UNIVERSITY OF GHANA COLLEGE OF HUMANITIES FACTORS INFLUENCING HOUSEHOLD COOKING FUEL CHOICE IN THE NSAWAM-ADOAGYIRI MUNICIPALITY OF GHANA COMFORT ADUBEA YEBOAH INSTITUTE OF STATISTICAL, SOCIAL AND ECONOMIC RESEARCH APRIL 2021 University of Ghana http://ugspace.ug.edu.gh UNIVERSITY OF GHANA COLLEGE OF HUMANITIES FACTORS INFLUENCING HOUSEHOLD COOKING FUEL CHOICE IN THE NSAWAM-ADOAGYIRI MUNICIPALITY OF GHANA BY COMFORT ADUBEA YEBOAH (10805276) THIS DISSERTATION IS SUBMITTED TO THE UNIVERSITY OF GHANA IN PARTIAL FULFILLMENT OF THE REQUIREMENT FOR THE AWARD OF MASTER OF ARTS IN DEVELOPMENT STUDIES INSTITUTE OF STATISTICAL, SOCIAL AND ECONOMIC RESEARCH APRIL 2021 University of Ghana http://ugspace.ug.edu.gh i DECLARATION I, Comfort Adubea Yeboah, affirm that this dissertation is my work under the supervision of Dr. Aba Crentsil, and does not contain any previously published work by any individual, except where due acknowledgment has been made in the text and at the end of the work. COMFORT ADUBEA YEBOAH … … (Student) Signature Date DR. ABA OBRUMAH CRENTSIL …………………… (Supervisor) Signature Date 24/04/2023 24/04/2023 University of Ghana http://ugspace.ug.edu.gh ii ABSTRACT There is heavy dependence on biomass fuels like firewood and charcoal which contribute to deforestation in Ghana. In Ghana, a majority of households (34.1%) use charcoal for their cooking needs. Others also use other types of cooking fuel like LPG. However, many households use a particular type of cooking fuel based on many reasons, including the price of cooking fuel and household average income. Therefore, this study sought to identify factors influencing household cooking fuel choice, specifically in the Nsawam-Adoagyiri Municipality. This study adopted a cross-sectional design to examine relationships between variables and a quantitative method approach was employed in the collection of primary data. A questionnaire was administered to 246 households in the Municipality. Data obtained from the field was coded and analyzed using STATA software. The study discovered that LPG was the most desired cooking fuel among households in the municipality. Also, the study revealed age, family size, educational level, age of household head, price of fuel, and indoor air pollution as the statistically significant factors that influence households’ choice of cooking fuel in the Nsawam-Adoagyiri Municipality. The analysis of the study helped make some recommendations such as subsidization of LPG prices, public education on the effects of the various cooking fuels, as well as free distribution of cylinders and burners to help achieve the SDG Goal 7, which seeks to achieve widespread access to affordable and reliable energy by the year 2030. It will also contribute to Ghana’s goal of transitioning 50% of households to the use of relatively clean cooking fuels such as LPG and biogas. University of Ghana http://ugspace.ug.edu.gh iii ACKNOWLEDGEMENTS Throughout the long hours and toils, this work has finally been completed and I could not have gotten to this point without the assistance and contributions of some great persons. First, I am very grateful to my supervisor, Dr. Aba Crentsil for her guidance, encouragement, forbearance, and assistance towards the completion of this work. It wouldn’t have been easy to complete this work on time without her scholarly advice and support. Her swift feedback and gentle criticisms on different chapters of this work motivated me to put in my best. I sincerely acknowledge the contributions of my bosses, Alfred George Thompson and Michael Oheneba Nti whose consistent support and encouragement propelled me to aim at achieving the best out of this program. To Oheneba, I am grateful for listening to all my complaints and always encouraging and believing in me that I could do it. To the households that willingly received me with smiles in the Nsawam-Adoagyiri Municipality, thank you so much! Even though being a master’s student is sometimes stressful, my dear friends, Aishat Tijani Najib, Rashid Alidu, Prince Shaibu, Frank Assiamah, Patience Abu Seidu, Gifty Blay, and Richard Akuoko Mensah made the journey an easy one for me and I am highly indebted to them. Thank you for making the journey easier and more fulfilling. A special thanks to Mr. Benjamin Osafo Kwaning. Sir, thank you for your support in diverse ways and for always believing in me. I am grateful. To my mom, Kingsley, and Jonas, I couldn’t have come this far without your support. ABOVE ALL, I GIVE GLORY TO GOD. University of Ghana http://ugspace.ug.edu.gh iv DEDICATION I dedicate this work to my lovely mom, Rosemary Animah. I love you, mom. University of Ghana http://ugspace.ug.edu.gh v TABLE OF CONTENTS DECLARATION ............................................................................................................................. i ABSTRACT .................................................................................................................................... ii ACKNOWLEDGEMENTS ........................................................................................................... iii DEDICATION ............................................................................................................................... iv TABLE OF CONTENTS ................................................................................................................ v LIST OF FIGURES/MAPS ......................................................................................................... viii LIST OF TABLES ......................................................................................................................... ix ACRONYMS .................................................................................................................................. x CHAPTER ONE ............................................................................................................................. 1 STUDY BACKGROUND .............................................................................................................. 