 1 related to misgivings about the call for women to 
have direct access to grain barns. Their misgivings were communicated through the 
following quotations: 
“There was one thing that the intervention messages carried which I didn’t want. 
The message said that, since men are those doing farming and the women are 
those doing cooking, we should always give the grains or the major food items to 
them so that they’ll use them to cook (...). We don’t want it that way: a woman 
doesn’t know how to keep items (…) so if you give her the bags of maize, it 
wouldn’t even get to 3 months and she’ll finish up everything” (Post-intervention 
FGD Y29, men; Category 1). 
 
On the contrary, the sole challenge revealed by the men in Category 3 pertained to 
misgivings about the calls to give women and children rather than men better portions of 
chicken. Their misgivings about the calls were stated as follows: 
“The one thing I've difficulty with is the issue about we men giving what an old 
person should take to a child. Giving men’s shares to children is my biggest 
problem. For instance, giving the children the thighs and I eating the head or the 
legs (…). In our culture, if an old man should give what he’s supposed to take to 
children and rather take the children’s share, he won’t be respected”(Post-
intervention FGD Y32, men Category 3). 
 
 175 
University of Ghana http://ugspace.ug.edu.gh
The only challenge reported by key informants related to community members’ 
expectations of monetary incentives for their participation in the intervention. The 
following quotation relates to the challenge: 
“Well, for me, I’ll say that all that was done was perfect. What I would 
recommend is the issue of money but there’s a saying that if you teach someone 
how to fish, it’s better than feeding him with fish all the time. Some of the 
community members were expecting something. Maybe, they’re thinking that 
money would be given to them to invest in agriculture or any other form of 
development that wasn’t part of the intervention”(Post-intervention KII Y26). 
 
4.4 Summaries of results 
Presented in Tables 4.13a, 4.13b, 4.13c, 4.13d, 4.13e and 4.13f are the summaries of 
results at baseline and endline. The summaries are presented in line with the objectives 
of the study. 
Table 4. 13a: Summary of results –objective 1- baseline and endline 
Objective  Indicators Baseline results for both Endline 
communities Yilkpene (intervention Kpachilo 
community) (comparison 
community) 
1. The Access to -Women had limited access to land, Community sensitized about No change 
food productive and labour, credit and household food need for women to be more 
system household food resources. empowered in terms of access 
and resources and control over food 
gender resources. 
issues Foods produced -Men dominate in producing for Sensitized about the need to No change 
and use home consumption and cash. begin cultivating vegetables 
-Women dominate in producing for household to improve 
vegetables usually from the wilds for supply all year round 
home use. 
-Men control income from household 
produce. 
Food Inadequate preservation and storage Sensitized on proper  
processing, facilities for women in particular techniques of preservation and 
preservation and storage of food 
storage  
Distribution and -Women dominate the sale of crops No change  
intra-household but price are determined by both. 
allocation of -Both men and women contribute to 
food household food needs. 
Preferential Women are disadvantaged in food Sensitized about need to share No change 
food allocation sharing. food according to need 
system 
 
 176 
University of Ghana http://ugspace.ug.edu.gh
Table 4.13b: Summary of results- Objective 2 -baseline and endline   
Objective Indicators Baseline results for both Endline 
communities Yilkpene (intervention Kpachilo 
community) (comparison 
community) 
2. Dietary Knowledge of Most participants could not Participants could differentiate No change 
knowledge the 3 food differentiate between the three food between the three food groups 
and groups groups 
attitudes 
(perceptio
ns and 
beliefs) 
 Knowledge of Aware of good foods for women Improved knowledge by No change 
health- particularly pregnant women adding a wider variety of 
enhancing foods foods 
for women 
 Sources of Health facilitates, the home, tradition Intervention was the No change 
dietary and NGOs additional source 
knowledge 
 Perceptions -Meals containing dawadawa, fish, Addition: Food safety and not No change 
about healthy meat, beans over-doing vegetables in 
foods -Foods that enhance health: sweating, particular 
drinking water, giving blood and 
energy 
-Hygienic food 
-Recommended by health worker 
 Food Soya bean, sweet potato, pigeon pea Only foods disliked were No change 
preferences and cassava-based foods disliked by soya-bean and bouillon cubes 
both men and adolescent girls by most men  
Bouillon cubes disliked by men 
 Perception of Divergent views on quality of Participants perceived No change 
adequacy of women’s diets: women’s diets to be lacking in 
women’s diets Some women felt their foods were quality 
adequate in terms of filling and 
keeping healthy 
Fewer women felt their foods were 
inadequate in terms of quality and 
variety 
 Perception of Problem of effects of women’s diets Problem of effects of No change 
the effects of not apparent  women’s diets more apparent:  
poor diets on Effects were associated with Affects unborn child’s health 
women’s health women’s physical and psychological and causes anaemia in mother, 
ability of women and not their difficult delivery  
reproductive health 
 Adherence to Strict adherence to food taboos Restrictions on consumption No change 
food taboos and affecting all WRAs  of plant-based foods modified 
restrictions  
affecting all 
WRA 
 Perceptions of Problem of food taboos affecting Effects of food taboos on No change 
the effects of quality of women’s diets not apparent  quality of women’s diets 
food taboos on apparent 
women’s diets 
 
 
 177 
University of Ghana http://ugspace.ug.edu.gh
Table 4.13c: Summary of results - objective 3- baseline and endline   
Objective Indicators Baseline results for both Endline 
communities Yilkpene (intervention Kpachilo 
community) (comparison 
community) 
3. Kinds of foods Mainly grains, some vegetables, No change No change 
Women’s consumed groundnut and anchovies 
dietary 
practices 
 Frequency of Majority consume 4 meals or more  No change No change 
meals 
 Preparation of -Most vegetables overdone -Vegetables not-overdone  No change 
food -Little use of anchovies  -Use of more anchovies in 
-Little use of vegetables  meals,  
-Do not add soya bean products to Add soya bean flour to meals 
foods 
 Minimum 55% in either community achieved 75% 90% 
dietary diversity MDD-W 
of women 
 
Table 4.13d: Summary of results –objective 4 
Objective Drivers 
4. Drivers of women’s dietary choices Food security in households 
 Food taboos and restrictions 
 Women’s physiological status 
 Perceived healthiness of foods 
 Large sizes of households 
 Sources of information on food 
 Tastes of foods 
 
Table 4.13e: Summary of results –objective 5 
Objective Readiness level of  Not repeated at end-line 
Yilkpene at baseline 
5. Community readiness to Low readiness score (3)   
modify socio-cultural 
barriers to women’s optimal 
diets in Yilkpene 
 
 178 
University of Ghana http://ugspace.ug.edu.gh
Table 4.13f: Summary of results – Community perceptions of intervention 
Indicators for process evaluation in only intervention community (Yilkpene) 
 Coverage Overall: 38%   
Men=33%  
Women=42% 
55 out of 56 household participated 
 Exposure Mean attendance =7  
 8-12 meetings  37.4%  
6-7 meetings 30.5% 
5-3 meetings 32.7% 
 Modifications to Extended period of intervention to  
intervention address bottle-necks and 
misconceptions 
Partici Messages recalled -Women recalled more (15) than men -Both: good sanitation practices, nutritive 
pants’ (11) than men benefits of vegetables and fruits and consuming 
percep - More massages recalled by a variety of foods 
tions of participants with more exposure  -Men: food preservation and storage, 
interve  importance of good diets for pregnant and 
ntion  adolescent girls 
Need to give children meat 
-Women: ensuring the availability of vegetables 
all year round through backyard gardening and 
proper storage techniques, improved food 
preparation methods and ensuring good hygiene 
among women 
 Messages not Empowering women to have more  
recalled control over food resources 
 
Modifications of unfounded beliefs 
about food 
 Messages More messages practised by women (8) -Men: consuming egg and meat 
practised than men (4) - Women: improved storage of vegetables, food 
preparation practices and observing good 
hygiene during cooking 
 Message not -Consumption of fermented grains  
practiced Not allowing women direct access to 
grains 
 Benefits of Category 1 identified more benefits of -Improved food hygiene, household and 
intervention the intervention (7) than Category 2 (3) community sanitation practices 
and 3 (2) -Improved consumption of vegetables, fish and 
fermented maize 
 
 Perceived Men:   
challenges -Uncomfortable about empowering 
associated with women to have more access to 
intervention household food resources 
Disagree with giving women and 
children better portions of meat 
Non-monetary rewards for community 
volunteers  
 
 179 
University of Ghana http://ugspace.ug.edu.gh
CHAPTER FIVE 
5.0 DISCUSSION OF FINDINGS 
The present study was a participatory action research aimed at identifying socio-cultural 
barriers to women’s diets and designing and implementing a culturally-appropriate 
attitudinal and behavioural change communication intervention addressing the identified 
barriers. The study adopted the social cognitive theory by Albert Bandura (1999). The 
social cognitive theory describes the reciprocal relationship between three constructs: 
cognitive/personal, behavioural and environmental factors. These factors can act 
individually but may also interact with one another to produce change. The theory was 
found to be relevant to this study and the factors have been integrated into the discussion 
along the objectives of the study.  
 
The pathway to attitudinal and behavioural change, which incorporated the social 
cognitive theory and community readiness assessment proposed for the study, was 
followed through as depicted in Figure 5.1 
 
 
 
 
 
 
 
 180 
University of Ghana http://ugspace.ug.edu.gh
 
 Pathway to attitudinal and behaviour change at Yilkpene 
 
Identification of Community Design and implementation 
 
individual-, readiness of intervention  
 household- and assessment  
community-level 
 drivers 
  Community-prioritized Interventions 
-Low access to food challenges by 
 resources consensus  Nutrition/dietary 
-Land and climatic 
education and 
 factors 1.Inadequate knowledge training on 
-Techniques of 
on nutritional qualities of preservation and 
 preservation and preparation of food 
preparation of food food and weak link 
 Advocacy on need 
-Poor dietary between women’s diets 
 for women to be 
knowledge and and reproductive health more empowered 
attitudes  
 to access more food -Food taboos and 2.Misperceptions about and modifying food 
restrictions healthy foods 
 taboos and -Low and irregular 
restrictions  
income 3.Women’s poor access to 
   and control over 
 productive and food 
resources 
 Poor access to adequate food 
4.Adherence to food 
resources 
 taboos and restrictions 
 Inadequate techniques of 5.Inadequate techniques 
 
preservation, storage and of food preservation, 
 preparation of food storage and preparat ioInm proved dietary knowledge 
and attitudes 
 Inadequate dietary quality 6. Poor environmental 
sanitation 
   Reduction in food restrictions 
 
 
 
 
 Improved techniques of food 
preparation 
Figure 5.2: Pathway of attitudinal and behaviour al change at Yilkpene 
  
The pathway to attitudinal and behaviour change fo r the study was followed through as 
shown in figure 5.1. The pathway shows the identified baseline findings of individual-, 
household- and community-level challenges to women’s diets during the exploratory 
 181 
Improved dietary quality 
University of Ghana http://ugspace.ug.edu.gh
phase of the study. This phase was followed by a community assessment of Yilkpene’s 
readiness for a socio-cultural intervention addressing the identified challenges. There 
was also a stakeholder consensus meeting to prioritize the identified problems. Six 
challenges were prioritized. This prioritization led to the design and implementation of 
an intervention (with two components) with active community participation. The results 
of the intervention were improved dietary knowledge and attitudes, modifications of 
plant-based food restrictions and improved proportion of women achieving minimum 
dietary diversity. 
 
5.1 Traditional norms and gender affect the local food system  
Gender-based barriers within the local food system were identified as posing challenges 
to women achieving optimal diets. The inability to acquire land, labour and other 
farming inputs; differences in the kinds of foods produced by men and women and their 
uses; and distribution and intra-household allocation, preparation and sharing of food 
were some of the gender-based issues identified as relevant barriers. 
 
Women’s limited access to land, labour and other farming inputs was identified as a 
challenge to producing a variety of crops needed to enhance their diets. Similar findings 
have been reported in sub-Saharan Africa and other parts of Ghana (Bambangi & 
Abubakari, 2013; Mohammed et al., 2013; Whitehead, 1999). FAO (2012) reported that 
land sizes in Ghana are generally small but women tend to have less access than men. 
 
Women’s inability to access land, labour and other farming inputs derives from the 
interplay of socio-cultural and physical environmental factors. Inheritance of land is 
restricted to male descendants mainly because married women belong to their husbands’ 
 182 
University of Ghana http://ugspace.ug.edu.gh
rather than their fathers’ households (Bambangi & Abubakari, 2013; Seebens, 2010). In 
essence, family land is distributed in a way that perpetuates control in men’s hands. 
 
Acquisition of land from other families is not a favourable option for women partly 
because they lack the money to hire land and partly because of lack of support from their 
husbands (Seebens, 2010). Natural and physical environmental factors such as climate 
change, environmental degradation, deforestation and bush burning may also contribute 
to women’s unfavourable access to land. The combined effect of the natural and 
environmental factors is a reduction in the acreage of arable land, especially the sizes 
available to women (Kuhnlein & Pelto, 2009). 
 
Even though women form an important part of the labour force for all farming activities 
- about 43% (FAO, 2012), they do not have rights to their husbands’ and other male 
household members’ labour (Seebens, 2010). Women traditionally need their husbands’ 
approval to engage in financial and land transactions with external parties (such as 
members of other families and NGOs). This practice limits women’s access to labour 
and farming inputs. 
 
The results call for women’s enhanced access to productive resources such as land 
through strategies such as co-operative farming and negotiations with land-owners and 
improved access to their husbands’ and other household members’ labour. There is also 
the need for the acquisition of farming inputs to be enhanced through more access to 
credit facilities for both men and women. 
 
 183 
University of Ghana http://ugspace.ug.edu.gh
The two communities studied were primarily subsistence farming communities where 
food production is a household activity involving men and women. While men produce 
more quantities of both subsistence and cash crops, women mainly produced for home 
consumption. A number of nutritious foods were currently less produced while leafy 
vegetables and fruits were harvested from the wilds by women. Even though livestock 
production was carried out by both genders, men took decisions for their care and sale. 
 
The findings agree with the report by FAO (2012) that Ghana is predominantly an 
agrarian economy and composed mainly of peasant farming. The report indicates that 
agriculture accounts for 82% of the total amount of land dedicated to farming activities 
and 80% of the country’s food production. The findings also support an earlier WFP 
(2012) report on the small sizes of farmland in Northern Region, where 62% of 
households own between 2 and 5 acres of farmland. The results further reflect women’s 
involvement in the various aspects of the food system (Mohammed et al., 2013 ; 
Seebens, 2010; Kuhnlein & Pelto, 2009). 
 
Differences in the quantities and kinds of foods produced by men and women have been 
reported in other countries (Seebens, 2010; Doss & Morris, 2001). The differences may 
be partly attributed to access to farmland, labour and capital as discussed earlier. Men 
typically produce larger quantities and varieties of foods for both domestic and 
commercial purposes. Women, on the other hand, rely on their husbands’ goodwill to 
access small portions of their land to grow mainly staple foods for home consumption 
(Bambangi & Abubakari, 2013). 
 
 184 
University of Ghana http://ugspace.ug.edu.gh
Labour plays a major part in women’s inability to grow certain categories of crops such 
as root crops and legumes except in the case of soya-bean, which women are supported 
by some external agencies to cultivate. Participants explained that these categories of 
food crops are labour intensive and women do not to have the requisite physical strength 
to cultivate them. Women’s inability to access their male household members’ labour 
coupled with their inability to hire labour limit their ability to cultivate large quantities 
and varieties of crops. 
 
Climate change has contributed to the lower scale of cultivation of some nutrient-dense 
foods in the communities (WFP, 2012). Foods such as cotton and melon seeds, yam and 
millet are perceived as nutritious food items in the communities. The production of these 
crops, has, however, been adversely affected by the poor patterns of rainfall and the 
acute water shortage in the communities, limiting the varieties of nutritious foods 
available to women. Similarly, foods harvested in the wilds, consisting mainly of green-
leafy vegetables and fruits, contribute to household and women’s dietary quality. 
According to participants, in the past, these foods were collected in relatively large 
quantities by women for both home consumption and sale but, owing to climate change 
and human activities, they are currently in short supply. 
 
Women’s inability to produce large and varied quantities of crops has implications for 
their dietary quality since they are traditionally required to supplement their husbands’ 
grain allocations with other foods needed to prepare meals for their households 
(Colecraft, 2012). The sale of most micronutrient-rich foods such as legumes and nuts 
cultivated by both men and women reduces household and women’s dietary quality. The 
male participants explained that, since the market values of such crops are high, they sell 
 185 
University of Ghana http://ugspace.ug.edu.gh
them to enable them to pay back loans used for their farming activities. Additionally, the 
attitude towards access to foods harvested from the wilds has negative implications for 
sustaining their availability. The result is a decline in the variety of foods rich in 
micronutrients available to women. 
 
Sale at farm gates to market women is necessitated by the unfavourable socio-economic 
conditions such as the low levels of women’s productivity, poor transport system and the 
challenge posed by the excessive demands made on women’s time by their multiple roles 
such as child care, cooking and fetching of water (Seebens, 2010). Women may find it 
more expedient and feasible to sell their produce to market women at rates far below the 
market values; however, such transactions perpetuate poverty among the women, which, 
in turn, deprives them of access to quality diets. 
 
Sale of foods at both farm gates and external markets requires the empowerment and 
mobilization of local women through producers’ or farming co-operatives and other 
groups. These measures would offer women better bargaining power and greater levels 
of financial security against shocks such as unfavourable climate and transport 
difficulties. There is also the need to empower women to add value to their produce 
through improved techniques of production, processing and preservation to raise their 
levels of incomes. 
 
Both men and women made provisions for household food needs. While men provided 
grains, women complemented the grains with soup ingredients and milling of the grains 
(Colecraft et al., 2012). The allocation of grains to women from the food barns has 
manifest and latent explanations. The manifest reason for the taboo against women 
 186 
University of Ghana http://ugspace.ug.edu.gh
entering the food barns is the belief that a woman who goes into any barn to take grains 
gives birth to children who grow to be thieves. 
 
On the contrary, the latent reason is linked with lack of trust, perceived mismanagement 
of resources and the need to keep track of stocks for future restocking. Men’s control of 
food barns proved to be a contentious issue among the women. According to those who 
support a male-only access, the practice relieves women of part of the burden to worry 
about grains when they are running out. Those who oppose this view felt that giving 
women direct access offers them better opportunity to have adequate access to food for 
the household. 
 