1 1.1 Introduction ........................................................................................................................... 1 1.2 Problem Statement ................................................................................................................ 3 1.3 Objectives ............................................................................................................................. 5 1.4 Research Questions ............................................................................................................... 5 1.5 Scope of Study ...................................................................................................................... 6 1.6 Justification of the study ....................................................................................................... 6 1.7 Limitation of the Study ......................................................................................................... 7 1.8 Organization of the Study ..................................................................................................... 7 CHAPTER TWO ............................................................................................................................ 8 LITERATURE REVIEW ............................................................................................................... 8 2.1 Introduction ........................................................................................................................... 8 2.2 Kinds and Samples of Cooking Fuel .................................................................................... 8 2.2.1 Firewood ....................................................................................................................... 8 2.2.2 Charcoal ........................................................................................................................ 9 2.2.3 Kerosene........................................................................................................................ 9 2.2.4 Biogas .......................................................................................................................... 10 2.2.5 Liquefied Petroleum Gas (LPG) ............................................................................... 10 2.3 Cooking Fuel Programs ...................................................................................................... 14 2.3.1 Cooking Fuel Programs in Africa ............................................................................ 15 2.4 Factors Affecting the Choice of Cooking Fuel ................................................................... 17 2.4.1 Socio-Economic Factors ............................................................................................ 17 University of Ghana http://ugspace.ug.edu.gh vi 2.4.2 Environmental Factors .............................................................................................. 19 2.4.3. Cultural and Behavioural Factors ................................................................................... 21 2.4.4. Policies and Market Regulation characteristics ............................................................. 21 2.5 Theoretical and Conceptual Framework ............................................................................. 22 2.5.1 The Energy Ladder Theory ...................................................................................... 22 2.5.2 The Fuel Stacking Theory ......................................................................................... 24 2.6 Conceptual Framework ....................................................................................................... 26 2.7 Conclusion .......................................................................................................................... 28 CHAPTER THREE ...................................................................................................................... 29 RESEARCH METHODOLOGY.................................................................................................. 29 3.0 Introduction ......................................................................................................................... 29 3.1 Research Design.................................................................................................................. 29 3.2 Profile of the Study Area .................................................................................................... 29 3.3 Study population ................................................................................................................. 31 3.4 Sample Size and Sampling Technique ................................................................................ 32 3.5 Type and Sources of Data ................................................................................................... 33 3.6 Data Collection Method ...................................................................................................... 33 3.7 Data Analysis ...................................................................................................................... 34 3.7.1 Descriptive Analysis ................................................................................................... 34 3.7.2 Econometric Analysis ................................................................................................ 35 3.8 Ethical Consideration .......................................................................................................... 38 CHAPTER FOUR ......................................................................................................................... 40 RESULTS AND DISCUSSION ................................................................................................... 40 4.2 Demographic Characteristics of Sample ............................................................................. 40 4.2.1 Sex of Respondents .................................................................................................... 40 4.2.2 Age of Respondents .................................................................................................... 43 4.2.3 Family Size of Household .......................................................................................... 43 4.2.4 Price of Various Cooking fuels ................................................................................. 44 4.2.5 Type of Kitchen .......................................................................................................... 44 4.2.6 Household Decision Maker for Cooking Fuel ......................................................... 44 4.2.7 Level of Education of Household Head .................................................................... 44 4.2.8 Average Income of Households................................................................................. 45 University of Ghana http://ugspace.ug.edu.gh vii 4.2.9 Quality of Housing ..................................................................................................... 45 4.2.10 Preference for Cooking Fuel ................................................................................... 46 4.2.11 Sources of Cooking Fuel .......................................................................................... 47 4.2.12 Household Cooking task and Fuel consumption ................................................... 47 4.2.13 Reasons for a Household’s Choice of Fuel ............................................................. 48 4.3. Socio-Economic Factors Influencing the Choice of Cooking Fuel ................................... 50 4.3.1 Income level of a household ...................................................................................... 50 4.3.2 Gender of Household Decision-maker and Choice of Fuel .................................... 51 4.3.3 The Age of Household Head...................................................................................... 53 4.3.4 The Level of Education of a Household Head ......................................................... 54 4.4. Environmental Factors influencing the choice of cooking fuel ......................................... 56 4.4.1. Availability................................................................................................................. 56 4.4.2. Quality of Housing (Dwelling Unit) ......................................................................... 57 4.5 Multinomial Logit Regression to determine fuel stacking ................................................. 59 4.5.1. LPG and Firewood .................................................................................................... 61 4.5.2 LPG and Charcoal ..................................................................................................... 61 4.6 Conclusion .......................................................................................................................... 64 CHAPTER FIVE .......................................................................................................................... 65 SUMMARY, CONCLUSION, AND RECOMMENDATION .................................................... 65 5.1 Summary of Findings .......................................................................................................... 65 5.2 Conclusion .......................................................................................................................... 68 5.3 Recommendations ............................................................................................................... 69 REFERENCES ............................................................................................................................. 71 APPENDICES .............................................................................................................................. 85 APPENDIX 1: Questionnaire ....................................................................................................... 85 APPENDIX 2: Cross Tabulation outputs ..................................................................................... 91 APPENDIX 3 ................................................................................................................................ 93 University of Ghana http://ugspace.ug.edu.gh viii LIST OF FIGURES/MAPS Figure 2.1: Electricity/LPG ........................................................................................................... 12 Figure 2.2: Animal Dung ............................................................... Error! Bookmark not defined. Figure 2.3: Firewood ..................................................................................................................... 24 Figure 2.4: The Energy Ladder ...................................................... Error! Bookmark not defined. Figure 2.5: Energy Ladder to Fuel stacking.................................................................................. 31 Figure 2.6: Conceptual framework ............................................... Error! Bookmark not defined. Figure 3.1: Map of Study Area….……………………………………………………………….47 Figure 4.1: Distribution of the average monthly income of households………………………....44 Figure 4.2: Choice of cooking fuel………………………………………………………………45 Figure 4.3: Sources of cooking Fuel…………………………………………………………….46 Figure 4.4: Household cooking task and fuel consumption……………………………………...47 Figure 4.5: Reasons behind household cooking fuel choice……………………………………..48 Figure 4.6: Influence of gender role in household decision……………………………………..50 Figure 4.7: Age of household head and the choice of cooking fuel……………………………..52 Figure 4.8: Level of education of household head and the Choice of cooking fuel.................................................................................................................................................54 Figure 4.9: Knowledge on cooking fuels availability……………………………………………56 Figure 4.10: The type of dwelling unit and the choice of cooking fuel………………………….57 University of Ghana http://ugspace.ug.edu.gh ix LIST OF TABLES Table 3.1: Description of dependent variables ............................................................................. 