Women’s relatively-low incomes hinder their performance of their roles such as milling 
the grains allocated by the men and getting ingredients for soups and sauces, which, are 
more expensive than accessing grains. This deficiency leads to poor quality of diets for 
households and women. This challenge calls for access to credit facilities for women to 
engage in alternative income-generating activities to enable them to access other food 
items to supplement their household food allocations from their husbands. 
 
The current study found the practice of giving out grains as gifts a common practice 
which offered very poor women the opportunity to access food. Food gifts are made in 
obedience to one of the five pillars of Islam called Zakat (Hossain, 2012). Zakat is a 
form of worship in which relatively wealthier Moslems are obliged to share their wealth 
(or harvests in the context of this study) including food with the poor or needy in society. 
Hossain (2012) claimed that the concept of Zakat, as practised among Moslems, could 
serve as a tool for alleviating poverty and hunger in poor communities since the poor 
 187 
University of Ghana http://ugspace.ug.edu.gh
have the right to the wealth of the rich and the latter have an obligation towards the 
former. Donor goodwill in the study communities, however, depends on the levels of 
production, magnitude of overall demand by potential beneficiaries and their own 
domestic circumstances. 
 
Gender and age influence the division of roles for the preparation and sharing of food for 
the household. Even though preparation of food is primarily done by women, it is 
subjected to division of labour on the basis of age and order of marriage among the 
women in the household. Women generally serve the food they prepare but the serving 
of meat is done by men. The influence of gender, age and order of marriage in 
preparation and sharing of food and eating arrangements corroborates findings by 
Gittelsohn and Vastine (2003); Leroy et al. (2008), and Dickson et al. (2016). The 
findings also confirm the report by Kavle and Landry (2013) that in many lower- and 
middle-income countries, men and children eat before women are served. The finding 
implies that women’s access to lower quantities and less-nutritious components of foods, 
may affect their dietary intake and quality. 
 
Tradition around the performance of specific domestic chores such as the preparation of 
food as the responsibility of women has been reported (Seebens, 2010). Cooking for 
large households exerts a heavy toll on women financially, physically and 
psychologically -with potential adverse effects on their diets and health. The decision to 
prepare a particular type of food in a household is dependent on the preferences of men, 
who allocate food items to the women usually on a weekly basis. Older women, usually 
husbands’ mothers, are the other members of the household who make decisions 
regarding food consumption. 
 188 
University of Ghana http://ugspace.ug.edu.gh
Restriction of the preparation of food to raw materials allocated by men has some socio-
cultural connotations. It could reflect the imbalance in power and authority, which is 
traditionally skewed in favour of men in the patriarchal cultural system. Involvement of 
mothers and mothers-in-law in making household food decisions further depicts the 
influences of patriarchal elements. The elderly women’s power and authority are derived 
from their kinship ties with the male heads of households and so they act in place of their 
sons not daughters. 
 
Issues of power, authority, gender, age and the related traditional perceptions of value 
placed on contributions to the acquisition of food are at play in the food-sharing system 
and eating arrangements. It is, thus, allegiance to and respect for tradition that accounts 
for the preferential treatment of men and the aged at the expense of women and children 
respectively. The woman who prepared the food is the first to have her bowl served prior 
to the rest of the household being served (with the better portion). The manifest 
explanation was that she should have the top-most portion, which is usually not so good. 
The latent reason for that practice, however, seems to be linked to the traditional mutual 
mistrust and scepticism among co-wives in a polygamous household. The woman who 
prepared the food is, thus, required to have the first taste of it ostensibly to assure the 
other women of its safety. Additionally, eating after everybody else is satisfied deprives 
the women of the benefit of eating to their satisfaction. 
 
Men traditionally serve the meat because, first, it is perceived as a sign of respect for the 
men who provided it. The second reason is to ensure fairness to other members of the 
household since there are typically several wives and any woman sharing the meat might 
be tempted to favour her children. Third, women are believed to be ignorant of the 
 189 
University of Ghana http://ugspace.ug.edu.gh
specific parts of the animals that should go to the various members of their households. 
Women are given the bony parts of an animal such as the back of a chicken or a 
ruminant ostensibly to enhance the development of strong bones to facilitate their 
carrying of pregnancy and going through child-birth. 
 
The coping strategies adopted by communities and women to mitigate the effects of food 
shortage have been reported in other parts of northern Ghana (Saaka, 2016; Armar-
Klemesu et al., 2015). The two studies reported reductions in the sizes of women’s 
portions of food and resorting to preferential food allocation, in which women go 
without food to ensure that other members of the household, especially children, get 
enough food to eat. Purchase of food through the sale of livestock and other items is a 
common practice among most households in Ghana (Armar-Klemesu et al., 2015). The 
coping strategies adopted by women attest to their contribution to all aspects of food 
security in Ghana: availability, accessibility and utilization of food (Mohammed et al., 
2013). 
 
Sale of households’ livestock, other foods and assets is traditionally done by men as is 
the disposal of income earned from any sale. Women have no access to such incomes. 
Younger women often migrate to southern Ghana during periods of low-farming 
activities to earn income as casual labourers (Abu, Louw, Raubenheimer, & Berg, 2014 ) 
and the incomes earned are usually controlled by women and used to buy household 
needs including food to supplement what the men allocate to them. This finding supports 
the claim by Duflo and Udry (2004) that women spend their incomes on household food 
needs and, particularly, on their children’s. The findings, does not, however, fully 
support the claim by Duflo and Udry (2004) that men spend their money on personal 
 190 
University of Ghana http://ugspace.ug.edu.gh
needs (Duflo & Udry, 2004). The findings reveal that men spend their incomes on 
grains, children’s education, festivals and acquiring additional wives. 
 
Some of the coping strategies adopted by households, and women in particular, to 
mitigate the effects of food shortage are likely to contribute to women’s poor dietary 
status. Reductions in grain allocations and sizes of women’s portions of food, for 
instance, are likely to reduce women’s food intake.  
 
5.2 Community members’ dietary knowledge and attitudes have implications for 
quality of women’s diets 
A fundamental finding in this study is that community members’ dietary knowledge and 
attitudes are important determinants of the quality of women’s diets. Diet is an integral 
part of culture and knowledge on it generally forms part of the socialization process. Diet 
is also an integral part of oral tradition that is passed on from one generation to the next. 
Health workers also provide dietary information to women during ante-natal and post-
natal clinics.  
 
The finding on sources of women’s dietary information identified by participants 
corroborate the report by de-Graft Aikins (2014) that women’s nutrition and dietary 
knowledge comes from multiple sources including “lifeworlds” and the health system. 
The role of family members, particularly mothers, mothers-in-law and friends, as trusted 
sources of dietary and health information, which is passed on from one generation to the 
next, has been reported in other low- and middle-income countries (Kavle & Landry, 
2018). The sources of information attested to the influence of informal and non-formal 
forms of community education, especially in rural communities. Access to information 
 191 
University of Ghana http://ugspace.ug.edu.gh
through the school and television to the group of adolescent girls pointed to the potential 
effectiveness of a segmented-target population approach to the implementation of a 
nutrition and dietary intervention.   
 
Participants were of the opinion that the foods they consumed enhanced their health. The 
finding on women’s dietary knowledge corroborates findings from a report by de-Graft 
Aikins (2014) that indicated that women’s lay knowledge of nutrition was not too 
different from that of experts or medical nutrition knowledge on the basis of the foods 
being generally healthy. In this study, however, participants at baseline could not 
differentiate between high-nutrient-density foods and low-nutrient-density foods. The 
apparent factors included the influence of tradition, lack of exposure to a sustained 
intervention and their low levels of formal training and education on diets (Amugsi et al., 
2016; Dunneram & Jeewon, 2015; FAO, 2012a; WFP, 2005). Participants’ lack of 
adequate knowledge on the nutritive qualities of different varieties of foods may prevent 
them from ensuring adequate balance in the varieties of foods contained in their diets, 
thus, limiting their dietary quality. 
 
Participants also had some misconceptions about what healthy foods were at baseline. 
For instance, they labelled groundnut an unhealthy food, particularly for pregnant 
women, for two main reasons. First, it was perceived as a “cold food” - that is, food that 
made the foetus lazy. Second, groundnut was associated with diarrhoea. Pregnant women 
were, therefore, prohibited from eating groundnut-based foods though it is essentially 
one of the main sources of protein and fat in the communities. Groundnut contains 47% 
fat and 38% protein (Atasie, Akinhanmi, & Ojiodu, 2009). Groundnut has also been 
reported to be forbidden for women during pregnancy among the Akwapims of Eastern 
 192 
University of Ghana http://ugspace.ug.edu.gh
Region of Ghana (de-Graft Aikins, 2014). In Nigeria, groundnut was reported to be 
associated with giving birth to very big babies and, hence, its prohibition among 
pregnant women (Girard et al., 2012).  
 
The similarities between the lists of foods liked and disliked at Kpachilo and Yilkpene 
identified in the study could be attributed to their common socio-cultural factors such as 
ethnicity and religion. Other socio-cultural factors including access to and affordability 
of externally-produced foods, migration, the media and interaction with non-indigenes 
may also contribute to their food preferences. The non-socio-cultural factors for the 
findings include the physical (natural environment), especially climate, vegetation, relief 
and drainage, that influence the elements of the food system such as the main varieties of 
locally-produced staples. 
 
The finding that women and adolescent girls had preference for sugar-sweetened 
beverages presumably relates to the popular perception that they have inclination 
towards sweet and processed products. On the contrary, the findings that men had 
preference for the energy-giving starchy staples and dislike for bouillon cubes, which 
they perceived to cause sexual impotence, might be traced to men’s perceived attraction 
to symbols of power, strength and sexual potency.  
 
Some participants perceived women’s dietary quality to be inadequate in the 
communities. They linked women’s poor diets to the food insecurity situation and not to 
socio-cultural factors such as denying women access to reproductive resources and 
restricting their consumption of some categories of foods. The high cost of quality foods 
such as green-leafy vegetables and fish and the limited varieties of foods consumed were 
 193 
University of Ghana http://ugspace.ug.edu.gh
the two main reasons for women’s poor dietary quality. The limited quantities of 
nutrient-dense foods in women’s diets explain why women’s diets are of low quality. 
The perception among participants that women’s diets are inadequate reflects the sub-
optimal intake of food in low-income countries reported by Dickson et al. (2016) and 
Martin-Prevel et al. (2017). The perception also supports the earlier claim by 
Ramakrishnan (2002) that micronutrient malnutrition is very common among WRAs in 
low-income countries. Participants’ awareness of nutrient gaps in women’s diets can 
serve as an opportunity for an intervention to address them provided the necessary 
resources for it are available. 
 
The findings that other participants, particularly women, perceived women’s diets to be 
adequate in the study communities might reflect a false feeling that women’s diets were 
sufficient in terms of quality and quantity. Entrenchment in the local dietary culture 
coupled, perhaps, with their relatively low acquaintance with different socio-cultural 
environments might have misled the women to consider the qualities of their diets as 
adequate. Availability of fish and vegetables, which was one of the reasons for their 
assessment of women’s diets as adequate, was a necessary but not a sufficient condition 
for dietary adequacy. The health benefits of the available foods needed to be 
supplemented by those of the less- frequently consumed foods such as meat, legumes 
and fruits to ensure that foods belonging to the energy-giving, body-building and 
protective categories were consumed in appropriate proportions. Satisfaction derived 
from the consumption of the foods was only quantitative rather than qualitative. The 
perception that the foods did not make their consumers sick reflected their prior 
identification of absence of disease and preventive potency among the attributes of good 
health. This perception also related to the perception of the presence of good physical 
 194 
University of Ghana http://ugspace.ug.edu.gh
energy to perform chores as an attribute of health. In other words, as long as the 
consumption of the foods did not interfere with the women’s performance of their 
multiple roles, they perceived them to be adequate. There is the need for promoting the 
culture of eating not just for satisfaction but also for health (including micronutrient 
intake). 
 
The effects of food shortages on the community and women identified in this study have 
been reported (Saaka 2013). The non-awareness at baseline of the link between women’s 
diets and their reproductive health outcomes has also been reported by studies in Nigeria 
and Ethiopia by Girard et al. (2012) and Saldanha et al. (2012). Girard et al. (2012) and 
Saldanha et al. (2012) indicated that both community members and health promoters 
related poor pregnancy outcomes to the “divine will” or to God, who determines the 
unborn child’s health. Other studies revealed that community members believed that the 
foetus and food were both contained in the stomach so eating too much food would not 
create enough space for the foetus (Huybregts, Roberfroid, Kolsteren, & Camp, 2009; 
Christian et al., 2006).  
 
Participants’ inability to relate women’s diets to their reproductive health may have 
accounted for community members’ failure to give challenges to women’s dietary 
quality due consideration. There is the need for continuous community-based nutrition 
education activities targeting all community members to understand the importance of 
women’s diets and their reproductive health. Men, who dominate in the production of a 
wider variety of foods, must be encouraged to reserve more of the nutrient-dense foods 
for home consumption. 
 
 195 
University of Ghana http://ugspace.ug.edu.gh
Participants in both communities had positive attitudes towards food taboos and 
restrictions at baseline. Food taboos and restrictions pertaining to WRA were based 
manifestly on superstition and misconceptions about their effects on the women’s and 
their unborn babies’ health (health motives). Society’s eagerness to safeguard its 
collective dignity, spiritual purity and survival is the latent force behind the restrictions 
(Senah, 2003). The perception of some foods as inappropriate for consumption by WRA 
owing to cultural taboos and misconceptions has been reported in other low- and middle-
income countries such as Burkina Faso, India, Indonesia, Nepal, Nigeria and Senegal 
(Kavle & Landry, 2018).  
 
The identification of food taboos and restrictions for WRA agrees with the two sets of 
food beliefs labelled by Gittelson and Vastine (2003) as either permanent or temporary 
(transitory). Identification of the flesh of the dog as a general food taboo on grounds of 
its status as a totem for the Dagomba tribe and in the Islamic faith relates to the claim by 
Senah (2003) that, in almost all cultures, totems exist to create a unique identity for 
people who belong to or are from the same clan or religion. The prevalence of food 
taboos and restrictions for WRA relative to general food taboos confirms the report by 
Gittelson and Vastine (2003) that food taboos and restrictions mostly affect women and 
to some extent, the sick and children. In the current study, very few taboos such as those 
that bordered on cultural identity (e.g. flesh of the dog) and religion (e.g. flesh of the pig 
and, again, the flesh of the dog) affected men. 
 
Several of the temporary food taboos and restrictions identified in the study were similar 
to those found in other cultures. Examples are perceptions about eating oily foods during 
pregnancy, “cold” and heavy foods during lactation in Pakistan, meat and fish in Sudan, 
 196 
University of Ghana http://ugspace.ug.edu.gh
buffalo milk in India and fish in Vietnam (Niloufer et al., 2003). Other similar restricted 
foods were leafy and green vegetables and snail in Nigeria (Ogunjuyigbe & Ojofetimi, 
2006) and “mrigal maach” (a type of fish) in Bangledesh (Levey et al., 2013). Finally, 
the reasons for the taboos and restrictions identified in the study- that is, the desire to 
protect the WRA’s heath and their unborn babies as well as the family’s image and 
dignity - were similar to those reported in other parts of Ghana (Arzoaquoi, 2014a; 
Senah, 2003).  
 
Communities’ positive attitudes towards these taboos and restrictions have implications 
for the quality of women’s diets. Unfortunately, most participants in both communities at 
baseline could not link the negative effects of food taboos and restrictions to the quality 
of women’s diets. Community members’ perception that food taboos and restrictions had 
no negative effects on women’s diets agrees with the report by Levey et al. (2013) that 
the influence of food taboos on women’s dietary practices was low in Bangladesh.  
 
Armar (1989), Niloufer et al., (2003); Perez and Gracia (2013) and Kavle and Landry 
(2018), however, indicated that positive attitudes to food taboos and restrictions led to 
the elimination of nutrient-dense foods from women’s diets and, hence, exposed them to 
malnutrition. Kavle and Landry (2018) reported that in Indonesia, for instance, about 14 
kinds of vegetables, 10 kinds of fish, 10 kinds of animal-source foods and as many as 14 
different types of fruits are avoided by pregnant and lactating women. Non-consumption 
of pigeon-pea, wild (bush) meat, chewable foods, fruits such as dawadawa and, 
especially, green-leafy vegetables could deprive lactating women of their body-building 
and protective nutrients. Potential implications of practising these taboos are the low 
 197 
University of Ghana http://ugspace.ug.edu.gh
quality of women’s diets and their subsequent effects on their well-being and 
reproductive health (Martínez Pérez & Pascual García, 2013).  
 
 
5.3 Women’s dietary practices are likely to improve if the drivers of their dietary 
choices are addressed 
Another essential finding of the study, which explained women’s poor dietary status in 
the communities, was that starchy foods dominated their diets. Majority of the foods that 
were reported by women in their 24-hour dietary recall were starch-based with little 
variety from the other food groups. The inadequate inclusion of foods from other groups 
in the diets leads to poor dietary quality as has been reported in the Ghana Nutrition 
profile (FAO, 2009) and GDHS (2015). The range of foods consumed by women also 
agrees with the findings of two recent studies conducted in Northern Region of Ghana 
which indicated that women’s and young children’s diets there were dominated by grain-
based foods (Armar-Klemesu et al., 2018; Armar-Klemesu et al., 2014). The findings 
also confirm the report by Zereyesus et al. (2014) that more than 40% of women in the 
three northern regions of Ghana ate diets that were characterized by low-diversity scores- 
that is, foods belonging to less than five food groups- an indication of the low quality of 
their foods including their nutrient densities. The current study found 45% of women not 
meeting their minimum dietary diversity requirement of five or more food group at 
baseline. Other studies by Kobati et al. (2012) and Koryo-Debrah (2012) reported that 
women in the savannah zone could not meet most of their nutrient requirements.  
 
Low diversity of women’s diets is not peculiar to northern Ghana.  The low consumption 
of meat, egg, milk, vegetables and fruits reported by Nti (2002), Kobati et al. (2012) and 
Amoateng et al. (2017) has been confirmed by the current study. Studies in other parts of 
 198 
University of Ghana http://ugspace.ug.edu.gh
Africa, Asia, Latin America and the Caribbeans have reported similar findings 
(Workicho et al., 2016; Lee, Talegawkar, Merialdi, & Caulfield, 2013).  
 
Even though majority of women (55%) consumed foods belonging to 5 or more groups, 
the consumption of anchovies, green-leafy vegetables and legumes (particularly 
groundnut) was observed to be low in their diets resulting in sub-optimal quality. The 
small quantities of anchovies (inexpensive small fish) observed in women’s meals could 
be attributed to their inability to afford adequate quantities owing to their low levels of 
incomes. The small quantities of legumes and nuts could be explained by their 
cultivation for cash purposes. Men’s dislike for soya-based foods could have accounted 
for their low consumption by women.   
 
The findings on the frequency of meals consumed in this study contradict report in the 
Ghana Nutrition profile (FAO, 2009) that women consume three meals a day instead of 
four reported in the current study. Another contradictory finding is the less consumption 
of ready-to eat meals in the current study compared to findings from other studies 
(Andaleeb & Caskey, 2007). Andaleeb and Caskey explained that ready-to-eat meals had 
become very common as a result of more people working out of home, particularly in 
urban communities. In the current study, however, owing to the absence of paid 
employment outside the home and low income levels of households, buying ready-to-eat 
meals was uncommon. 
 
The drivers/determinants of women’s dietary choices identified in the current study 
included issues related to food security, perceived healthiness of foods, tastes of foods, 
women’s physiological statuses, food restrictions and taboos, large sizes of households 
 199 
University of Ghana http://ugspace.ug.edu.gh
and sources of information on foods/diets. The drivers identified in this study have been 
reported in earlier studies carried out among women in urban Africa and other low and 
middle-income countries (Kavle & Landry, 2018; Gissing, Pradeilles, Osei-Kwasi, 
Cohen, & Holdsworth, 2017; Monterrosa, 2017). The influence of household food 
insecurity further reflects the barrier posed by economic constraints to access to more 
nutritious foods such as dairy products, fish, chicken, eggs and meat among women in 
general in low-income countries such as Ethiopia, Tanzania and Kenya (Young and Pike, 
2012). Similarly, the influence exerted by sources of information on diets agrees with the 
earlier findings that most pregnant women consumed foods recommended by members 
of their families (especially mothers and mothers-in-law) and friends owing to their 
respect for these trusted traditional sources of information about diets (Choudhury & 
Ahmed, 2011).  
 
Strategies to enhance WRA’s diets should take due cognizance of the seven factors 
identified in the current study that influence women’s choices of food. Neglect of any 
one or some of the factors would not augur well for WRA’s dietary status because of the 
strong inter-relationships among them. 
 
5.4 Community readiness and intervention coverage  
A cardinal finding was that even though low levels of readiness and intervention 
coverage were reported at Yilkpene, the intervention led to some improvements in 
participants’ dietary knowledge, attitudes and practices. Practices relating to gender 
imbalances in the food system, however, did not change. The intervention also led to the 
sensitization of members of the intervention community on the need to address socio-
 200 
University of Ghana http://ugspace.ug.edu.gh
cultural barriers to women’s diets such as women’s access to and control of household 
food resources, preferential food allocation practices and taboos affecting women’s diets. 
 
Low readiness scores have been reported in other public health-related interventions 
such as weight/obesity and HIV/AIDS (Pradeillis et al., 2016; Findholt, 2007; Kennedy 
et al., 2004). The level of readiness was low in this study presumably because members 
of the community who had the general feeling that the issue of WRA’s diets needed to be 
addressed felt they were not empowered enough in terms of financial resources to 
embark on such an intervention. Another presumption may have emanated from the 
general feeling of helplessness and inertia that dominates traditional rural communities, 
where external efforts targeted at local problems tend to be more welcomed than the 
internal.  
 
The low coverage of the intervention (in respect of the number of participants attending 
group meetings) might be explained by the pro-female nature of the intervention, which 
sought to address women’s dietary challenges. At baseline, most participants perceived 
women’s dietary quality as the norm and not peculiar to only women. Hence, there was 
no need to address it. The second reason might be the socio-cultural nature of the 
intervention, which sought to encourage community members, particularly men, to 
empower women in terms of accessing more food resources.  
 
The nutrition and dietary education led to improved dietary knowledge and perceptions 
and improved methods of food preparation among participants at Yilkpene. The 
improvements in dietary knowledge reported in this study agree with reports of earlier 
evaluation studies that community-based nutrition education improves participants’ 
 201 
University of Ghana http://ugspace.ug.edu.gh
knowledge on nutrition (Otoo & Adam, 2016; Abu, Louw, Raubenheimer, & Berg, 
2014; Tchum et al., 2009). Improvements in the proportion of women achieving the 
minimum dietary diversity after an intervention have been reported in Mali (Bonde, 
2016). Contrary to the findings by Otoo and Adam (2016), this study did not find 
improvements in the post-intervention intake of micronutrient-dense foods. Anchovies, 
were universally consumed at all major meals in the households owing to their relatively 
high scopes of availability and affordability but quantities consumed remained relatively 
small. Groundnut was also commonly consumed because women kept parts of their own 
produce and what they earned as payments for labour services for consumption by their 
households but quantities consumed were very small. 
 
The lower scopes of consumption of legumes such as cow- and pigeon-peas and soya-
bean, meat, egg, milk and fruits could be attributable to economic and attitudinal factors. 
The peas, soya bean and livestock were produced mainly for commercial purposes and, 
hence, their low domestic consumption. In the case of soya bean, in particular, the 
additional reason for its low scope of consumption was the low level of preference for it 
by the menfolk in the communities who associated it with stomachache, diarrhoea and 
stroke. 
 
The reasons for the lower levels of consumption of fruit and vegetables could be both 
economic constraints and seasonal variations in levels of availability (Abizari, Azupogo, 
Nagasu, Creemers, & Brouwer, 2017). Economically, women’s levels of income are too 
low to afford them fruits and vegetables when their seasons are over. Additionally, since 
women are not supported financially by men to acquire other household food items, they 
are unable to afford them as and when needed. 
 202 
University of Ghana http://ugspace.ug.edu.gh
Advocacy on effects of food taboos and restrictions on women’s diets yielded some 
behaviour modification. Pregnant and lactating women reported consuming most of the 
plant-based foods such as slimy foods, which were hitherto considered restricted foods 
far lactating women; however, attitudes towards all the animal-source foods taboos did 
not change in both communities. The probable explanations could be the entrenched 
socio-cultural nature of such practices bordering on superstition and respect for tradition 
as indicated by Senah (2003). According to Senah, taboos exist to create a cultural 
identity among groups of people or a clan and that a violation of such beliefs attracts 
punishment from either the group or gods. Failure to abide by such practices may also 
contribute to one losing one’s cultural identity. Hence, there is the need to find other 
strategies to address these entrenched beliefs and practices. 
 
The two components of the intervention - that is, nutrition/dietary education and 
advocacy on women’s access to and control of food resources - were both implemented 
as planned. Participants were sensitized on the need to empower women with more food 
resources and not necessarily to bring about a change in behaviour since socio-cultural 
issues are entrenched and may need prolonged periods of sensitization to bring about 
change. Raising consciousness is suggested as the first step in a behaviour change 
process (Prochaska, Colleen, & Evers, 2008). Raising people’s consciousness on a 
particular behaviour can be achieved through advocacy and sensitization.  
 
The intervention succeeded in raising Yilkpene’s level of consciousness on some socio-
cultural practices affecting women’s diets even though the feedback from male members 
was not favourable. For instance, the male participants demonstrated their displeasure at 
sensitizing the community on the need to allow women more access to household food 
 203 
University of Ghana http://ugspace.ug.edu.gh
resources. They indicated that women could not be trusted to manage food resources 
effectively. The misgivings about the call for women to have direct access to grain barns 
as well as those about the calls to give women and children, rather than men, better 
portions of food derive from objections to any suggestion suspected to be aimed at 
tampering with sensitive socio-cultural systems such as power and authority, social 
control and stratification. Women and children’s diets could be the victims of such acts 
of cultural reaction. 
 
Post-intervention results showed improved dietary knowledge and the proportion of 
women achieving minimum dietary diversity in both the intervention and comparison 
communities. Ngowi et al. (2008) reported similar findings in a randomised control trial 
in a community-based intervention to estimate the effectiveness of health and pig-
management education intervention to reduce the incidence rate of a pig disease in 
Tanzania. The study attributed the improved knowledge of pig-management in both 
intervention and control groups to their inability to control the experimental conditions, 
which resulted in information contamination either from the intervention group or 
another external source and the Hawthorne effect (Ngowi et al., 2008). The Hawthorne 
effect is a situation where an individual changes his or her behaviour because he or she is 
aware of being observed. Thus, giving a false effect of the intervention being introduced. 
 
In the current study, however, the experimental condition was controlled to a large extent 
by monitoring both the intervention and comparison communities to account for other 
community activities outside of the intervention that might help explain the post-
intervention outcomes. Data from the monitoring activities indicated that a possible 
explanation for the improvement in the proportion of women achieving the minimum 
 204 
University of Ghana http://ugspace.ug.edu.gh
dietary diversity at Kpachilo was the nutrition-related activities of NGOs in the 
community as well as the involvement of the District Nutrition Officers in the 
intervention activities. The inability of the intervention to effectively control for 
intervention contamination is a major weakness associated with quasi-experimental 
designs (Fisher & Foreit, 2002).  
 
The success of the intervention may have derived partially from its participatory nature 
in which the principles of adult education were applied throughout the intervention 
process. The active participation of community stakeholders in the selection of identified 
challenges to women’s diets, their proposed solutions to addressing the challenges and 
contributions to the design of a communication strategy for the intervention encouraged 
ownership of the intervention among members of the community. In addition, the 
community selected some of its own members to form part of the implementation team. 
 
The adult learning principles which the intervention drew on included the principles of 
involving adults in the planning and evaluation of their learning exercises as indicated 
above; using participants’ experiences as the basis for the learning activities and topics 
or subjects learnt having immediate relevance to participants’ personal lives (Knowles, 
1984). Participants’ rich stores of experiences on diets and their readiness to learn about 
their dietary situation were utilized during the group meeting sessions, when participants 
were encouraged to share their personal experiences with their groups before discussing 
topics scheduled for the day. 
 
Facilitating the group meetings and other sessions in this manner assured participants of 
their self-respect, thus, allowing them to fully participate in the intervention activities. 
 205 
University of Ghana http://ugspace.ug.edu.gh
The nutrition and dietary education as well as the training on proper methods of 
preservation, storage and preparation of food helped to increase participants’ self-
efficacy to be able to practise some of the messages such as improving their cooking and 
practices of environmental hygiene. 
 
Other factors that might have accounted for the successful implementation of the 
intervention include the use of the local dialect of the community as the medium of 
facilitation and communication; rapport and mutual trust between the research team and 
participants and relevance of messages to women’s and household dietary needs. Other 
factors include the convenient timing of the activities, use of homogenous groups of 
participants, convenient seating arrangements and the interactive approach to facilitation. 
The suggestions made by participants for the intervention to be sustained through 
continuation of meetings and their readiness to provide accommodation facility for the 
research team to live with them in the community attest to their appreciation of its 
benefits to the community. 
 
Implications: The findings give credence to the fact that when participatory interventions 
addressing socio-cultural barriers to women’s diets are sustained, they are able to make 
the desired impact on communities in terms of improved knowledge and attitudes as well 
as creating community members’ awareness of socio-cultural issues that contribute to 
women’s poor diets. On the contrary, the low consumption of soya bean and other 
nutrient-dense foods augurs ill for WRA’s diets partially because of their deprivation of 
the benefits of soya bean, a protein-rich food (Sanjukta & Rai, 2016). Additionally, 
women’s non-access to grain barns could continue to contribute to women’s poor dietary 
quality. 
 206 
University of Ghana http://ugspace.ug.edu.gh
CHAPTER SIX 
6.0 SUMMARY, CONCLUSIONS AND RECOMMENDATIONS 
In this chapter, a summary of the study and main findings, their implications for 
women’s diets, conclusions and recommendations are presented.  Also, contributions the 
study has made to knowledge and areas for further study relevant to WRA’s diets are 
indicated. 
 
6.1 Summary of main findings 
This study is a qualitative community participatory quasi-experimental research 
conducted in rural Northern Region of Ghana. The study adopted the social cognitive 
theory by Albert Bandura to explore how individual, community and environmental 
factors affect women’s diets. Sixty-seven interviews were conducted using focus group 
discussions, in-depth interviews and key informant interviews among others at Yilkpene 
and Kpachilo in Savelgu-Nanton District at baseline. Thereafter, a community readiness 
assessment was conducted using the community readiness model as a tool. 
 
The findings of the baseline and community assessment were used to design an 
attitudinal and social behavioural change intervention in consultation with key 
community stakeholders. The implementation of the intervention activities lasted a 
period of seventeen weeks. Assimilation over a period of three months was allowed 
before a post-evaluation of the intervention was carried out using the same data 
collection tools as baseline. Thematic analysis was conducted with emphasis on 
comparing results from baseline and endline to assess the changes resulting from the 
intervention. The main findings of the study were as follows:  
 207 
University of Ghana http://ugspace.ug.edu.gh
1. Women’s involvement in all aspects of the production chain was low owing to 
inadequate access to productive resources. 
2. Women’s non-access to food barns and incomes from household produce 
alongside their earnings of relatively-low and irregular incomes hinders their 
access to optimal diets.  
3. The pro-male nature of utilization of food at the household level together with the 
scoping strategies adopted to mitigate the effects of food shortage are major 
barriers to women’s optimal diets.  
4. The overall readiness in the intervention community for an intervention 
addressing socio-cultural barriers to women’s dietary quality was the stage of 
vague awareness-that is, some members of the community had begun to 
appreciate the sub-optimal quality and quantity of women’s diets but, 
unfortunately, they lacked the motivation to do anything about the problem. 
5. Dietary knowledge was sub-optimal at baseline but improved at endline in both 
communities 
6. Perceptions on women’s adequate dietary quality and its relations to women’s 
reproductive health improved substantially after the intervention. 
7. The intervention made positive change in dietary knowledge and practices 
despite low levels of coverage and community readiness. 
8. The intervention also demystified socio-cultural food restrictions in   plant-source 
foods during pregnancy and lactation.  
9. The majority of women in both communities consumed at least four meals daily 
and micronutrient-dense foods were either not consumed frequently or were 
consumed in very small quantities at both baseline and endline. 
 208 
University of Ghana http://ugspace.ug.edu.gh
10. More females than males could recall more key messages during post-
intervention evaluation. 
 
6.2 Implications of findings for women’s diets 
Women’s diets will continue to suffer in terms of quality and variety if the socio-cultural 
barriers identified in this study are not addressed. Interventions addressing socio-cultural 
barriers have a high potential of achieving success if members of communities 
participate actively in the design and implementation process and support to sustain the 
intervention activities for long durations. Sustained intervention activities are feasible if 
existing community structures or resources are utilized in facilitating them. The link 
between agricultural productivity, nutrition and health must be strengthened since rural 
women’s diets depend, to a large extent, on the local production of a variety of food 
items. The use of agro-chemicals needs to be critically examined and farmers educated 
on it. The need for communities, and women in particular, to gain more knowledge 
through education, either formally or non-formally, on dietary issues is critical for 
enhancing the quality of women’s diets. 
 
6.3 Conclusions 
The following conclusions are drawn in reference to the objectives of the study:  
Women in Savelugu-Nanton District interact closely with all aspects of the food system- 
that is, production, processing, preservation and storage, preparation, distribution and 
consumption; however, the pro-male nature of the food system serves as a strong barrier 
to women’s optimal diets. 
 
 209 
University of Ghana http://ugspace.ug.edu.gh
Sub-optimal dietary knowledge and attitudes hinder women’s access to optimal diets at 
Yilkpene and Kpachilo. The situation is reflected in the conservative attitudes towards 
socio-cultural food restrictions on some animal-source foods and vegetables, thereby 
depriving women of some essential nutrients including micronutrients.  A false feeling of 
women’s dietary adequacy also emanates from their sub-optimal dietary knowledge and 
attitudes. 
 
Some aspects of women’s dietary practices at Yilkpene and Kpachilo, which are more or 
less uniform, do not promote their dietary status. The main factor responsible is their 
sub-optimal dietary diversity such as inadequate consumption of micronutrient-dense 
foods. The main drivers of women’s dietary choices, which together contribute to 
depriving them of optimal diets, are mainly socio-cultural in nature - perceived 
healthiness of foods, food taboos and restrictions, tastes of foods, large sizes of 
households and sources of dietary information. Other factors are household food security 
situation and women’s physiological statuses. 
 
The enormity of the threat posed by socio-cultural barriers to women’s optimal diets at 
Yilkpene is revealed by the low scope of readiness on the part of members of the 
community to modify them (the barriers). The situation might be traced to the entrenched 
nature of the barriers including women’s low decision-making power on food resources, 
low levels of formal education, earnings of low, irregular incomes and large sizes of 
households. 
 
Attitudinal and social-behavioural change communication interventions addressing 
socio-cultural issues require a lot of time and other resources to create remarkable 
 210 
University of Ghana http://ugspace.ug.edu.gh
change in community knowledge and attitudes to women’s diets. Adequacy of resources 
is necessary to engage communities to begin to reflect on their current practices and to 
assess how best to address them without compromising their cultural identity. 
Additionally, the findings indicate that adopting a participatory approach to nutrition and 
dietary intervention adds to the body of evidence that engaging communities to deliver 
dietary information leads to improvements in their levels of knowledge and attitudes 
(perceptions and beliefs).  
 
6.4 Recommendations 
Based on the results and conclusions of the study, the following recommendations are 
made to enhance WRA’s diets in Savelgu-Nanton District: 
Community 
1. Women’s interaction with the food system should be enhanced by assisting them 
to transport their produce through arrangements with men who have access to 
vehicles. This strategy would increase the quantities of foods available to the 
households as well as reduce the negative effects of heavy workload on women’s 
health. In order to reduce the scope of loss through sun-drying, shade- and solar-
drying should be encouraged especially in respect of green leafy vegetables. 
Fermentation of grains should not be limited to those used for porridges but 
extended to those used for other dishes such as TZ. Women should be assisted to 
acquire the specially-treated bags for storage of produce from the ministry of 
agriculture. They should also be educated on the importance of proper 
preservation and storage after drying. 
 