37 Table 3.2: Explanatory variables used in the Multinomial Logit Regression and the expected effect on the choice of cooking fuel .............................................................................................. 38 Table 4.1: Descriptive Characteristics .......................................................................................... 41 Table 4.2: Mean values for the explanatory variables .................................................................. 42 Table 4.3: Cross-tabulation of average household income over the past 12 months and choice of cooking fuel .................................................................................................................................. 51 Table 4.4: Chi-Square Test result for Gender Role in Household Decision-making and Cooking Fuel Choice ................................................................................................................................... 52 Table 4.5: Chi-Square Test result for household cooking fuel choice and age of household head ....................................................................................................................................................... 54 Table 4.6: Cross-tabulation of the level of education of household head and the choice of cooking fuel .................................................................................................................................. 56 Table 4.7: Cross-tabulation of Dwelling Unit and the Choice of Cooking Fuel .......................... 58 Table 4.8: Multinomial Logit Regression on the determinants of cooking fuel choice ............... 60 Table 1: Cross-tabulation of average household income over the past 12 months and choice of cooking fuel .................................................................................................................................. 91 Table 2: Chi-Square Test result for Gender Role in Household Decision-making and Cooking fuel Choice .................................................................................................................................... 91 Table 3: Chi-Square Test result for age of household head and cooking fuel choice .................. 92 Table 4: Cross tabulation of level of education of household head and the choice of cooking fuel ....................................................................................................................................................... 92 Table 5: Cross tabulation of dwelling unit and the choice of cooking fuel .................................. 92 Table 6: Multinomial Logistic Regression Output ....................................................................... 93 University of Ghana http://ugspace.ug.edu.gh x ACRONYMS ACCES Africa Clean Cooking Energy Solutions EAP CSI East Asia and Pacific Clean Stove Initiative ECREEE Ecowas Centre for Renewable Energy and Energy Efficiency EFA Eco-Fuel Africa ESMAP Energy Sector Management Assistance Program GACC Global Alliance for Clean Cookstoves GLPGP Global LPG Partnership GoG Government of Ghana IEA International Energy Agency LPG Liquefied Petroleum Gas SDG Sustainable Development Goals UNSEA United Nations Sustainable Energy for All WACCA West African Clean Cooking Alliance University of Ghana http://ugspace.ug.edu.gh 1 CHAPTER ONE STUDY BACKGROUND 1.1 Introduction Cooking fuel is crucial in every home as it assists in cooking activities. Cooking energy comes in many forms and so households decide on the suitable type for their homes based on several reasons such as availability of cooking fuel and household characteristics (Nielsen, 2017). In many instances, the cooking fuel choice of a household depends on income, price, and a household head’s level of education (Pundo & Fraser, 2006). However, households also make choices under several constraints which propel them to use either ‘traditional’ cooking fuels such as firewood and charcoal or ‘modern’ cooking fuels such as Liquefied Petroleum Gas (LPG) and electricity, and for some households, a mixture of both (Truneh, 2014). Presently, the world is striving to achieve Sustainable Development Goal 7 (SDG 7) which aims to ensure universal access to reliable, affordable, and modern energy by 2030 (Franco, Power, & Whereat, 2020). The United Nations Environment Programme (2019) reports that in 2016, over three billion people worldwide, especially in Asia and Africa depended on biomass fuels – wood, leaves, crop waste, dung, etc. for their daily cooking needs. About 1.3 billion folks worldwide did not have access to electricity, and 2.6 billion people relied on traditional biomass (fuelwood, charcoal, dung, and agricultural residues) for cooking largely in rural areas in Africa and Asia as of 2011 (International Energy Agency (IEA,2013). This indicates a decrease in the number of households who depended on biomass fuel in 2006 as reported by the World Health Organization. Literature suggests that an increase in income, taste and preference, as well as level of education, account for the decrease in biomass fuel usage (Atanassov, 2010; Berhe, 2014; Karimu, 2015). University of Ghana http://ugspace.ug.edu.gh 2 As opined by Zhang & Hassen (2017), an increase in well-being and income in developed countries like China leads to a shift from traditional cooking fuels like firewood to LPG and electricity. Similarly, in Ghana, a rise in income, the level of education of household heads, access to modern infrastructure, among others contribute to the choice of cooking fuel in homes (Heltberg, 2004; Kwakwa, Wiafe, & Alhassan, 2013; Karimu, 2015). It is reported that charcoal is the most used cooking fuel (34.1 percent) by households in Ghana. The next most used cooking fuel is firewood, accounting for 33.3 percent, followed by LPG with a household percentage usage of about 25 percent (Ghana Statistical Service, 2019). However, preference for cooking fuel types differs between households in urban areas and rural areas. The Ghana Statistical Service (2019) reports that while charcoal and LPG are the most used cooking fuels in cities, firewood is rather preferred among rural households. Among the determining factors that influence this disparity, Karimu et al., (2016) assert that homes choose a cooking fuel type based on socio-economic factors. For example, Karimu (2015) and Alhassan (2018) cite the level of income, level of education of household head, and price of cooking fuel as notable socioeconomic factors that affect a family’s cooking fuel choice. Also, Nyankone & Waithera (2016) and Makonese, Ifegbesan & Rampedi (2018) discover that most households in rural areas prefer solid cooking fuels such as firewood and charcoal due to their availability and easy access. They further reveal that most households collect firewood freely from their farms and those who do not have trees on their farms either receive them as gifts from their neighbors or purchase them at relatively low prices. Dry and semi-arid areas in the country contain most of the forests and are good sources of biomass, hence this provides easy access to firewood and charcoal for residents in those areas. According to the UNFAO (2010), Ghana’s rate of deforestation is 3 percent per annum and is estimated to University of Ghana http://ugspace.ug.edu.gh 3 increase over time. In the year 2000, annual wood production was about 30 million tons, and about 18 million tons were used as wood fuel. This practice continues to harm the country’s forest reserves although the exploitation of wood resources is not the major cause of deforestation in the country (Trossero, 2002). In current times where the world is consistently experiencing negative consequences of climate change, it is important to implement safe, effective, and renewable energy use among municipalities in Ghana to enhance environmental sustainability (Bawakyillenuo & Agbelie, 2015). Households are, therefore, encouraged to use relatively safe cooking fuels like biogas. Unfortunately, many households prefer other alternatives because biogas (fecal gas), due to its nature, is not attractive to households, especially Muslims in livestock rearing areas in the Northern part of Ghana (Ahiekpor, Antwi, Bensah, & Ribeiro, 2015c). Consequently, to better understand household cooking fuel choice in a given geographic location, it is important to incorporate the sources of energy that provide households with their cooking and heating needs (Wiesmann, Azevedo, Ferrão & Fernandez, 2011; Soltani, Rahmani, Pour, Ghaderpour, Ngah & Misnan, 2019). 1.2 Problem Statement There is a rise in urbanization across Sub-Saharan Africa and this makes urban household energy a critical issue for developing countries as there is heavy dependence on biomass fuels which contribute to deforestation (Adeyemi & Adereleye, 2016). According to IEA (2019), biomass fuels such as firewood and charcoal are widely used among households in sub-Saharan Africa. The available literature on household cooking fuel usage reverberates the conception that urbanization and lifestyle changes cause households to use modern fuels than traditional fuels. However, in Sub-Saharan Africa, many countries, including Ghana still rely on biomass fuels and are adopting University of Ghana http://ugspace.ug.edu.gh 4 modern fuels such as LPG at a slower pace (Barnes, Krutilla & Hyde, 2005; Mokennen & Kohlin, 2008; Truneh, 2014). The heavy reliance on solid fuel by urban households in developing countries is due to complex economic and cultural restraints (World Bank, 2011). Socio-economic factors such as education and income as stated by Karimu (2015), also affect the choice of households’ cooking fuel. Where the income of a household is relatively low, the household would opt for cheaper cooking fuels like firewood and charcoal (Alhassan, 2018; Nlom & Karimov, 2015). Underprivileged households burn large amounts of firewood and charcoal with inefficient stoves to meet their cooking needs. This exposes those involved in cooking for households, particularly women, to high levels of air pollution. The combustion of biomass fuels such as firewood and charcoal is known to produce high levels of pollutants, including carbon monoxide, nitrogen dioxide, particulate matter, and other organic compounds that threaten respiratory health and the environment (Fullerton, Bruce, & Gordon, 2008). Despite the numerous disadvantages such as the emission of smoke and less heat associated with biomass fuels, lots of Ghanaians, especially those in rural areas still use firewood, cow dung, and charcoal for cooking (Ahiekpor, Antwi, Bensah & Ribeiro, 2015). Cognizant of the fact that poverty influences the choice of household cooking fuel, the Government of Ghana (GoG) has made efforts to make alternative cooking fuels accessible to many households all over the country. Thus, the government has resorted to diversifying cooking fuel sources among households by introducing Liquefied Petroleum Gas (LPG) onto the market some years back. The LPG promotion project sells both cylinder and stove to support consumers who until then used to buy the cylinder and stove separately (Amissah-Arthur & Amonoo, 2004). The program, however, has not helped the ordinary Ghanaian who does not have a regular flow of income, hence University of Ghana http://ugspace.ug.edu.gh 5 depending on solid fuel for cooking. Additionally, some households who can afford modern fuels still prefer traditional cooking fuels like firewood and charcoal. To this effect, identification of the factors that influence the choice of household cooking fuel in Ghana would be helpful because this will assist policy-makers in designing sustainable intervention mechanisms to enhance the adoption of environmentally friendly cooking fuels to help combat climate change. Several gaps still exist, as few studies by Ahiekpor, Antwi, Bensah & Ribeiro (2015) and Karimu (2015) have focused on peri-urban areas in Ghana. Therefore, this study examines cooking fuel at the household level and factors influencing choices among peri- urban households in the Nsawam-Adoagyiri Municipality. 