 211 
University of Ghana http://ugspace.ug.edu.gh
2. To enhance women’s diets through their dietary practices and choices as well as 
their interaction with the food system, conscious efforts must be taken by 
members of the communities to cultivate green-leafy vegetables for the home to 
sustain availability throughout the year. Perennial availability of the vegetables 
will assure women of their health benefits such as iron and vitamins. In this 
respect, members of the communities must be discouraged from thinking that 
“one does not need to look for what one already has” and rather begin to think 
more positively that “one can actually improve upon what one already has while 
looking for what one has not but needs” by consciously cultivating such crops 
during the peak periods of the year and embarking on backyard gardening during 
the dry season to ensure all year availability.  
 
3. Community opinion leaders such as traditional rulers, religious leaders and health 
volunteers should organize activities to raise and sustain members’ readiness to 
modify socio-cultural norms that act as barriers to women’s diets. Relevant 
activities include public fora, home visits and group meetings, which will sustain 
community members’ attention on issues at stake and, thus, motivate them to deal 
with the issues on a sustained basis. 
 
4. Identification of religion, culture and superstition as the forces behind the food 
taboos points to the need to adopt taboo- and restriction-specific strategies to 
tackle them. The culture-related taboos and restrictions could be modified, 
presumably, through the co-operation of traditional community leaders and heads 
of households. The superstition-related taboos and restrictions require the 
creation of awareness about the health benefits of the taboo foods. Food taboos 
 212 
University of Ghana http://ugspace.ug.edu.gh
and restrictions on women’s diets would, thus, be mitigated only through an 
intervention that caters for established socio-cultural structures of power, 
authority and communality. 
 
Ghana Health Service and the media 
1. To enhance community members’ dietary knowledge and attitudes, Ghana Health 
Service should implement sustained dietary education through community-based 
interventions, ANC and PNC sessions as well as advocacy on socio-cultural 
barriers to women’s diets. GHS needs to work in collaboration with the media for 
the interventions to benefit from the latter’s immense influence on public opinion 
and mass mobilization. The roles of the media as key educators and providers of 
information should be utilized by GHS to enhance the effectiveness of its 
interventions targeted at socio-cultural barriers to women’s diets especially in 
relatively lowly-populated rural communities in which members have close 
interactions. The collaboration will, thus, enhance dietary knowledge and 
attitudes in the district. Women’s diets will, again, benefit from the interaction 
through improvements in women’s interaction with the food system, their dietary 
practices and choices. 
2. Adverse effects of the sale of legumes and nuts, which constitute the main 
nutrient-dense foods in the communities, on women’s dietary quality need to be 
addressed through nutrition educational activities of Ghana Health Service. The 
activities should include the education of members of communities, particularly 
men, who are customarily the heads of households about the health benefits of 
these products and the need to reserve sufficient quantities for home 
consumption. Improved awareness and knowledge of the nutritive benefits of 
 213 
University of Ghana http://ugspace.ug.edu.gh
anchovies, green-leafy vegetables, legumes and nuts would encourage WRA to 
consume them more frequently. In general, frequent consumption of fruits such 
as water-melon, one of the seasonal fruits cultivated in the communities, and 
meat, a rich source of iron, could help to improve women’s dietary quality and 
micro-nutrient status.  
 
3. Ghana Health Service should organize training for local health volunteers on 
nutrition and healthy dietary practices.  Improved dietary knowledge and 
attitudes among local facilitators would enhance the effectiveness of the 
behavioural change interventions that the service should organize with the media 
and other stakeholder-organizations.   
4. The interventions should exploit relevant results from the study. An example is 
the finding that deviant WRA experienced neither spiritual nor physical (health) 
side-effects of their consumption of the restricted foods. 
5. The coping strategies adopted by women to mitigate the effects of food shortages 
call for WRA-friendly strategies. These strategies include backyard cultivation of 
green-leafy vegetables in the dry season and improved preservation and storage 
of farm produce to improve women’s access to adequate diets.  
 
6. The findings on community members’ dietary knowledge and attitudes call for 
interventions to enhance awareness of women’s diets to be linked with their 
preferred foods and their methods of processing, preservation, storage and 
preparation. The enhanced awareness would motivate consumption of healthier 
foods to improve their nutritional status. Additionally, a communication 
intervention would be useful to enhance participants’ knowledge of dietary 
 214 
University of Ghana http://ugspace.ug.edu.gh
qualities of less-preferred foods and, thus, eliminate misperceptions regarding 
such foods. 
7. Strategies to enhance WRA’s diets should take due cognisance of the seven 
drivers identified in the current study that influence women’s choice of food. 
Neglect of any or some of the drivers would not augur well for WRA’s dietary 
status because of the inter-relationships among them. 
NGOs in Savelugu-Nanton District: RING, RAIN, World Vision and other 
institutions 
1. To ensure that women’s diets benefit adequately from their interaction with the food 
system, the acquisition of farming inputs needs to be enhanced through more access to 
credit facilities for both men and women. There is also the need for women to have 
improved access to their male household members’ labour through negotiations and 
hired labour through access to improved credit facilities. 
 
2. Sale of foods at both farm gates and external markets requires the empowerment and 
mobilization of local women through producers’ (farmers’) co-operatives and other 
groups. These measures would offer them strong bargaining power and greater levels of 
financial security against shocks such as unfavourable climate and transport difficulties. 
Women’s relatively-low incomes hinder their performance of their roles such as milling 
the grains allocated by the men and acquiring ingredients for soups and sauces. This 
deficiency, which leads to poor quality of diets for the households in general and women 
in particular, would be mitigated through improved earnings. 
 
 215 
University of Ghana http://ugspace.ug.edu.gh
3. There is also the need to empower women to add value to their produce through 
improved techniques of production, processing and preservation. This strategy would 
help to raise their levels of income.  
 
6.5 Contributions to knowledge 
One key contribution this study has made to the existing body of knowledge is that the 
local food system in rural Savelgu-Nanton District is driven by gender and socio-cultural 
norms, which put women at a disadvantage in respect of their access to food resources at 
the household level. These inequities need to be addressed holistically if women’s 
dietary quality is to reach optimal levels on a sustained basis. Second, even though 
community readiness and intervention coverage may be low in community-based 
interventions, the interventions may still be successful if other process indicators such as 
level of community participation and community perceptions of the intervention are 
assessed using qualitative methods. Third, in spite of modernization and extensive 
facility- and community-based nutrition and dietary education, women’s diets still suffer 
from food taboos in most rural communities. 
 
Theoretical contribution 
The CRM had previously been used mostly in public health related interventions such as 
management of obesity and weight, HIV/AIDS and domestic violence but not in an 
intervention addressing socio-cultural barriers to women’s optimal diets. This study has 
demonstrated that the model can be effectively used in public health-related 
interventions in which sustainability is usually a daunting challenge. 
 
 216 
University of Ghana http://ugspace.ug.edu.gh
The study combined the SCT and CRM in the design and implementation of a 
community-based behavioural change communication intervention to complement each 
other. While the SCT focused on inter-personal self- and collective-efficacy to accept, 
modify and sustain change, the CRM concentrated on the readiness of a community to 
effect and sustain change in attitudes and, eventually, behaviour. The theoretical 
contribution of the study to knowledge is, thus, that combining the two models in 
community-based interventions could lead to increased community participation and 
ownership of the intervention. 
 
Methodological contribution 
Perhaps for the first time, and to the best of my knowledge, an intervention study in 
Ghana has applied the CRM to determine the stage of readiness of a community to 
implement an intervention to address a public health issue. The five dimensions by 
which the stage of readiness of the community was assessed were relevant in 
determining whether its resources and structures could be effective to implement and 
sustain the intervention especially when it involved socio-cultural issues. The study led 
to the design and development of a communication strategy through a participatory 
approach to addressing socio-cultural barriers to women’s diets in a typical rural setting. 
 
Policy contributions 
The study gives credence to the fact that if communities are given the necessary support 
by researchers, they can design and implement context-and culturally-specific 
interventions to address issues affecting them. 
 
 217 
University of Ghana http://ugspace.ug.edu.gh
Implications for further research 
A mixed methods approach to a similar study is suggested since this study was unable to 
state quantitatively the extent to which dietary knowledge and attitudes improved at the 
post-intervention stage. Additionally, women reported adopting improved methods of 
preparation of food and non-adherence to restrictions against the consumption of slimy 
foods at post-intervention stage but the scopes of these changes could not be verified 
quantitatively. 
 
6.6 Limitations of the study 
The study used only qualitative methods in data collection and analysis and, therefore, 
was unable to quantify the margins of change in the indicators measured. For instance, 
even though knowledge improved in both communities, the study could not indicate the 
proportion of participants whose knowledge improved. The 24-hour dietary recall could 
only recall the foods women consumed the previous day, which may not be a true 
representation of all the varieties of foods they usually consumed. 
 
Second, the design of the intervention, quasi-experimentation, which does not allow 
participants equal chance of being assigned to the different groups, may have introduced 
selection bias into the study. Last but not least, the intervention suffered from 
information contamination through similar activities introduced into the comparison 
community, which may have accounted for the improved outcome indicators there.  
 
 
 
 
 218 
University of Ghana http://ugspace.ug.edu.gh
REFERENCES 
Aasoglenang, A. T., & Bonye, S. Z. (2013). Rural Livelihoods Diversity: Coping 
strategies in Wa West District in northern Ghana. European Scientific Journal, 
9(35), 139-156.  
Abizari, A.-R., Azupogo, F., Nagasu, M., Creemers, N., & Brouwer, I. D. (2017). 
Seasonality affests dietary diversity of school-age childen in northern Ghana. 
PLoS One, 12(8).  
Abu, B. A. Z., Louw, V. J., Raubenheimer, J. E., & Berg, L. V. V. d. (2014 ). 
Effectiveness of a nurtition education intervention to improve knowledge, 
attitudes and practices regarding iron deficiency among mothers in Ghana. 
African Journal of clinical nutrition, 27(3).  
Abu, B. A. Z., Louw, V. J., Raubenheimer, J. E., & Berg, V. d. (2014). Designing 
interventions for resource poor communities with low literacy: An example of an 
iron deficiency (ID) education program in Ghana.  . 
http://micronutrientforum.org/wp.  
Agarwal, B. (2014). Food security and democratic choice: Critical contradictions. The 
Journal of Peasant Studies, 41(6), 1247-1268.  
Ahrari, M., Houser, R. F., Yassin, S., Mogheez, M., Hussaini, Y., Crump, P., . . . 
Levinson, J. (2006). A Positive Deviance-Based Antenatal Nutrition Project 
Improves Birth-weights in Upper Egypt Journal of Health, Population and 
Nutrition, 24(4), 498-509.  
Aikins, A. d.-G. (2014). Food beliefs and practices during pregnancy in Ghana: 
Implications for maternal health interventions. Health care for Women 
Internations, 35(7-9), 954-972.  
Allison, R., Lecky, D. M., Town, K., Rugman, C., Ricketts, E. J., Ockendon-Powell, N., 
. . . McNulty, C. A. M. (2017). Exploring why a complex intervention piloted in 
general practices did not result in an increase in chlamydia screening and 
diagnosis: a qualitative evaluation using the fidelity of implementation model. 
BMC Family Practice, 18(43). doi:10.1186/s12875-017-0618-0 
Amoateng, A. Y., Doegah, P. T., & Udomboso, C. (2017). Socio-Demographic Factors 
Associated with Dietary Behaviour among Young Ghanaians Aged 15-34 Years. 
J Biosoc Sci, 49(2), 187-205. doi:10.1017/S0021932016000456 
Amugsi, D. A., Lartey, A., Kimani-Murag, E., & Mberu, B. (2016). Women's 
participation in household decision-making and higher dietary diversity: Findings 
from nationally representive data from Ghana. Journal of Health, Population and 
Nutrition, 35(16).  
Andaleeb, S., & Caskey, A. (2007). Satisfaction with food services:Insights from a 
college cafeteria Journal of Food Service Business Research, 10(2), 51-65.  
Armar, M. (1989). Maternal energy status, lactation capacity and infant growth in 
Southern Ghana: A study of the interactions of cultural and biological factors.  . 
(PhD), University of London, London, UK.    
Armar-Klemesu, M. (2000). Food habita and attitudes study. Project commissioned by 
Unilever Ghana Ltd. Accra. Retrieved from ( Unpublished):  
 219 
University of Ghana http://ugspace.ug.edu.gh
Armar-Klemesu, M., Osei-Menya, S., Zakariah-Akoto, S., Tumilowicz, A., Lee, J., & 
Hotz, C. (2018). Using ethnography to identify barriers and facilitators to optimal 
Infant and Young Child Feeding in rural Ghana: Implications for programs. Food 
and Nutrition Bulletin, 1-15.  
Armar-Klemesu, M., Pelto, G. H., Folson, G., Osei--‐Menya, S., Zakariah--‐Akoto, S., & 
Monterrosa, E. (2014). A focused ethnographic study of food consumption 
patterns and their social and cultural context in pregnant and breastfeeding 
women and in adolescents. DSM Sight and life. Retrieved from 
https://www.noguchimedres.org/index.php/prof-margaret-armar-klemesu 
Arzoaquoi, S. K. (2014a). Common food Taboos and Beliefs During Pregnancy in Yilo 
Krobo District, Ghana. (Masters in Public Health Dissertation), University of 
Ghana, Ghana.    
Arzoaquoi, S. K. (2014b). Common Food Taboos and Beliefs During Pregnancy in Yilo 
Krobo District, Ghana. (M'Phil in Public Health Dissertation), Uninversity of 
Ghana, Ghana.    
Atasie, V. N., Akinhanmi, T. F., & Ojiodu, C. C. (2009). Proximate Analysis and 
Physico-Chemical Properties of Groundnut )Arachis hypogaea L. Pakistan 
Journal of Nutrition, 8(2).  
Baden, S., Green, C., Otoo-Oyortey, N., & Peasgood, T. (1994). Background paper on 
gender issues in Ghana. Retrieved from Brighton:  
Bambangi, S., & Abubakari, A. (2013). OWNERSHIP AND ACCESS TO LAND IN 
URBAN MAMPRUGU, NORTHERN GHANA. International Journal of 
Research In Social Sciences, 2(2).  
Bandura, A. (1997). Self-Efficacy: The exercise of control (Vol. 1). USA: W.H. Freeman 
Company. 
Bell, S., & Aggleton, P. (2016). Monitoring and Evaluation in Health and Social 
Development: Interpretive and Ethnographic Perspective. 2 Park Square, Milton, 
Abingdon, Oxon OX14 4RN: Routiedge 711 Third Avenue, New York, NY 
10017. 
Black, R. E., Victora, C. E., Walker, S. P., Bhutta, Z., Christian, P., Onis, M. d., . . . 
Group, t. M. a. C. N. S. (2013). Maternal and Child Nutrition 1: Maternal and 
child undernutrition and overweight on low-income and middle-income countries  
The Lancet, 382(1), 427-451.  
Boakye-Achampong, S., Mensah, J. O., Aidoo, R., & Osei-Agyemang, K. (2012). The 
role of rural women in the attainment of household food security in Ghana: Acase 
study of women-farmers in Ejura Sekyeredumasi District. International Journal 
of Pure and Applied Sciences and Technology http://www.ijopaasat.in, 12(1).  
Bonde, D. (2016). Impact of agronomy and livestock interventions on women and 
children's dietary diversity in Mali. Field Exchange, 51.  
Braimah, M. M., Dukuh, I. G., Oppong-Sekyere, D., & Momori, P. H. (2014). Assessing 
the socio-cultural factors that affect women's  contribution to food security in the 
Kasena Nankana Municipality, Upper-East Region, Ghana. International Journal 
of Plant, Animal and Environment Sciences 4(1).  
 220 
University of Ghana http://ugspace.ug.edu.gh
Braun, V., & Clarke, V. (2006). Using thematic Analysis in psycholoogy. Qualitative 
reseach in psychology. http://www.informaworld.com/smpp/content db=all 
content=a795127197 frm=titlelink. , 3(2), 77-101.  
Cassidy, A., & Patterson, B. (2008). The planner's Guide to the urban food system. 
Retrieved from University of Southern Califonia:  
Choudhury, N., & Ahmed, S. M. (2011). Maternal care practices among the ultra poor 
households in rural Bangladesh: a qualitative exploratory study. BMC Pregnancy 
and Childbirth, 11(15), 1471-2393.  
Christian, P., Srihari, S. B., Thorne-lyman, A., Kharty, S. K., LeClerq, C. S., & Shrestha, 
S. R. (2006). Eating down in pregnancy: exploring food related beliefs and 
practices of pregnancy in rural Nepal. Ecology of Foood and Nutrition, 45(4), 
253-278.  
Colecraft, E., Marquis, G. S., Aryeetey, R., Sakyi-Dawson, O., Lartey, A., Ahunu, B., . . 
. Huff-Lonergan, E. (2006). Constraints on the Use of Animal Source Foods for 
Young Children in Ghana: A Participatory Rapid Appraisal Approach. Ecology 
of Food and Nutrition, 45(5), 351-377. doi:10.1080/03670240600985464 
Colecraft, E. K., Marquis, G. S., Sakyi-Dawson, O., Lartey, A., Butler, L., Ahunu, B., . . 
. Canacoo, E. (2012). Planning, designing and implementation of the Enhancing 
Child Nutrition through animal source food management (ENAM) Project. 
AJFAND, 12(1).  
Contenta, I. R. (2008). Review of nutrition education research. Jounal of nutrition and 
behaviour, 30(4), 456-471.  
Creswell, W. J. (2009). Research Design: Qualitative, Quantitative and Mixed Methods 
Approaches (Third edition ed.). USA: Sage Publications, Inc. 
Darfour, B., & Rosentrater, K. A. (2016). Agriculture and food security in Ghana. 
Agriculture and Biosystems Engineering: Conference proceedings and 
presentations.  
Davis, R., Cambell, R., Hildon, Z., Hobbs, L., & Susan, M. (2014). Theories of 
behaviour and behaviour change across the social and behaviour sciences: a 
scoping review. Health  Psychology Review.  
Delormier, T., Frohlich, K. L., & Potvin, L. (2009). Food and eating as social practice – 
understanding eating patterns as social phenomena and implications for public 
health. Sociology of Health & Illness  Vol. 31(No. 2 ), 215–228. 
doi:10.1111/j.1467-9566.2008.01128.x 
Doss, C. R., & Morris, M. (2001). How Does Gender Affect the Adoption of 
Agricultural Innovations? The case of improved Maize Technology in Ghana. 
Agricultural Economics.  
Doss, C. R., & Morris, M. L. (2000). How does gender affect the adoption of agricultural 
innovations? Agricultural Economics, 25(1), 27-39.  
Douglas, M. (2014). Food in the social order: Routledge. 
Ducharma, F., & Trudeau, D. (2002). Qualitative evaluation of a stress management 
intervention for elderly caregivers at home: A constructivist approach. Issues in 
mental health nursing, 23(7), 691-713.  
 221 
University of Ghana http://ugspace.ug.edu.gh
Duflo, E., & Udry, C. (2004). Intrahousehold resource allocation in Cote d'Ivoire: 
Social norms, seperate accounts and consumption choices, . Retrieved from 
http://www.econ.yale.edu/~cru2/pdf/ivoire_RES_comp.pdf 
Dunneram, Y., & Jeewon, R. (2015). Healthy diet and nutrition education program 
among women of reproductive age: A necessity of multilevel strategies or 
community responsibility. Health Promotion Perspectives, 5(2), 116-127.  
Eaves, E. R., Howerter, A., Nichter, M., Floden, L., Gordon, J. S., Ritenbaugh, C., & 
Muramoto, M. L. (2017). Implementation of tobacco cessation brief intervention 
in complementary and alternative medicine practice: qualitative evaluation. 
BMC complementary and  alternative medicine, 17(331).  
Edwards, R. W., Jumper-Thurman, P., Plested, B. A., Oetting, E. R., & Swanson, L. 
(2000). Community Readiness: Research to Practice. JOURNAL OF 
COMMUNITY PSYCHOLOGY, 28(3), 291-307.  
Ersino, G., Henry, C. J., & Zello, G. A. (2017). A nutrition education intervention affects 
the diets-health related practices and nutritional status of mothers and children in 
a pulse-growing community in Halaba, south Ethiopia. The FASEB Journal, 31(1 
(suppleent 786.39)).  
Etiévant, P., Bellisle, F., Dallongeville, J., Donnars, C., Etilé, F., Guichard, E., . . . Tibi, 
A. (2010). Dietary behaviours and practices Determinants, action, outcomes. 
Retrieved from 
http://inra.dam.front.en.pad.brainsonic.com/ressources/afile/224968-6e1fe-
resource-dietary-behaviours-scientific-expertise-summary.html 
FAO. (2007). Climate change and food security: A framework documentary summary. 
Retrieved from http://www.fao.org/3/k2595e/k2595e00.htm 
FAO. (2009). Ghana Nutrition Profile Retrieved from Ghana: 
http://www.moh.gov.gh/wp-content/uploads/2016/02/Nutrition-Country-Profile-
Ghana.pdf 
FAO. (2011). Guidelines for Measuring Household and Individual Dietary Diversity. 
Retrieved from 
http://www.fao.org/fileadmin/user_upload/wa_workshop/docs/FAO-guidelines-
dietary-diversity2011.pdf 
FAO. (2012a). Gender inequalities in rural employment in Ghana: an overview. 
Retrieved from http://www.moh.gov.gh/wp-content/uploads/2016/02/Nutrition-
Country-Profile-Ghana.pdf 
FAO. (2012b). Sustainable nutrition security: Restoring the bridge between agriculture 
and health. Retrieved from http://www.fao.org/3/me785e/me785e.pdf 
FAO, & FHI360. (2016). Minimum D:ietary Diversity for Women: Aguide for 
measurement. Retrieved from Rome: http://www.fao.org/3/a-i5486e.pdf 
Farahmand, M., Tehrani, F. R., Amiri, P., & Azizi, F. (2012). Barriers to healthy 
nutrition: Perceptions and experiences of Iranian women. BMC Public Health, 
12(1064). doi:10.1186/1471-2458-12-1064 
Findholt, N. (2007). Application of the Community Readiness Model for Childhood 
Obesity Prevention. Public Health Nursing, 24(6), 565-570.  
 222 
University of Ghana http://ugspace.ug.edu.gh
Fisher, A. A., & Foreit, J. R. (2002). Designing HIV/AIDS Intervention Studies: An 
operations research handbook: Population Council. 
Fonteyna, M., & Bauer-Wub, S. (2005). Using qualitative evaluation in a feasibility 
study to improve and refine a complementary therapy intervention prior to 
subsequent research. Complementary Therepies in Clinical Practice, 11, 247-
252.  
Gadegbeku, C., Wayo, R., Badu, G. A., Nukpe, E., & Okai, A. (2013). Food tabbos 
among residents at Ashongman; Accra, Ghana. Food Science and Quality 
Management, 15.  
Galbete, C., Mary Nicolaou, Karlijn A. Meeks, Aikins, A. d.-G., Addo, J., Amoah, S. K., 
. . . Danquah, I. (2017). Food consumption, nutrient intake, and dietary patterns 
in Ghanaian migrants in Europe and their compatriots in Ghana. FOOD & 
NUTRITION RESEARCH, 61.  
GHS. (2014). Savelgu-Nanton Municipal Health Directorate: Annual report 2014. 
Retrieved from Ghana: 
https://www.ghanahealthservice.org/downloads/2014_FHD_ANNUAL_REPOR
T.pdf 
GHS. (2018). District Health Information Mangement System (DHIMS), Savelugu -
Nanton District. Retrieved from Ghana: Savelugu District health Mnangement 
Team, Savelugu-Nanton District 
Giddens, A. (1993). Sociology (2nd ed.): Polity Press. 
Gissing, S. C., Pradeilles, R., Osei-Kwasi, H. A., Cohen, E., & Holdsworth, M. (2017). 
Drivers of dietary behaviours in women living in urban Africa: a systematic 
mapping review. Public Health Nutrition.  
Gittelsohn, J., Haberle, H., Vastine, A. E., Dyckman, W., & Palafox, N. A. (2003). 
Macro- and microlevel processes affect food choice and nutritional status in the 
republic of the marshall islands. J Nutr, 133(1), 310S-313S.  
Gittelsohn, J., & Vastine, A. E. (2003). Sociocultural and household factors impacting on 
the selection, allocation and consumption of animal source foods: current 
knowledge and application. J Nutr, 133(11 Suppl 2), 4036S-4041S.  
Glanz, K., Rimer, B. K., & Viswanath, K. (2008). Health Behavior and Health 
Education:Theory, research and practice (Fouth ed.). CA: Jossey-Bass. 
Global Panel on Agriculture and Food System for Nutrition. (2016). Fodd systems and 
diets: Facing the challenges of the 21st century. Retrieved from London, UK:  
GOG. (2013). Ghana Nutrition Policy. Retrieved from Ghana: 
http://ndpc.gov.gh/downloads/Policy Almanac/Nutrition Policy_September 2013 
(Draft).pdf 
GSS. (2012). 2010 Population and housing census: summary report of final results 
Retrieved from 
http://www.statsghana.gov.gh/gssmain/storage/img/marqueeupdater/Census2010
_Summary_report_of_final_results.pdf 
GSS. (2014a). 2010 Population and Housing Census: District Analytical Report- 
Savelgu-Nanton District Retrieved from 
 223 
University of Ghana http://ugspace.ug.edu.gh
http://www.statsghana.gov.gh/docfiles/2010_District_Report/Northern/SAVELU
GU.pdf 
GSS. (2014b). Population and Housing Census: A district analytical report (Central 
Gonja). Retrieved from Ghana: https://www.worldcat.org/title/2010-population-
housing-census-report-district-analytical-report-name-of-district-or-municipal-
assembly/oclc/908660319 
GSS, GHS, & DHS, T. (2015). Ghana Demographic and Healh Survey 2014:Key 
indicators  copy. Retrieved from USA:  
GSS, GHS, & ICF. (2015a). Ghana demographic and Health Survey 2014. Retrieved 
from Rockville, Maryland, USA: http://www.ghana.gov.gh/index.php/media-
center/news/2073-2014-ghana-demographic-and-health-survey-gdhs-report-
launched-in-accra 
GSS, GHS, & ICF. (2015b). Ghana Demographic and Health Survey 2014: Key 
indicators. Retrieved from Accra, Ghana: 
http://www2.statsghana.gov.gh/docfiles/DHS_Report/Ghana_DHS_2014-KIR-
21_May_2015.pdf 
GSS, GHS, & Macro, I. (2009). Ghana Demographic and Health Survey 2008. Retrieved 
from Accra, Ghana: https://dhsprogram.com/publications/publication-fr221-dhs-
final-reports.cfm 
Haileslassie, K., Mulugeta, A., & Girma, M. (2013). Feeding practices, nutritional status 
and associated factors of lactating women in Samre Woreda, South Eastern Zone 
of Tigray, Ethiopia. Nutrition Journal, 12(28). doi:10.1186/1475-2891-12-28 
Harika, R., Faber, M., Samue, F., Kimiywe, J., Mulugeta, A., & Eilander, A. (2017). 
Micronutrient Status and Dietary Intake of Iron, Vitamin A, Iodine, Folate and 
Zinc in Women of Reproductive Age and Pregnant Women in Ethiopia, Kenya, 
Nigeria and South Africa: A Systematic Review of Data from 2005 to 2015. 
Nutrients, 9(1096).  
Hossain, M. Z. (2012). Zakat in Islam: A powerful poverty aleviation instrument for 
Islamic countries. International Journal of Economic Development Research and 
Investment, 3(1).  
Huybregts, L. F., Roberfroid, D. A., Kolsteren, P. W., & Camp, J. H. V. (2009). Dietary 
behaviour, food and nutrient intake of pregnant women in a rural community in 
Burkina Faso. Matern Child Nutr, 5(3), 211-222.  
IFAD. (2013). Reducing rural women's domestic workload through labour saving-
technologies and practices: Gender, targeting and inclusion. Retrieved from 
https://www.ifad.org/documents/38714170/40196082/Teaser_workload_web.pdf/
c8b175be-f4cf-4f97-a3bf-d6720cc08aaf 
Johnson-Welch, C. (1999). Focusing-on-Women-Works-Research-on-Improving-
Micronutrient-Status-through-Food-Based Interventions. Retrieved from 
https://www.popline.org/node/528618 
Kavle, J. A., & Landry, M. (2018). Addressing barriers to maternal nutrition in low- and 
middle-income countries: A review of the evidence and programme implications. 
Maternal and Child Nutrition, 14(1).  
Kearsley, G. (2010). Andragogy (M. kNOWLES). The theory into practice database.  
 224 
University of Ghana http://ugspace.ug.edu.gh
Kennedy, S., Johnson, K., & Alberto. (2004). Evaluation. AIDS, 18(3).  
Knowles, M. (1984). The adult learner: A neglected species (3rd ed.). Houston TX: Gulf 
Publisher. 
Kobati, G. Y., Lartey, A., Marquis, G. S., Colecraft, E. K., & Butler, L. M. (2012). 
Dietary intakes and body mass indices of non-pregnant and lactating women 
(NPNL) from the coastal and guinea savannah zones of Ghana. AJFAND, 12(1).  
Kong, A., Tussing-Humphreys, L. M., Odoms-Young, A. M., Stolly, M. R., & 
Fitzgibbon, M. L. (2014). Systematic review of behaviour interventions with 
culturally-adapted strategies to improve diet and weight outcomes in African 
American women. National Institute of Health, 15(4), 62-92.  
Koryo-Dabrah, A., Nti, C. A., & Adanu, R. (2012). Detary Practices and Nutrient Intakes 
of Pregnant Women in Accra, Ghana. Current Research Journal Biological 
Sciences 4(4), 358-365.  
Kostadinov, I., Daniel, M., Stanley, L., Gancia, A., & Cargo, M. (2015). A systematic 
Review of Community Readiness Tool Application: Implications for Reporting. 
International Journal of Environmental Research and Public Health, 12, 3453-
3468.  
Kpormegbe, S. K., & Ahorlu, C. K. (2014a). THE ROLE OF COMMUNITY 
PARTICIPATION IN INTERMITTENT PREVENTIVE TREATMENT OF 
CHILDHOOD MALARIA IN SOUTH- EASTERN GHANA. Ghana Medical 
Journal, 48(2). doi:http://dx.doi.org/10.4314/gmj.v48i2.1 
Kpormegbe, S. K., & Ahorlu, C. K. (2014b). The role of community participation in 
intermittent preventive treatment of childhood malaria in South-Eastern Ghana. 
Ghana Medical Journal, 48(2).  
Kuhnlein, & Pelto, G. H. (2014). Ethnographic aspects of human nutrition, inThe role of 
food, Agriculture, Forestry, and Fisheries in Human Nutrition (Vol. 4). Paris, 
France (http://www.eolss.net): Encyclopedia of Life Support Systems (EOLSS). 
Kuhnlein, H., & Pelto, G. H. (2009). Ethnographic Aspects of Human Nutrition (Vol. 4). 
Oxford, United Kingdom: UNESCO. 
Kuhnlein, H., & Receveur, O. (1996). Dietary change and traditional food systems of 
indegenous peoples. Annual Review of Nutrition; 
http://www.annualreviews.org/aronline, 16, 417-442.  
Lartey, A. (2008). Maternal and child nutrition in Sub-Saharan Africa: challenges and 
interventions. Proc Nutr Soc, 67(1), 105-108. doi:10.1017/S0029665108006083 
Lee, S. E., Talegawkar, S. A., Merialdi, M., & Caulfield, L. E. (2013). Dietary intakes of 
women during pregnancy in low- and middle-income countries. Public Health 
Nutrition, 16(8), 1340-1353.  
Leroy, J. L., Razak, A. A., & Habicht, J. P. (2008). Only children of the head of 
household benefit from increased household food diversity in northern Ghana. J 
Nutr, 138(11), 2258-2263. doi:10.3945/jn.108.092437 
Levay, A. V., Mumtaz, Z., Rashid, S. F., & Willows, N. (2013a). Influence of gender 
roles and rising food prices on poor, pregnant women's eating and food 
provisioning practices in Dhaka, Bangladesh. Reproductive Health, 10(53).  
 225 
University of Ghana http://ugspace.ug.edu.gh
Levay, A. V., Mumtaz, Z., Rashid, S. F., & Willows, N. (2013b). Influence of gender 
roles and rising food prices on poor, pregnant women’s eating and food 
provisioning practices in Dhaka, Bangladesh. Reproductive Health, 10(53).  
Lieb, S. (1991). Principles of adult learning, Phoenix, AZ: Vision – South Mountain 
Community College, http://honolulu.hawaii.edu/intranet/committees/Fa 
cDevCom/ guidebk/teachtip/adults-2.htm (& 1. htm).  
Lincoln, Y. S., & Guba, E. G. (1985). Naturalistic Enquiry. Newbury Park, CA: Sage 
Publications. 
Liu, N., Mao, L., Sun, X., Liu, L. i., Yao, P., & Chen, B. (2009). Effects of health and 
nutrition education intervention on women's postpartum beliefs and practices: a 
randomised controlled trial. BMC Public Health, 9(45).  
Loewenson, R., Laurell, A. C., Hogstedt, C., D'Ambruoso, L., & Shroff, Z. (2014). 
Participatory action research in health systems: a methods reader: TARSC, 
AHPSR, WHO, IDRC Canada, EQUINET, Harare. 
MacDonald, C., Neequaye, M., Namarika, R., & Yiannakis, M. (2009). Anemia - can its 
widespread prevalence among women in developing countries be impacted? A 
case study: Effectiveness of a large-scale, integrated, multiple-intervention 
nutrition program on decreasing anemia in Ghanaian & Malawian women. 
Retrieved from https://www.researchgate.net/publication/289736454_ 
Martin-Prevel, Y., Arimond, M., Allemand, P., Wiesmann, D., Ballard, T. J., Deitchler, 
M., . . . Moursi, M. (2017). Development of a Dichotomous Indicator for 
Population-Level Assessment of Dietary Diversity in Women of Reproductive 
Age. Current Development in Nutrition.  
Martínez Pérez, G., & Pascual García, A. (2013). Nutritional Taboos among the Fullas in 
Upper River Region, The Gambia. Journal of Anthropology, 2013, 1-9. 
doi:10.1155/2013/873612 
McAlister, A. L., Perry, C., & Parcel, G. (2008). How individual, environment and 
health interact: The social cognition cognition theory. 
MOFA. (2010). Medium Term Agriculture Sector Investment Plan (METASIP)-2009-
2015. Retrieved from Accra, Ghana: http://agricinghana.com/wp-
content/uploads/2013/02/medium-term-agriculture-sector-investment-plan-
ghana.pdf 
Mohammed, A. S., Solikua, O., & Mansega, B. C. K. (2013). Challenges of Women 
Contributing to Food Security in Ghana: Perspectives of Women Farmers in 
Northern Ghana. Sociology Study, 3(11), 835 - 864.  
Mohammed, A. S., Solikua, O., & Mansega, B. C. K. (2013 ). Challenges of Women 
Contributing to Food Security in Ghana: Perspectives of Women Farmers in 
Northern Ghana. Sociology Study, 3(11), 835-846.  
Monterrosa, E. C. (2017). The socio-cultural drivers of food choices: Formative research 
among pregnant and lactating women i Rajasthan. Sight and Life, 3(2).  
Moore, F. G., Audrey, S., Barker, M., Bond, L., Bonell, C., Hardeman, W., . . . Baird, J. 
(2015). Process evaluation of complex interventions: Medical Research Council 
guidance. BMJ, 350.  
 226 
University of Ghana http://ugspace.ug.edu.gh
Nana, A., & Zema, T. (2018). Dietary practices and associated factors during pregnancy 
in northwestern Ethiopia. BMC Pregnancy and Childbirth, 18(183).  
Ngowi, H. A., Carabin, H., Kassuku, A. A., Mlozi, M. R. S., Mlangwa, J. E. D., & III, A. 
L. W. (2008). A health-education intervention trial to reduce porcine 
cysticercosis in Mbulu District, Tanzania. Preventive Veterinary Medicine, 85(1-
2), 52-67.  
Niloufer, S. A., Syed, I. A., & Rahat, N. (2003). Women's Beliefs Regarding  Food 
Restrictions During Common Childhood Illnesses: A hospital based study.  
Nti, C. A., Inkumsah, D., & Fleischer, G. (1999). Influence of women’s workload on 
their nutritional status in selected communities in Ghana. Journal of Consumer 
Studies & Home Economics, 23(3), 165-170. doi:10.1046/j.1365-
2737.1999.00106.x 
Nti, C. A., Larweh, P. M., & Gyemfua-Yeboah, Y. (2002). Food consumption patterns, 
dietary quality and health status of expectant mothers: case studies in suburban 
and rural communities in Ghana. International Journal of Consumer Studies, 
26(1), 7-14. doi:10.1046/j.1470-6431.2002.00211.x 
Ogunjuyigbe, P. O., & Ojofetimi, E. O. (2006). Culture and feeding practices: Major 
underlying causes of childhood malnutrition in developing countries  
Oluwafolahan, O. S., Catherine, A. B., & Olubukunola, A. J. (2014). Dietary habits of 
pregnant women in Ogun-East Senatorial Zone, Ogun State, Nigeria: A 
comparative study. International Journal of Nutrition and Metabolism, 6(4), 42-
49. doi:10.5897/ijnam2014.0170 
Otoo, G., & Adam, Y. (2016). Effects of nutrition education with an emphasis on 
consumption of iron-rich foods on hemoglobin levels of pregnant women in 
Ghana. The FASEB Journal, 30(1).  
Patton, M. Q. (2002). Qualitative research and evaluation methods (3rd Edition ed.). 
Thoudand Oaks. London. New Delhi: Sage Publications. 
Pope, C., Ziebland, S., & Mays, N. (2000). Analysing Qualitative Data: In Qualitative 
Research in Health-care. BMJ Clinical Research.  
Pradeillis, R., Rousham, E. K., Norris, S. A., Kesten, J. M., & Griffiths, P. L. (2016). 
Community readiness for adolesccents' overweight and obesity prevention is low 
in urban South Africaa: a case study. BMC Public Health, 16(763).  
Prochaska, J. O., Colleen, C. A., & Evers, k. E. (2008). Transtheoretical model and 
stages of change. 
Puoane, T., Matwa, P., Bradley, H., & Hughes, G. (2006). Socio-cultural Factors 
Influencing Food Consumption Patterns in the Black African Population in an 
Urban Township in South Africa Ecology, Culture, Nutrition, Health and 
Disease, 14, 89-93.  
Quaye, W. (2008). Food security situation in northern Ghana, coping strategies and 
related constraints. African Journal of Agricultural Research, 3(5), 334-342.  
Ramakrishnan, U. (2002). Prevalence of Micronutrient Malnutrition Worldwide. 
International Life Sciences Institute.  
Ransom, E. I., & Elder, L. K. (2003). Nutrition of Women and Adolescent Girls: Why it 
matters. Population Reference Bureau.  
 227 
University of Ghana http://ugspace.ug.edu.gh
Russ, S. J., Sevdalis, N., Moorthy, K., Mayer, E. K., Rout, S., Caris, J., . . . Darzi, A. 
(2014). A Qualitative Evaluation of the Barriers and Facilitators Toward 
Implementation of the WHO Surgical Safety Checklist Across Hospitals in 
England. Annal of Surgery, 00(00).  
Saaka, Oladele, J., Larbi, A., & Zeledon, I. H.-. (2017). Household food insecurity, 
coping strategies, and nutritional status of pregnant women in rural areas of 
Northern Ghana Food Science Nutrition, 5, 1154-1162.  
Saaka, M. (2012). Maternal Dietary Diversity and Infant Outcome of Pregnant Women 
in Northern Ghana. International Journal of Child Health and Nutrition, 1, 148-
156.  
Sanjukta, S., & Rai, A. K. (2016). Production of bioactive peptides during soybean 
fermentation and their potential health benefits Trends in Food Science & 
Technology, 50, 1-10.  
Savin-Baden, M., & Major, C. H. (2013a). Qualitative research: Essential guide to 
theory and practice. USA and Canada: Routiedge. 
Savin-Baden, M., & Major, C. H. (2013b). Qualitative Research: The essential guide to 
theory and practice. USA and Canada: Routiedge. 
Seebens, H. (2010). Intra-household bargaining, gender roles in agriculture and how to 
promote welfare enhancing changes:ESA Working Paper No. 11-10. Retrieved 
from  
Senah, K. (2003). Maternal Mortality  in Ghana: The other side. Research Review, 19(1), 
47-55.  
Seshadri, S. (2001). Prevalence of micronutrient deficiency particularly of iron, zinc and 
folic acid in pregnant women in South East Asia. British Journal of Nutrition, 85.  
Shenton, A. K. (2004). Strategies for ensuring trustworthiness in qualitative research 
projects. Education for information. IOS Press, 22, 63-75.  
Sliwa, S., Goldberg, J., Clark, V., Collins, J., Edwards, R., Hytt, R. m., . . . Economos, C. 
D. (2011). Using the Community Readiness Model to Select Communities for a 
Community -Wide Obesity Prevention Intervention. Public Health Research, 
Practice and Policy, 8(6).  
Sloane, D. C., Diamant, A. L., Lewis, L. B., Yancey, A. K., Flynn, G., Nascimento, L. 
M., . . . Cousineau, M. R. (2003). Improving the nutritional resource envirnment 
for healthy living through community-based participatory research. JGIM, 18.  
Stanley, L. R. (2014). Community Readiness for Community Change: Tri-Ethnic Centre 
Community Readiness Handbook (2nd edition). 
Stok, F. M., Renner, B., Hoffmann, S., Volkert, D., Boeing, H., & Ensenauer, R. (2016). 
The DONE framework: Creating, evaluating and updating of an interdisciplinary, 
dynamic framework 2.0 of determinants of nutrition and eating. PLoS One, 12(2).  
Tacoli, C., Burkhari, B., & Fisher, S. (2013). Urban Poverty: Food security and climate 
change. Human Settlement Groups,. International Institute for Environment and 
Development (IIED).  
Tchum, S. K., Newton, S., Tanumihardjo, S. A., Fareed, K. N. A., Tetteh, A., & Owusu-
Agyei1, S. (2009). Evalution of a green leavy vegetable intervention in Ghanaian 
Postpartum mothers. AJFAND, 9(6).  
 228 
University of Ghana http://ugspace.ug.edu.gh
Tong, A., Sainbury, P., & Craig, J. (2007). Consolidated criteria for reporting qualitative 
research (COREQ): A 32-item checklist for interviews and focus groups. 
International journal of quality in health care, 19, 349-357.  
Tull, K. (2017). Behaviour-change communication on health related issues (part one).  
Underwood, B., & Galal, O. (2004). Human Nutrition- An Overview, In the Role of 
Food, Agriculture, Forestry and Fisheries in Human Nutrition. Paris, France 
(http://www.eolss.net): Encyclopedia Of Life Support System (EOLSS). 
United Nations. (1990). Women and Nutrition-Nutrition policy discussion papaer No.6. 
Retrieved from 
https://www.unscn.org/layout/modules/resources/files/Policy_paper_No_6.pdf 
USAID. (2012). Maternal dietary diversity and the implications for children’s diets in the 
context of food security. USAID's Infant and Young Child Nutrition Project. 
Retrieved from http://iycn.wpengine.netdna-
cdn.com/files/IYCN_Brief_Maternal_Dietary_Diversity0112.pdf 
USAID. (2017). EFFECTIVE AT-SCALE NUTRITION SOCIAL AND BEHAVIOR 
CHANGE COMMUNICATION: TECHNICAL GUIDANCE BRIEF. Retrieved 
from https://www.usaid.gov/what-we-do/global-health/nutrition/technical-
areas/effective-scale-nutrition-social-and-behavior 
Warren, R. L. (1978). The community in America. Chicago: Rand McNally. 
WFP. (2005). Manual Measuring and Interpreting Malnutrition and Mortality Retrieved 
from https://www.unhcr.org/45f6abc92.pdf 
WFP. (2012). Comprehensive food security and vulnerability analysis Ghana: Focus on 
Northern Ghana. Retrieved from 
https://documents.wfp.org/stellent/groups/public/documents/ena/wfp257011.pdf?
iframe 
White, H., & Sabarwal, S. (2014). Quasi-experimental Design and Methods (Vol. 8). 
Florence: UNICEF Office of Research,. 
Whitehead, A. (1999). Poverty in Upper East Region. A report by ESCOR, ODA, 
London.  
Whitehead, T. (1984). Cultural systems and cultural meaning: Class and cultural factors 
underlying food habitsin a southern US community. In M. Douglas, ea., Food 
and the social order. New York: Russell Sage Foundation. 
Workicho, A., Belachew, T., Feyissa, G. T., Wondafrash, B., Lachat, C., Verstraeten, R., 
& Kolsteren, P. (2016). Household dietary diversity and Animal Source Food 
consumption in Ethiopia: evidence from the 2011 Welfare Monitoring Survey. 
BMC Public Health, 16(1192).  
Zereyesus, Y. A., Ross, K. L., Amanor-Boadu, V., & Dalton, T. J. (2014). Baseline Feed 
the Future Indicators for Northern Ghana 2012. Retrieved from Kansas State 
University: https://pdf.usaid.gov/pdf_docs/pnaed041.pdf 
  