1.3 Objectives The key objective of the study is to examine factors that influence the choice of household cooking fuel in the Nsawam-Adoagyiri Municipality. The specific objectives are to: 1. Identify the kinds of cooking fuel choices available to households and their preferences in the Nsawam-Adoagyiri Municipality. 2. Examine different socio-economic factors influencing household cooking fuel choice in the Nsawam-Adoagyiri Municipality. 3. Determine the environmental factors influencing household cooking fuel choice in the Nsawam-Adoagyiri Municipality. 1.4 Research Questions The study seeks to answer the following research questions; University of Ghana http://ugspace.ug.edu.gh 6 1. What are the kinds of cooking fuel choices available to households and what are their preference in the Nsawam-Adoagyiri Municipality? 2. What are the socio-economic factors influencing household cooking fuel choice in the Nsawam-Adoagyiri Municipality? 3. What are the environmental factors influencing household cooking fuel choice in the Nsawam-Adoagyiri Municipality? 1.5 Scope of Study This study is limited to Nsawam-Adoagyiri Municipality in the Eastern Region of Ghana. This area is chosen for the study because of the limitation imposed by the Institute of Statistical, Social and Economic Research (ISSER) concerning designated areas of study and partly due to the peri- urban nature of the area. The focus of the study is on the factors that influence household cooking fuel choice in the municipality. Also, Nsawam-Adoagyiri is cosmopolitan with 1.6 Justification of the study A growing body of literature such as (Aberilla, Gallego-Schmid, Stamford & Azapagic, (2020) and the Food and Agriculture Organization of the United Nations (2010) have noted various types of cooking fuels as significant contributors to climate change. For Ghana to achieve its national goal of clean energy, policymakers must be abreast with the various kinds of cooking fuels being adopted at the household level and their related determinants. This, therefore, gives rise to the need for information on the types of cooking fuels used in various municipalities in Ghana. The study will contribute insight into the major sources of cooking fuel as well as factors influencing households’ choices. These factors can help stakeholders and policymakers to design strategies that will increase peoples’ awareness of alternative and cleaner energy cooking fuels. Again, the discoveries of the study would be important for future cooking fuel-related studies. Thus, University of Ghana http://ugspace.ug.edu.gh 7 individuals and groups interested in conducting similar research work could refer to this work as a source of literature. 1.7 Limitation of the Study The study focused on household cooking fuel choice and does not address other units such as restaurants. Also, the unit of analysis was limited to households in the Nsawam-Adoagyiri Municipality in the Eastern Region of Ghana. Therefore, the setting of the study may not lead to a generalization of the findings across the country. Again, there was limited time and financial resources to undertake the study. Under these limitations, the researcher put in much effort to ensure that the quality of the research was not negotiable. The aforementioned limitations, which the researcher had no control of, may impact the level of generalization of this work. 1.8 Organization of the Study This research consists of five chapters. The first chapter gives a background to the study, explains the problem, and states the objectives of the research and the questions it seeks to answer, followed by the scope, limitation, and significance of the study. Chapter Two reviews relevant literature and provides theoretical and conceptual insights into factors affecting household cooking fuel choice. Chapter Three also explains the methodology for the research and includes a description of the study area, research design, data type and source, sampling procedure, data instruments and collection, data collection procedure, the definition of variables, and processing of data. Chapter Four describes the demographic profile and socio-economic background of the sample respondents. Additionally, it analyses data and highlights findings from this research, and presents results. This chapter also introduces discussion points based on the results. Lastly, Chapter Five concludes the study and presents recommendations. University of Ghana http://ugspace.ug.edu.gh 8 CHAPTER TWO LITERATURE REVIEW 2.1 Introduction This chapter examines existing literature on cooking fuel and factors affecting households’ choices. The first section of the chapter introduces the kinds of cooking fuels, followed by fuel programs, factors influencing the choice of household cooking fuel, theoretical review, and then a conceptual framework to help answer the research questions. 2.2 Kinds and Samples of Cooking Fuel Different terms are used in describing household cooking fuel types based on their characteristics. According to Malla & Timilsina (2014), household cooking fuels can be characterized into "traditional" (animal dung, agricultural residues, and fuelwood), "intermediate" (wood pellets, charcoal, briquettes, coal, and kerosene), and "modern" (LPG, solar, biogas, natural gas, electricity, plant oils, and dimethyl ether) based on the level of energy development. In Central America, Wang, Franco, Masera, Troncoso & Rivera (2013) assert that nearly 20 million people use biomass, mainly fuelwood for cooking. Furthermore, coal is a popular household cooking fuel in China, animal dung in India, charcoal in Sub-Saharan Africa, kerosene in Djibouti, and electricity in South Africa (Smith et al., 2012). According to the Ghana Statistical Service (2014), some common types of cooking fuel in Ghana include firewood, charcoal, kerosene, LPG, and sometimes, electricity. 