 229 
University of Ghana http://ugspace.ug.edu.gh
APPENDICES 
Appendix A: Key Informant Interview Guide on the Food System  
Target respondents will include District agricultural extension officer, men and women 
farmers, community elders, District nutrition officer, community volunteer  
Respondent’s /ID…………… 
A: Participants’ socio-demographic background  
Name of Community ………………          Category of respondent……………. 
Respondent’s sex………………………….      Age………….. Religion…………. 
Educational level completed…………………Marital status ………. 
Occupation…………………………. 
Place of work…………………………. 
Size of household…………………………. 
Respondent’s average weekly/monthly income…………………………. 
B: Understanding the food system  
Food production 
1. What foods are produced in this community?  
(Probe for production of crops and animals across different food groups: cereals, 
vegetables, roots and tubers, fruits, fish, poultry, livestock, etc.). Use table below 
to ask further questions on food items mentioned and their characteristics. 
Free list record form for foods produced in community and characteristics 
Traditional Food group Varieti Seasonality (all Time Cultiva Part of Frequency Perception of 
food (grain/cereal, es  year round or some of year ted or local of use nutritive/health 
item/food meat/fish, parts of the year) harvest wild diet? value or quality 
name vegetable, fruit, ed of foods 
etc.  
         
 
 
 
2. For foods which do not form part of the local diet, what happens to them when 
they are harvested? 
 230 
University of Ghana http://ugspace.ug.edu.gh
3. Are there foods which form part of your local diets but which are not produced in 
this community? (Also probe for food items such as sugar, magi and skimmed 
milk)  
4. What are the sources of these other foods?  
5. What resources are required for the production of these foods?  
Probe for:  
a) Inputs such as seeds, fertilizers and implements 
b) Land; how is land accessed and controlled? 
c) Labour;- what is your source of labour for farm work? Hired, household 
members’ labour members? What activities are performed by men and 
women?  
d) Are there any crops and animals produced by men or women? If so, why is it 
so? 
Food processing, preservation, storage and distribution 
6. When crops are produced/harvested from the farm or animals mature, what 
happens to them? Probe for:  
a) Processing techniques/methods 
b) Food storage, food barns and other facilities 
c) Marketing/distribution: are there foods that are produced mainly for cash? 
Where are they marketed and who does the marketing? 
d) Are some of the foods given out as gifts? If so who are the beneficiaries? 
e) What happens to the animals reared in this community when they are mature? 
Food Consumption and preparation 
7. In a typical home, how is food prepared for consumption? Probe for: 
a) Who prepares the food and why? 
b) When is food prepared and how is it shared among members of households 
(who is served fast and who is served last, which people eat together and who 
eat alone and why? Who shares the food when it is prepared and why? 
Food security 
8. How easy or difficult is it to have enough food to be eaten by everyone in the 
community?   
a) Food availability: Is food available all year round? In what time of the year is 
food abundant in this community? When is food scarce in the community? How 
does scarcity of food affect: 1) household diets 2) women in this community? 
b) How do people cope during times of food shortages in the community? 
c) What is your perception of the cost of food?  Probe:  
i) During harvest?  
ii) After harvesting and other periods of the year? 
d) What foods do community members have preference for or like? Why are these 
foods preferred to others?  
e) Are there foods community members do not generally like but eat anyway? What 
is it about the foods that make people dislike them? When are these foods usually 
consumed? 
 231 
University of Ghana http://ugspace.ug.edu.gh
f) How do community members get the money to acquire food? (Probe for different 
sources of income to acquire food). 
 
 
 
 
 232 
University of Ghana http://ugspace.ug.edu.gh
Appendix B: Key informant and FGD interview guide on community perceptions of 
intervention activities 
Target participants: Participants in the intervention 
1). What were some of the messages you heard about? 
2) Which of the messages have you been able to practise? 
3) Which messages have you not practised and why? 
4) What are the benefits of the intervention to the community? 
5) What are some of the challenges you had with the intervention? 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 233 
University of Ghana http://ugspace.ug.edu.gh
Appendix C: Assessment of community readiness to address women’s dietary 
challenges 
Target respondents will include chief, Assemblyman, men and women leaders, 
social workers, NGO representative in community, teachers, healthcare workers, 
volunteers and community members 
 
Background Characteristics 
Category of respondent……..      Sex…      Respondent’s Age…… 
Educational level completed       Marital status……………..Size of household…. 
Religion……   Occupation……  
Respondent’s average weekly/monthly income 
1. Knowledge about the issue of women’s poor dietary quality 
a. What do members of the community think about women’s diets? (Probe: Are 
women’s diets adequate in terms of nutrients and quantity? Please explain your 
response) 
b. What do you think are the causes and consequences of women’s poor dietary 
quality? 
c. How does women’s poor dietary quality affect them and the community at large? 
 
2. Community efforts to addressing women’s poor dietary problem 
d. What activities/programmes are in place to address challenges to women’s diets 
in the community? (health facilities and community tradition?) 
e. What activities are carried out by community volunteers to address the issue?  
f. Are these activities effective in managing women’s dietary problems? 
g.  Are these activities community-wide such that all segments of the community 
are reached? 
 
3. Leadership 
d. Are community leaders in support of the activities? 
e. How is this support shown? (What are they doing to support the activities?) 
 
4. Community climate 
a) How do community members feel about the issue? (Probe: Do they think they 
can do something about the problem?) 
b) If they can, what do they think they can they do about the problem? 
c) Do they think others can help solve the problem? (Probe for who can help?) 
 
 234 
University of Ghana http://ugspace.ug.edu.gh
5. Resources related to the issue 
a) What resources are available to support intervention activities addressing the 
problem of women’s diets? (Probe for people who have the time to be trained to 
deliver BC messages, money to support activities, existing programmes to tag on 
BC activities child welfare activities, social and religious programmes, etc) 
 235 
University of Ghana http://ugspace.ug.edu.gh
Appendix D:  Qualitative 24-hour recall guide  
Target respondents: Women of reproductive age 
Respondent ID 
A. Participant’s socio-demographic characteristics  
Community …. 
Respondent’s     Age………….. Religion……………… 
Physiological status……………… 
Educational level completed……………………………… 
Current primary occupation……………………………… 
Place of work……………………………… 
Size of household ……………………………… 
Respondent’s average weekly/monthly income……………………………… 
B. Women’s Time use 
Tell me about all the activities you carried out yesterday from the time you wake up until 
the time you went to bed at night.  
C. 24-hour recall of foods consumed 
Now, I will like us to discuss all the foods you ate yesterday from the time you first 
woke up until the time you went back to sleep at night. After getting all the foods you ate 
yesterday, we will then go over each of them to talk about the time they were eaten, 
whether they were bought or home prepared, how they were prepared and the quantity 
you ate.  
1. What was the first food you ate when you woke up yesterday? 
2. What time did you consume this food? 
(Go through respondent’s eating activities for yesterday and ask for all meals, drinks, 
snacks and other foods she ate for the rest of the day till she went to bed at night).  
Name of Description of dish Time of day Purchased If home prepared, If Amount 
Food/meal (ingredients etc.) consumed  or Home how is this food purchased consumed 
prepared  prepared?  how much (using 
household 
measures) 
       
4. Was food ……..(insert name of food) purchased or home prepared? 
 236 
University of Ghana http://ugspace.ug.edu.gh
5. Kindly tell me what ingredients were used in preparing the food…….(insert 
name of food)? 
6. How is food…..(insert name of food) prepared.  
7. Who else consumed (insert name of food) with you in the household?  
 
 
 237 
University of Ghana http://ugspace.ug.edu.gh
Appendix E: In-depth interview guide on food ownership and utilization at the 
household level 
Target respondents: Women of reproductive age 
A: Sources of food acquisition and decision-making regarding household food 
1. Where do you usually acquire the above-listed foods from? (Refer to the 24-hour 
recall by each respondent. For each of the foods/meals mentioned, probe for its  source 
to the point of production) 
2. How are decisions made about food consumed at home and why? (Probe for whose 
responsibility it is to provide food items for the home, role of men and women, husbands 
and wives in feeding the household) 
3. Who is responsible for preparing food for the family and why? 
5. If you have to purchase the ingredients for preparing food, who goes to purchase 
them?  
6. How are you able to buy food? (Probe for access to resources/money to buy food?) 
7. Apart from buying food, how else are you able to access food? 
B Food distribution/sharing and eating schedules at the household level 
1. How is food shared for the household after it has been prepared?  
(Probe for: a). Who is (are) responsible for sharing the main dish/meal in the 
household after it has been prepared? Does a different person apart from the one 
who prepared the main dish share the dish? (Probe for reasons) 
a) Describe what happens during meal times, e.g. eating together, etc.? 
(Probe for: who eats with who in the household? Women and men: similarities 
and differences in what they eat?) 
b) When the family eats together, 
a)  Who is served first when the dish is prepared? Why is this person or 
persons served first? b) Who is served last and why? 
C) Which people eat together from the same bowl and why? Which people 
eat separately and why? 
c) Are there times when you eat away from home? When are these times and 
what will you normally consume? What are the sources of the foods? 
 238 
University of Ghana http://ugspace.ug.edu.gh
Appendix F: Focus group discussion guide (Community men and women) 
A. Demographic Characteristics 
ID Age Marita No. of HH Physio. Educat Occup Average Religion 
No. l status children size ion weekly/m
onthly 
income 
01          
          
 
B. Community knowledge, attitudes and beliefs about food and nutrition and health 
Food choices and preference 
1. What are the foods/meals people like to eat in this community? (Probe for 
different kinds of food and explain why these foods are preferred) 
2. What foods do members of the community not like? (Probe for all foods not liked 
and discuss the reasons).  
3. Are there foods members of the community like but do not eat? List them and 
explain why you do not eat them. (Probe: Are there foods pregnant or lactating or 
children like but do not eat them because of their physiological stage or gender? 
Why can’t they eat such foods?). 
4. Are there foods the community does not like but eat them? (Probe for these foods 
and discuss reasons) 
5. Are there foods pregnant and lactating women do not like but are made to eat 
because they are pregnant/lactating?  
6. What about children and the aged? 
Food taboos, beliefs and norms 
7. Vignette 
In almost every community, there are food taboos different segments of the 
population adhere to. For instance, in upper East region, women are not allowed 
to eat fowls, pregnant women are not allowed to eat groundnut or groundnut soup 
for fear they may not be easily delivered of their babies and children are not 
allowed to eat eggs for fear they might end up becoming thieves in the future. 
a) What is the situation in this community? Are there certain foods in this 
community (in the past or present) that some categories of persons are not 
allowed to eat? (Probe for pregnancy, lactation, adolescents, the aged, children 
and the sick. explain the nature of these taboos. What are the reasons for the  
taboos?) 
 239 
University of Ghana http://ugspace.ug.edu.gh
b) Are these taboos being practised currently? Which ones are practised and 
which category of people practise them and why? Why are the others taboos no 
more practised? 
Health and nutrition 
8. Vignette: 
Our health workers recommend that a pregnant woman should visit the midwife 
at least four times before she gives birth. This measure is to ensure that the 
woman will be healthy and the baby will be healthy at birth and also grow well. 
They also recommend that pregnant and lactating mothers eat well to enhance 
their health and that of their children. 
a) Tell me about what foods women need to enhance their health. (Probe for: 
pregnant women, lactating women, adolescent girls and other categories of 
women of reproductive age.  Further probe for what they have been told at 
the health facility and what is traditionally known or local knowledge) 
b) Tell me about foods that are healthy. What are the attributes of healthy foods? 
c) List such foods and perceptions of their health and nutritive values (Probe for 
what they learnt from the health facility and what they learnt from other 
people in the community and what tradition has handed down to them,  
d) Are some of these foods produced in the community? List them. How else 
can they be accessed? 
 
 
 
 
 
 
 
 
 
 
 240 
University of Ghana http://ugspace.ug.edu.gh
Appendix G: Observations of women’s time use (household level: from 6am to 6pm) 
Target Respondents: Women of reproductive age 
Respondent’s ID……………….. 
Background  Characteristics 
Community……  
Respondents Age………….. Physiological status…….. 
Educational level completed……………………………… 
Current primary occupation……………………………… 
Place of work……………………………… 
Size of household ……………………………… 
Respondent’s average weekly/monthly income……………………………… 
Items and Activities to observe 
 Time and duration of observation (Start and end time) 
 Number of people in household 
 Physical environment of household (hygiene and sanitation, etc.) 
 Types of foods consumed in the household 
 Cooking processes and people involved in cooking 
 Food items used in cooking and sources of each food item 
 Food distribution/sharing among members of households 
 Eating schedules and arrangements 
 Social interactions during eating 
 Men’s role in food preparation/ distribution/sharing processes 
 Time spent by women doing household chores and commercial activities 
 Women’s eating patterns: frequency of eating, time of eating types of food 
consumed, etc.  
 Time spent resting during the day 
 
 
 
 
 
 
 
 
 241 
University of Ghana http://ugspace.ug.edu.gh
Appendix H: Observational Checklist (Community level) 
Activities to observe and interact with: 
 Number of compounds in community 
 Transportation and communication channels and patterns 
 Livelihood activities in community 
 Men’s daily activities 
 Women’s daily activities 
 Other household members’ daily activities 
 Availability of the ff. facilities: educational, health, markets, water sources, 
sanitation, recreational, nutrition programmes especially for women 
 Presence of other organizations -e.g. NGOs 
 Observation of ceremonial activities such as market days, funeral, 
outdooring/naming/ marriage and gender-based roles in these events and food use 
 Farming, processing, food storage facilities such as food barns, food distribution 
channels, food vending activities, etc. 
 
 
 242 
University of Ghana http://ugspace.ug.edu.gh
Appendix I: Participant’s Consent Form 
Study title: A formative study of community perceptions and strategies to improve 
women’s dietary quality in Savelgu-Nanton District, Ghana 
Student Investigator: Sawudatu Zakariah 
Address: School of Public Health, Post Office Box 581, University of Ghana, Legon 
Introduction 
My name is Sawudatu Zakariah, a student of School of Public Health, University of 
Ghana. I am undertaking this research as part of the requirements for the award of a PhD 
Degree in Public Health. To be sure that you understand what you are getting yourself 
involved in, I have summarized the purpose of this study to make it easy for you to 
understand and to make an informed choice about either taking part or not taking part. If 
after the explanation you still feel uncomfortable about participating, you still have the 
right not to be involved. However, if you agree to be part of this research, then you are 
required to sign or thumbprint this agreement for us. 
General information about Research 
This research is about women’s eating practices and how our beliefs about food 
influence what they eat. In Ghana, women’s diets in particular are influenced by what 
people know and understand about food and its effect on their health so some categories 
of women are restricted from consuming certain foods which are considered taboos or 
harmful to their health. Because of this restriction, women’s diets are not well balanced 
(limited in quality and variety) - leading to poor nutritional statuses. We, therefore, need 
solutions that will address our beliefs and practices that affect our women’s diets 
negatively. This study is aimed at identifying these social and cultural barriers and 
bringing the community together to find ways of modifying these practices in order to 
improve women’s diets.  
If you agree to be part of this study, then I will take about one hour of your time to 
discuss some of the issues. 
Possible risks:  
This study is a minimal risk study. Apart from spending some time discussing these 
issues, this study does not pose any risks to participants.  
Possible benefits 
There are no immediate rewards from this study for participation. However, in the long 
run, the findings will help inform policies on more appropriate solutions for our health. 
 
 243 
University of Ghana http://ugspace.ug.edu.gh
Confidentiality 
The information you give us will be kept confidential. No names of participants will be 
seen in our reports and the data will be used solely for the purpose for which it is 
collected. 
Contact for additional information 
If you need further information or persons to contact concerning this research, 
kindly contact: 
Sawudatu Zakariah 
School of Public Health, 
University of Ghana, Legon 
Cell phone number: 0268628793 or email: sodeyakoto@gmail.com 
If you agree to be part of this study, please, sign or thump print. 
Participant:………………………………                   
Researcher……………………………... 
 
 
 244 
University of Ghana http://ugspace.ug.edu.gh
Appendix J: Child Assent Form 
Introduction 
My name is Sawudatu Zakariah and I am from the School of Public Health at 
University of Ghana.  I am conducting a research study entitled “ A formative study of 
community perceptions and strategies to improve women’s dietary quality in Savelgu-
Nanton District, Ghana”.  I am asking you to take part in this research study because I 
am trying to learn more about our community norms that prevent women and young girls 
attaining optimal diets.  This will take about one hour of your time. 
General Information 
This research is about dietary practices of women who are in their child-bearing age and 
how our beliefs and usual practices influence what women eat. In Ghana, women’s diets 
in particular are influenced by what people know about food and its relationship to their 
health so some categories of women are restricted from consuming certain foods which 
are considered taboos or harmful to their health. Because of this restriction, women’s 
foods are not well balanced (limited in quality and variety) - leading to poor nutritional 
statuses. We, therefore, need solutions that will address our beliefs and practices that 
affect our women’s diets negatively. This study is aimed at identifying these social and 
cultural barriers and bringing the community together to find ways of modifying these 
practices in order to improve women’s diets.  
If you agree to be in this study, you will be asked to participate in group discussions and 
one-on-one interviews to discuss these issues.  
Possible Benefits 
Your participation in this study has no immediate rewards. However, in the long run, the 
findings will help inform policies on more appropriate solutions for our health.  
Possible Risks and Discomforts 
Apart from spending about an hour of your time discussing issues, this study does not 
pose any risks to you.  
Voluntary Participation and Right to Leave the Research 
You can stop participating at any time if you feel uncomfortable. No one will be angry 
with you if you do not want to participate. 
Confidentiality 
Your information will be kept confidential. No one will be able to know how you 
responded to the questions and your information will be anonymous. 
 
 245 
University of Ghana http://ugspace.ug.edu.gh
Contacts for Additional Information 
You may ask me any questions about this study.  You can call me at any time [insert 
contact information] or talk to me the next time you see me. 
Please talk about this study with your parents before you decide whether or not to 
participate.  I will also ask permission from your parents before you are enrolled into the 
study.  Even if your parents say “yes”, you can still decide not to participate.   
             Your rights as a Participant 
This research has been reviewed and approved by the Institutional Review Board of 
Noguchi Memorial Institute for Medical Research (NMIMR-IRB).  If you have any 
questions about your rights as a research participant youcan contact the IRB Office 
between the hours of 8am-5pm through the landline 0302916438 or email addresses: 
nirb@noguchi.mimcom.org  
VOLUNTARY AGREEMENT 
By making a mark or thumb printing below, it means that you understand and know the 
issues concerning this research study. If you do not want to participate in this study, 
please do not sign this assent form. You and your parents will be given a copy of this 
form after you have signed it. 
This assent form which describes the benefits, risks and procedures for the research titled 
“ A formative study of community perceptions and strategies to improve women’s 
dietary quality in Savelgu-Nanton District, Ghana” has been read and or explained to 
me. I have been given an opportunity to have any questions about the research answered 
to my satisfaction.  I agree to participate.  
Child’s Name:………………………   Researcher’s Name:……………………… 
Child’s Mark/Thumbprint…………   Researcher’s Signature:…………… 
Date: ………………………………  Date: ………………………… 
 
 
 
 246 
University of Ghana http://ugspace.ug.edu.gh
Appendix K: Parental Consent Form 
Study title: A formative study of community perceptions and strategies to improve 
women’s dietary quality in Savelgu-Nanton District, Ghana 
Student Investigator: Sawudatu Zakariah 
Address: School of Public Health, Post Office Box 581, University of Ghana, Legon 
Introduction 
My name is Sawudatu Zakariah, a student of the school of Public Health, University of 
Ghana. I am undertaking this research as part of the requirements for the award of a PhD 
Degree in Public Health. To be sure that you understand what your child is getting 
herself involved in, I have summarized the purpose of this study to make it easy for you 
and your child to understand it and to make an informed choice about either taking part 
or not taking part in the study. If after reading it (or it being explained to you) you do not 
understand anything, you may ask for more explanations. If after the explanation you 
still feel uncomfortable about your child participating, you still have the right not to 
involve her. However, if you agree for your child to be part of this research, then you are 
required to sign or thump print this agreement on her behalf. 
General information about Research 
This research is about dietary practices of women who are in their child-bearing age 
including young girls like your child who has not yet given birth but is in the process of 
getting to the stage of motherhood. In Ghana, women’s diets in particular are influenced 
by what people know and understand about food and how it affects their health so some 
categories of women are restricted from consuming certain foods which are considered 
taboos or harmful to their health. Because of this restriction, women’s diets are not well 
balanced (limited in quality) - resulting in poor nutritional status. We, therefore, need 
solutions that will address our social practices that affect our women’s diets negatively. 
This study aims to identify these social and cultural barriers to women’s optimal diets 
and bring the community together to find ways of modifying these practices in order to 
improve women’s diets. 
If you agree for your child to be part of this study, then I will take about one hour of her 
time to discuss some of the issues. 
 247 
University of Ghana http://ugspace.ug.edu.gh
Possible risks 
A part from spending some time discussing these issues, this study does not pose any 
risks to participants. 
Possible benefits: There are no immediate rewards from this study for participation. 
However, in the long run, the findings will help inform policies on more appropriate 
solutions to our nutritional problems. 
Confidentiality 
The information she gives us will be kept confidential. No names of participants will be 
seen in our reports and the data will be used solely for the purpose for which it is 
collected. 
Contact for additional information: If you need further information or persons to 
contact concerning this research, kindly contact: 
Sawudatu Zakariah, School of Public Health, University of Ghana, Legon. Cell 
phone number: 0268628793 or email address: sodeyakoto@gmail.com 
If you agree for your child to be part of this study, please, sign or thump print. 
Your child’s rights as a participant 
This research has been reviewed and approved by the Noguchi Memorial Institute for 
Medical Research Institutional Review Board (NMIMR-IRB). If you have any questions 
about your child’s rights as a research participant you can contact the IRB Office 
between the hours of 8am and 5pm through the landline 0302916438 or email address: 
nirb@noguchi.ug.edu.gh 
VOLUNTEER AGREEMENT  
The above document describing the objective and purpose of the study, the benefits and 
risks as well as the procedures of the research titled “a formative study of community 
perceptions and strategies to improve women’s dietary quality in Savelgu-Nanton 
District, Ghana” has been read and explained to me. I have been given the opportunity to 
have any questions about the research answered to my satisfaction. I agree that my child 
should participate as a volunteer. 
 248 
University of Ghana http://ugspace.ug.edu.gh
Date ………………………………                ……………………………...    
Parent/guardian’s name and 
signature/thump print  
If volunteers cannot read the form themselves, a witness must sign here: 
I was present while the objective, purpose, benefits, risks and procedures of the research 
were read to the child’s parent/guardian. All questions were answered and the child’s 
parent/guardian agreed that his/her child should participate in the research.  
Date……………………                      ……………………………...    
Witness’ name and signature/thump print  
I certify that the nature and purpose, the potential benefits and possible risks associated 
with participating in this research have been explained to the above individual. 
Date……………………….              ………………………………………….. 
Name and signature of person who obtained 
consent 
 
 
 
 
 
 
 249 
University of Ghana http://ugspace.ug.edu.gh
Appendix L: Intended Action plan for implementing attitudinal and behaviour change to improving women’s diets 
Week/Activities Audience Channels Content of act. Durations  Responsibili Expected outcomes/results 
ty 
Week 1 District nutrition officers, Stakeholders Baseline results and Two days PI Consensus on challenges facing 
Consensus community health volunteers, meeting designing women’s diets 
meeting with  extension officer, opinion intervention)  
community leaders, personnel in women Design of intervention messages 
stakeholders in agriculture, NGO reps. in to address challenges  
community 
Week 1 Whole community Community  Sensitization of Two hours Implementat Support and participation of 
Comm.  Durbar comm. about ion Team community in BCC activities 
sensitization  intervention  
-Importance of good sanitation 
Week 2 8 community volunteers, four Training  Training on Two days  PI  Improved understanding and 
Training of research assistants workshop  messages and data ability to disseminate messages 
community collection to target audience 
volunteers and 
research assistants  
Week 3 WRA, older women, 1st Group  Key message 1: One hour Implementat -Understanding the nutritrive 
Meetings with adolescent girls, Husbands, meeting Importance of ion team benefits of foods produced in 
men and women heads of households, young combining and community 
men, fathers, etc. consuming a variety -Need for women to eat a variety 
of foods in meals to of foods 
improve the quality -Need for women to eat a 
of women’s diets combination of foods of good 
qualities in good quantities 
 
-Importance of good sanitation 
Week 4 WRA, older women, 2nd Group Key message 2: One hour Implementat -Importance of women’s diets 
 250 
University of Ghana http://ugspace.ug.edu.gh
Meeting with men Adolescent girls, husbands, meeting Importance of ion team and their reproductive health 
and women heads of households, young women’s diets to development 
men, fathers, etc. their health and  
reproductive health --Understanding the effects of 
outcomes; poor diets on women’s health 
and reproductive outcomes 
 
-Importance of good sanitation 
Week 5 WRA, older women, 3rd Group Key message 3: One hour Implementat  Understanding the importance 
Meetings with adolescent girls, husbands, meeting Modification of ion team of women’s social and economic 
men and women heads of households, young unfounded food status in household food 
men, fathers, etc. beliefs, taboos and provision and consumption 
other food  
discriminatory -Addressing all forms of food 
dietary practices discrimination against women 
against WRA  
 
Week 6 Households/compounds with Visits to Review of messages  Two hours Implementat Addressing all forms of dietary 
1st set of home WRA households/ ion team discrimination against women 
visits compounds  
Week 7 WRA, older women, 4th Group  Key message 4: 0ne hour Implementi Understanding the importance of 
Meeting with men adolescent girls, men in meeting importance of ng team women’s social and economic 
and women community including empowering women status in household food 
husbands, heads of to have control over provisions and consumption 
households, young men, resources to enable 
fathers, etc. them to improve 
their dietary quality 
Week 8 Households /compounds with Visits to  Review of Two hours Implementat -Adoption of behavioural change 
Home visits WRA households/ messages  ion team communication messages 
 compounds 
 251 
University of Ghana http://ugspace.ug.edu.gh
Week 9 WRA, men, older women 5th Group  Key message 5: One hour Implementat  
Meeting with men meeting importance of ion tean Importance of preservation of 
and with women  improving methods food 
of food preservation 
and preparation to 
enhance the quality 
of women’s diets 
Week 10 WRA, older women and Group cooking  Two hours Implementat Learning other ways of using 
Food adolescents Processing of soya ion team  legumes, particularly soya bean 
demonstration bean into flour for at home to improve qualities of 
fortification of diets 
household meals  
 -Learning about other leafy 
vegetables and their consumption  
Week 11 WRA Group cooking Using soya bean in a 2 hours Implementat Learning other ways of using 
Food variety of local ion team legumes, particularly soya bean 
demonstration dishes at home to improve qualities of 
diets 
 
-Learning about other leafy 
vegetables and their consumption 
Week 12 Whole community Community Review of messages 2 hours Implementat Outcome of intervention activity 
Community durbar and sustainability of ion team for the past 12 weeks and way 
sensitization intervention,  forward 
Note: Key message 6 runs through the whole period of the intervention 
 
 
 
 252 
University of Ghana http://ugspace.ug.edu.gh
Appendix M: Implemented action plan for attitudinal and behaviour change to improve women’s diet 
 
Week   Activity/topic/content Audience Channels Durations  Responsibility Outcomes 
Week 1   District nutrition officer, Stakeholders Two days PI Consensus on challenges facing 
 Consensus meeting with community health nurse, meeting   women’s diets 
community community health volunteers,     
stakeholders agric. extension officer,    
Design of intervention messages 
 community opinion leaders,    
    and channels of communication chief and elders, research 
 assistants     
     Selection of community 
     volunteers to partner research 
     team to implement intervention 
 Whole community Community Two hours Stakeholders and  
Comm. sensitization on  durbar  PI 
Support and participation of  
baseline findings and 
intervention community in BCC activities 
 
       
Week 2  Community volunteers, four Training 2 days PI    
Training of community research assistants work shop Improved understanding and 
volunteers and research  ability to disseminate messages 
assistants on messages 
to target audience 
 
       
 
Week 3 :   (1st grp. Meeting) WRA, older women, adolescent  Group One and Implementation Outcome- 
 253 
University of Ghana http://ugspace.ug.edu.gh
 Introductory message girls, discussion half hours team Improved understanding of the 
on importance of Husbands, heads of households, nutritional benefits of foods 
growth and young men, fathers, etc. produced in community and 
development 
their  
 
Week 4:   (2nd grp meeting) WRA, older women, adolescent  Group One and a Implementation Outcome: 
 girls, discussions half hours team Improved understanding of the 
Key message 1: Women  nutritional benefits of foods 
need a variety of foods in Husbands, heads of household, 
produced in community 
their diets for good young men, fathers, etc. 
health and wellbeing   
 -Improved understanding of the 
need for women to eat a variety 
of foods 
 
-Improved understanding of the 
need for women to eat 
combinations of food of good 
qualities in good quantities 
 
Week 5:   (3rd grp meeting) WRA, older women, adolescent Group One and a Implementation Outcome: 
Key message 2: girls, discussions half hours team Improved understanding of the 
Combining and  importance of consuming a 
consuming a variety of Husbands, heads of households, 
variety of foods 
foods in our meals young men, fathers, etc. 
improve the qualities of 
our diets 
 
 254 
University of Ghana http://ugspace.ug.edu.gh
Week 6:   (4th grp meeting) WRA, Older women, adolescent Group One hour Implementation  
Key message 3: girls, husbands, heads of discussions team Outcome: 
Women’s diets are households, young men, fathers, Improved understanding of the 
important for their etc. 
importance of women’s diets 
health and reproductive 
and their reproductive health 
health outcomes 
development 
 
--Improved understanding of 
the impact of poor diets on 
women’s health and 
reproductive outcomes  
Week 7 :   (5th grp meeting) Husbands, heads of households Group On and a Implementation Outcome: 
Key message 4: Address heads, young men, fathers, discussion half hours team Discouraging food-related 
unfounded beliefs and women, adolescent girls discriminatory practices against 
taboos and  
women and children 
discriminatory practices 
to help improve women’s  
diets Discouraging food restrictions 
 against pregnant and lactating 
women 
Week 8:   (6th grp meet) Husbands, heads of household, Group One hour Implementation Reinforcing messages  
 Review of messages 1 young men, fathers, women, discussion  team  
through 4 adolescent girls     
     
 
     
 
     
First Home-visit Compounds in community One-on-one Two Implementation  
discussion hours  team  Adoption of behaviour change 
with 
 255 
University of Ghana http://ugspace.ug.edu.gh
households communication messages 
 
Week 9th    (7th grp meet) WRA, older women, adolescent Group 0ne hour Implementation Outcome:  
Key message 5: girls men in community discussion team Improved women’s access to 
Empower women to have including Husbands, heads of productive resources 
control over resources so households, young men, fathers, 
 
they can improve their etc. 
dietary qualities Improved understanding of the 
 importance of women’s social 
 and economic status in 
household food provision and 
consumption 
 
Improved men’s understanding 
of the need to support women to 
be economically self-sufficient 
Week 10  (8th grp meet) WRA, older women, adolescent Group One hour Implementing Understanding the importance 
Key message 5: girls, men in community discussion team of women’s social and economic 
Empower women to have including husbands, heads of status in household food 
control over resources so households, young men, fathers 
provision and consumption 
they can improve etc. 
qualities of their diets 
 
Week 11  (9th grp meet) All participants Group One hour  Outcome: 
 Review of messages  discussion   Reinforcing messages 
       
  Home visits Two Implementation 
Adoption of behaviour change 
Second Home-visits Compounds in community hours team 
communication messages 
 256 
University of Ghana http://ugspace.ug.edu.gh
Week 12 :(10th grp meet) WRA, older women, adolescent Group One hour Implementation  
Reviewed messages on girls, men in community discussion team  
importance of women’s including husbands, heads of Outcome: 
diets households, young men, fathers, 
Improved understanding of  
etc. 
effects of women’s diets on their 
health and well-being 
Week 13 Sanitation day Whole community Clean-up Two Community  Outcome:  
exercise hours members Improved community sanitation 
Week 14  (11th grp meet) Men and women Group 1 hours Community -Adoption of behaviour change 
11th  Group meeting discussion volunteers communication messages 
Week 15  (12th grp meet) Men and women Group 1 hours Community Learning other ways of using 
12th Group meeting discussion volunteers legumes particularly soya beans 
at home to improve qualities of 
diets 
 
-Learning about other leafy 
vegetables and their 
consumption 
Week 16 Key message 6: Men and women in community Presentation 2 hours Ministry of Outcome: Improved food 
Improving food , video show Agric. extension preservation practices 
preservation and and services 
preparation: Training demonstrati
and demonstration on  on  
Prevention of post- 
harvest losses and 
preservation of farm 
produce 
Week 17 Cooking demonstration Women and men Group 3 hours PI and personnel Improved qualities of foods 
 257 
University of Ghana http://ugspace.ug.edu.gh
cooking from Women in prepared at home by 
Agriculture encouraging use or addition of 
soya bean in all household 
foods,  
prepared, improved cooking of 
vegetables, encouraging the 
consumption of fermented TZ 
 
Week 17 Week17 Whole community Community Two PI Outcome: 
Community Durbar durbar hours Sustainability of intervention 
activities 
 
 
 
 
 
 
 
 
 
 
 258 
University of Ghana http://ugspace.ug.edu.gh
Appendix N: Messages disseminated 
Table 2: BC communication messages addressing challenges to women’s diets 
Issues and Current Communication messages Key message 
challenge behaviour/evidence 
Sub- Women’s diets were 1. The foods we produce are of different kinds; each provides the body with Combining and 
optimal found to be inadequate unique nutrients for energy, protection and health and, last, child growth consuming a variety of 
knowledge in variety and the 24- (when pregnant or breastfeeding): foods in our meals 
and hour recall of women’s Energy foods: cereals, grains and roots/tubers: maize, rice, millet, guinea-corn, improve the quality of 
perceptions diets confirmed this yams, cassava and sweet potato our women’s diets. 
about foods finding Health and Protective foods: green leafy vegetables: ayoyo, bra, kuka, alefu, 
cassava leaves 
Other vegetables: tomato, okro, pepper, garden-egg  
Fruits: mango, shea fruit, dawadawa fruit, ebony fruit (gaya), sinsaba (….), 
pawpaw, water-melon, orange, banana, pineapple  
Growth foods:  
legumes/seeds/nuts: groundnuts, beans, adua, bambara, soya bean, neri, 
kanton, agushie, dawadawa seed 
Meat/fish: goat, sheep, cattle, chickens, guinea fowl, rabbit 
 
2. Consumption of a combination of any of these foods in our meals on a daily 
basis will give us the different nutrients we need to grow and work hard.   
Consume different vegetables and fruits at different meals when in season 
because of the varying amounts of micro-nutrients contained in the different 
vegetables/fruits. 
Both men and women should cultivate a wider variety of foods for household 
consumption. 
 
3. Consume a lot of legumes and anchovies: good sources of iron. 
 259 
University of Ghana http://ugspace.ug.edu.gh
Eat enough to meet energy and nutrient needs. 
 
4. Beverages like tea (without sugar) do not contain energy or nutrients; they 
only fill the stomach. 
 
5. However, not eating a combination of these foods can make us get the 
following health conditions: 
Iron-deficiency anaemia: not enough quantity of iron in the body to make 
blood leading to tiredness, low productivity, birth complications 
IDD (iodine deficiency disorders): Inadequate consumption of iodine causes 
goitre, brain damage to the foetus, dwarfism 
Vitamin A deficiency: eye problems leading to blindness, dry skin, severe hair 
loss, etc. 
2. Poor Even though 6. Adolescent girls need good diets to develop well.  Women’s diets are 
knowledge communities were Focus especially on adolescent girls between the ages of 9 and 19 years. important for their 
about aware poor diets It is the time when they grow fastest and, therefore, their nutrient requirements health and reproductive 
women’s affected women’s are greatest. health outcomes. 
diets and its health, majority were They need to eat adequately (quality and quantity) to be able to grow well and 
relationship not aware it affected prepare for when they are ready to have children. 
to women’s women’s reproductive Therefore, adolescent girls need enough iron-rich foods (e.g. red meat, beans 
reproductiv health outcomes.  and dark-green leafy vegetables) and other vegetables and fruits to protect them 
e health from getting anaemia.  
outcomes  
7. They also need to avoid getting pregnant until they are 20 years and above. 
Avoid early marriage. 
 
8. Girls and women (whether pregnant or lactating) need to eat more energy 
foods, a lot of iron rich-foods to stay active and healthy, to produce healthy 
babies and breastfeed well. 
Women also need a lot of iron-rich foods because of menstruation.  
 
 260 
University of Ghana http://ugspace.ug.edu.gh
9. Well-nourished women are likely to have healthy pregnancies, healthy babies 
and healthy bodies to take care of their children and the community. 
 
10. During pregnancy and lactation, a woman’s need for energy and other 
nutrients increases and, therefore, she needs to eat more frequently. 
 
11. A poorly- nourished woman may develop complications and experience 
difficult labour that threatens her life. 
 
12. A poorly-nourished woman may also give birth to very small babies- who 
grow and develop very slowly and stand a greater risk of infection. 
 
13. Men need healthy diets to keep strong and productive. 
Men should consume enough energy-giving foods- to meet their need for health 
and physical activity. 
Men also need to eat from a wide variety of food groups to maintain healthy 
nutritional status and health. 
 
14. Iron-rich foods are helpful for reducing risk of anaemia. 
 
3. Some Positive attitudes to 15. Let’s not discourage some community members especially women and Address unfounded 
community food taboos and children from consuming certain foods in the community.  beliefs and taboos and 
beliefs and restrictions against Women (whether pregnant or lactating) as well as men and other community discriminatory practices 
practices women was high, members should be able to consume any food. to help improve women’s 
contributin preferential food  diets 
g to allocation practices 16. All foods produced in community can be nutritious depending on how we 
women’s where women are combine them to eat.  
poor dietary served food last and get Pregnant and lactating women as well as adolescent girls should consume 
practices bony parts of meat enough vegetables, fruits, legumes and fish. 
shared at meals  
 17. Family foods need to be shared according to individual needs. 
 261 
University of Ghana http://ugspace.ug.edu.gh
 Women and children have the greatest energy requirements and this must be 
considered when sharing food at the household level. 
 
18. Rest during daytime for women helps to conserve some of their energy to 
improve their health. 
Men and other household members can help women with household chores to 
allow women to have some rest during the day. 
 
 
4. Lack of Women had no rights to 19. Allowing women to have as much access to farmland   as men will improve Empower women to have 
sufficient farmland.  quantities and varieties of foods produced. control over resources so 
power and   they can improve their 
control over Women depended on 20. As the main people who prepare and feed families, allowing women more dietary quality. 
household husbands for small access to food resources will contribute to improving the qualities and 
food parcels of farmland to quantities of household food needs. 
resources grow household food  
by women. needs. 21. Women need to be economically supported to be able to contribute to food 
 for the family/household. 
Woman had no direct  
access to grains. 22. Income generating activities like trading, small-scale farming and food 
 processing do not need a lot of start-up capital but can serve as sources of 
Women had no regular income for women to contribute to household food needs. 
income- generating  
activities to enable  
them to acquire other 
food resources for the 
home. 
  
5. Poor Most households stored 23. Food-processing methods like drying can prolong shelf-lives and improve Improving food 
food their farm produce in availability of nutrient-rich vegetables and fruits. preservation and 
 262 
University of Ghana http://ugspace.ug.edu.gh
preservatio unprotected sacks and Most vegetables we consume can be preserved by drying so they be kept preparation enhances the 
n and kept them on the bare throughout the year. Some of these are being done already. Can we do similarly nutrient qualities of diets. 
preparation floors in their rooms. for others such as alefu, ayoyo, cassava leaf and bean leaf?  
methods  Inadequate drying of 24. Fermentation gives foods longer shelf-lives and also improves the digestion 
farm produce resulting and absorption of iron and other nutrients from such foods. 
in afflatoxin infestation 25. Cooking vegetables for an extended duration destroys some of the important 
Little use of nutrients they contain. 
fermentation as a Therefore, leafy vegetables should be cooked for very short durations. 
preservation method 26. Consuming well-handled and clean fruits and vegetables in their raw states 
and over-cooking of is one way of ensuring we get, as much as possible, all their nutrients to 
vegetables nourish our bodies. 
6. Poor Most compounds in the 27. Good environmental hygiene and sanitation enhance the health of all. Let’s keep our 
environmen community were Constantly keeping our compounds and community clean contributes to good environment clean to 
tal observed to be unkempt health. avoid contracting 
hygiene/san even during food  diseases. 
itation preparation. Animal 28. Keeping our drains and gutters clean will prevent the breeding of 
droppings and other mosquitoes, which causes malaria. 
debris were found all Keeping our environment clean prevents the outbreak of diseases such as 
over homes and cholera. 
surroundings.  
 
 
 
 263 
University of Ghana http://ugspace.ug.edu.gh
Appendix O: Cooking demonstration sessions 
A. Processing of soya bean flour. 
Ingredients:  
Whole dried soya bean and water 
Method 
1. Clean soya thoroughly by removing all debris such as stones, chaff, dirt and other 
foreign materials. 
2. Wash in clean water. 
3. Boil for 30 minutes in an open pot. 
4. Drain and wash in cold water while rubbing between the palms to hull. 
5. Dry hulled beans thoroughly. 
6. Mill to flour. 
7. Store in air-tight container and use within one month. 
B. Soya and “alefu” stew with boiled yam and rice 
Ingredients 
Soya beans, oil, pepper, tomatoes, saly, pounded anchovies, onion, water and “alefu” 
leaves 
Method: 
1. Heat oil (shea butter) on fire.  
2. Fry onion in oil for a while. 
3. Add ground tomato and pepper and stir till almost cooked. 
4. Add pounded anchovies. 
5. Mix some amount of soya flour with water into a paste and add into boiling stew 
and stir. 
6. Add chopped and “alefu” to the stew, stir and allow to cook for 5 minutes.  
7. Add a little salt to taste. 
8. Serve with boiled yam and rice. 
C. Weanimix porridge with soya flour 
 
 264 
University of Ghana http://ugspace.ug.edu.gh
Ingredients 
Maize, groundnuts, soya flour salt, sugar  
Method 
1. Take 4 parts of maize to 1 part of soya bean and half part of groundnuts. 
2. Roast groundnuts, hulled soya bean and maize separately. 
3. Put all three ingredients together, allow to cool and add spices if desired. 
4. Mill content into fine flour. 
5. Put water on fire with a little salt until boiling point. 
6. Mix flour into a liquid texture and to boiling water. 
7. Stir thoroughly to avoid lumps. 
8. Allow to boil for 5 to 10 minutes depending on the intensity of fire. 
9. Serve with a little sugar . 
E. Bra soup with soya flour 
Ingredients 
Groundnut paste, soya flour, pepper, tomato, onion, bra leaves, anchovies, water, 
dawadawa, salt 
Method 
1. Mix groundnut paste with water and cook for a while. 
2. Add water to make mixture lighter. 
3. Add onion, tomato and pepper to mixture and boil till tomato, onion and pepper 
are cooked. 
4. Remove and mash the tomato, pepper, onion and return to boiling mixture. 
5. Add anchovies, dawadawa and soya flour to boil till cooked. 
6. Add chopped bra to soup and allow to boil for 5 minutes. 
7. Add salt to taste.  
F.  Ayoyo soup with soya 
Ingredients 
Ayoyo leaves, soya flour, shea butter oil, tomato, onion, pepper, dawadawa, salt, 
pounded anchovies 
 265 
University of Ghana http://ugspace.ug.edu.gh
Method 
1. Heat shea butter in a pot. 
2. Add onion and fry till golden brown. 
3. Add ground or mashed tomatoes and pepper to heated butter with onion. 
4. Add dawadawa. 
5. Allow to cook for a while. 
6. Add water to boiling ingredients to desired quantity and allow to boil. 
7. Add a generous quantity of soya flour to boiling soup. 
8. Chop ayoyo leaves and boil for about 5 minutes. 
9. Add boiled ayayo to boiling soup and allow to boil for a minute or two. 
10. Add salt to taste and serve with TZ. 
G. TZ fortified with soya flour 
Ingredients 
Fermented maize flour, soya flour, water, cassava flour (optional) 
Method 
1. Put a pot of water on fire till boiling point. 
2. Mix maize flour with water into a porridge-like consistency. 
3. Add to boiling water amidst stirring to avoid lumps. 
4. Allow to boil for a few minutes. 
5. Mix maize, soya and cassava flour together. 
6. Add mixture of flour to boiling TZ porridge amidst stirring until a thick –like 
texture is achieved. Continue to stir thick –like TZ until cooked. 
7. Serve with any soup especially ayoyo and bra soup. 
H. Soya bean Tubani 
Ingredients 
Four parts of cowpea flour, one part of soya flour, water, salt, onion, roasted and 
pounded sesame seeds, salt-petre, shea butter oil,  dry yam leaves, transparent rubber 
bags for wrapping and steaming 
 
 266 
University of Ghana http://ugspace.ug.edu.gh
Method 
1. Mix cowpea and soya flour together.  
2. Add water to content and stir until fluffy and add salt to taste. 
3. Mix content until a butter like substance is achieved. 
4. Add a little salt petre to mixture. 
5. Fetch desired quantities of mixture and wrap into leaves for steaming. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 267 
University of Ghana http://ugspace.ug.edu.gh
Appendix P: Certificate of ethical clearance from Noguchi Memorial Institute for 
Medical Research 
 
 
 268 
University of Ghana http://ugspace.ug.edu.gh
 
Appendix Q: Certificate of ethical clearance from Noguchi Memorial Institute for 
Medical Research (Renewed) 
 
 269