2.2.1 Firewood Firewood is the commonest cooking fuel in most African countries and South East Asia. It is estimated that about 60 percent of households in these regions cook with solid fuels such as University of Ghana http://ugspace.ug.edu.gh 9 firewood and charcoal (Bonjour, Adair-Rohani, Wolf, Bruce, Mehta, Prüss-Ustün ,…& Smith, 2013). In Ghana, wood fuel provides about 71 percent of the total annual energy demand (Mason, 2008). Firewood is burned in open stoves, usually traditional three-stone stoves and mud stoves. This produces low energy density and low efficiency on combustion, between 10 to 20 percent (Bailis, 2004). Again, due to the open spaces in the stoves, it is difficult to control the level of heat. 2.2.2 Charcoal About 17 percent of all the wood used worldwide as fuel is converted to charcoal. There has been an increase in the production of charcoal in recent times as demand has grown among urban populations and enterprises predominantly in parts of Asia, Sub-Saharan Africa (SSA), and South America (Jones & Challinor, 2019). In the US for instance, people use charcoal for grilling due to the flavors it imparts to grilled food (Kammen & Lew, 2005). Charcoal is a solid residue that is produced from the carbonization of wood (Afrane & Ntiamoah, 2011a). In Ghana, charcoal production is popular in the transition belt between the tropical forests of the south and the savanna woodlands where fresh wood is harvested for processing. It is also produced from the harvesting of live trees and wastes generated from logging and sawmills’ operation and is often packaged in sacks for distribution by trucks from the production site to wholesalers and retailers (Obiri, Nunoo, Obeng, Owusu & Marfo, 2014). Many urban households in developing countries prefer charcoal to firewood because it has higher energy density and burns more cleanly with less smoke (Schlag & Zuzarte, 2008). 2.2.3 Kerosene Kerosene is a microscopic concentrate of petroleum. It is a combination of hydrocarbons – 12-18 carbon atoms per molecule and it boils between 190-250oC (Amakiri, Owen & Iboh, 2009). Some households in Sub-Saharan Africa use kerosene as their main source of cooking fuel and as an University of Ghana http://ugspace.ug.edu.gh 10 insect repellent because of its odor ( Amakiri, Owen & Iboh, 2009; Adeyemi & Adereleye, 2016). According to Bailis (2004), wick stoves and pressurized stoves are used for cooking with kerosene and both are easy to use. 2.2.4 Biogas Biogas is a mixture of carbon dioxide (CO2) and methane (CH4) and is produced from the process of anaerobic digestion of organic wastes like fecal matter (Feroskhan & Ismail, 2016). The availability of inputs like manure and water influence the use of bio-digesters for cooking at home. China and India are widely known for the use of bio-digesters and the spread of the technology to other parts of the world (Odoi-Agyarko, 2009). 2.2.5 Liquefied Petroleum Gas (LPG) LPG is a mixture of hydrocarbon compounds made up of propane (C3H8) and butane (C4H10) (Oyelami & Bolaji, 2015). According to the World Liquefied Petroleum Gas Association (2004), LPG demand grew from 200 million tonnes in the year 2000 to 237 million tonnes globally with Asia being the largest consumer with about 60 million tonnes. Many households in developing countries recorded an increase in LPG consumption by 5% between 1990 and 2011. In Ghana, LPG is produced by the country’s only refinery, the Tema Oil Refinery, and sometimes imported from Nigeria (Afrane & Ntiamoah, 2011). LPG is filled into steel cylinders at LPG fuel stations and later connected to a cookstove for use at home. According to the Ghana Statistical Service (2019), about 25% of households use LPG for cooking. 2.2.6 Electricity A very small percentage of Ghanaian households (less than 1 percent) depend on electricity for their cooking needs (World Health Organization, 2018). Out of this percentage, it is mostly high- University of Ghana http://ugspace.ug.edu.gh 11 income households with a relatively small household size that uses electricity for cooking (Ahiekpor, Antwi, Bensah & Ribeiro, 2015). According to Malla & Timilsina (2014), households that use electricity for cooking are usually urban dwellers. In some urban areas in South Africa, Gabon, and Cameroon, electricity is common among households that can afford grid connections (Blimpo & Postepska, 2017). However, grid-based electricity is not a feasible source of cooking fuel in rural areas of Sub-Saharan Africa where electrification is not common and where transmission distances are large (Puzzolo, Zerriffi, Carter, Clemens, Stokes, Jagger,… & Petach, 2019). On average, it is estimated that a household in Africa uses about 8kWh per day and 3232 kWh per year for cooking (Kammila, Kappen, Rysankova, Hyseni & Putti,2014). University of Ghana http://ugspace.ug.edu.gh 12 Images of some cooking fuels A: charcoal Source: africa-energy-portal.org B: Electricity/LPG Figure 2.1. Electricity/LPG Source: (Harsha K R. CC BY-SA 2.0 and)envirofit.org University of Ghana http://ugspace.ug.edu.gh 13 C: Animal dung Figure 2.2: Animal Dung Source: www.alamy.com D: Firewood Figure 2.3: Firewood Source:(Harsha K R . CC BY-SA 2.0 and)envirofit.org University of Ghana http://ugspace.ug.edu.gh 14 2.3 Cooking Fuel Programs Firewood, charcoal, cow dung, and crop waste are predominant cooking fuels in many households in developing worlds, including Sub-Saharan Africa. The use of solid fuels and potentially harmful fuels like kerosene causes major public health problems (ESMAP & GACC, 2015). Lambe, Jürisoo, Wanjiru & Senyagwa, (2015) reveal that the usage of inefficient fuels for cooking also levies major expenses on governments. While this may be worrying, it is worth noting that there would come a time that the world will adopt transformative measures to improve the cooking sector. Programs aimed at embracing relatively safer fuels such as LPG and renewable energy alternatives like biomass pellet fuels and biogas must be intensified to facilitate a high rate of adoption to save the environment and improve the health of individuals in households. Countries like Ghana, China, Bangladesh, and Peru have united with organizations and programs such as the United Nation’s Sustainable Energy for All (UNSEA), Global Alliance for Clean Cookstoves (GACC), Global LPG Partnership (GLPGP), Africa Clean Cooking Energy Solutions (ACCES), and East Asia and Pacific Clean Stove Initiative (EAP CSI) to deliver an enabling environment for clean cooking solutions (ESMAP & GACC, 2015). Most developing countries are faced with the problem of implementing improved cooking solutions to avoid the widespread harmful effects of smoky fuels. Available data from the Energy Sector Management Assistance Program (ESMAP) and Global Alliance for Clean Cookstoves (GACC) indicate that in 2010, only 10 percent of households in Sub-Saharan Africa had access to renewable fuels and advanced biomass solutions with few countries managing to introduce clean and improved ways of cooking on a broader scale to combat the effects of inefficient cooking fuels (ESMAP & GACC, 2015; Gold Standard, 2016). According to GACC (2011), poverty and lack of University of Ghana http://ugspace.ug.edu.gh 15 awareness among households in developing countries about the benefits of clean cooking fuels may also be the reason for the reliance on unclean cooking fuels. 2.3.1 Cooking Fuel Programs in Africa Several households in Africa rely on unclean and inefficient cooking fuels in their homes. About 80 percent of households in Sub-Saharan Africa use solid fuels such as charcoal and firewood for their basic cooking needs (Kumie, Emmelin, Wahlberg, Berhane, Ali, Mekonnen & Brandstrom, 2009; Gold Standard, 2016). Meanwhile, the use of solid fuel in developing countries has been identified as a major source of black carbon, contributing about one-fourth of total global emissions, second only to open burning of forests, savanna, grasslands, et cetera (Bond, Streets, Yarber, Nelson, Woo & Klimont, 2004). This has posed several threats to health and the environment and has triggered the conception of various interventions to minimize the effect. Impact Carbon, a partner of Global Alliance for Clean Cook Stoves (GACC), joined forces with BioLite in Uganda to implement a full-scale product launch of BioLite’s Home Stove, a fan stove that reduces household air pollution when used together with charcoal. Eco-Fuel Africa (EFA) also produces green charcoal from locally sourced biomass waste as an alternative to traditional wood fuel for Ugandans. The ECOWAS Centre for Renewable Energy and Energy Efficiency (ECREEE) also established the West African Clean Cooking Alliance (WACCA) in 2012 to encourage the enactment of policies on clean cooking initiatives in the ECOWAS region (ECREEE, 2013). According to ECREE, this initiative is to ensure that by 2030, all households in the ECOWAS region would have access to modern cooking fuels and devices. All these programs are aimed at reducing the effects of unclean cooking fuels. University of Ghana http://ugspace.ug.edu.gh 16 2.3.2 Cooking Fuel Programs in Ghana Ghana’s improved cooking fuel program can be traced back to the 1990s when the Government of Ghana initiated the Liquefied Petroleum Gas (LPG) promotion program. This program was aimed at encouraging households that used charcoal and firewood for cooking to rather adopt LPG for cooking (Quaye-Foli, 2002). As part of the program, 14.5kg and 5kg cylinders were distributed to the general public, and consumers were given free cylinders upon request or cylinders with LPG where they paid for only the gas (Ahunu, 2015). This initiative resulted in an upsurge in LPG usage from 5,267 MT in 1989 to 32,000MT in 1996 (Quaye-Foli, 2002). Before this, only about 6.5 percent of the population in urban areas used LPG, while no households in the rural areas used LPG (Quaye-Foli, 2002). According to the Ghana Statistical Service (2014), firewood and charcoal were the main cooking fuels in Ghana in 2013. About 34.1 percent of households used charcoal and 33.3 percent of households used firewood for their cooking needs. According to the report, LPG usage among households was just about 22 percent. This informed the Energy Ministry to reduce the use of wood fuel to 50% by 2020 through nationwide dissemination of clean and efficient cooking stoves and alternative cleaner cooking fuels such as LPG (Tettey, 2014). The LPG cookstove dissemination program aims to encourage the usage of cleaner fuels for cooking in both rural and urban areas. The Rural LPG promotion program was also launched in 2013 to increase LPG usage in rural areas and also reduce the usage of biomass fuel which has proven to be a major cause of respiratory diseases in Ghana (Asante, Afari-Asiedu, Abdulai, Dalaba, Carrión, Dickinson,…& Jack, 2018). In 2016, a total of 2000 sets of LPG stoves and accessories were distributed in ten districts, including Atebubu, Bole, Jirapa, Krachi West, Nanumba South, Nkoranza South, Nkwanta North University of Ghana http://ugspace.ug.edu.gh 17 & South, Pru, and West Gonja. The rural LPG promotion program is aimed at increasing the adoption of LPG usage among rural and peri-urban households from 5.5 percent to 50 percent by 2030 (Global Alliance for Clean Cookstoves, 2014; Tettey, 2014). The Council for Scientific and Industrial Research (CSIR) in collaboration with Plan Ghana (an NGO) also developed and distributed improved firewood stoves to about 1000 households in Tumu District in the Upper- East Region in the same year. Despite the increasing usage of LPG, many rural communities in Ghana are still reluctant to adopt it (Broni-Bediako & Amoritin, 2018). 2.4 Factors Affecting the Choice of Cooking Fuel For many households around the world, the choice of a particular cooking fuel depends on several factors. Such factors include socio-economic, cultural, and environmental conditions (Pundo & Fraser, 2006). Household characteristics also vary from country to country and so the factors may not necessarily be universally applicable. 2.4.1 Socio-Economic Factors Socio-economic factors such as household income, home-ownership, price of cooking fuel, and the educational level of a household head can all affect a household’s choice of cooking fuel (Berhe, 2014; Karimu, 2015; Soltani, Rahmani, Pour, Ghaderpour, Ngah & Misnan, 2019). In finding the driving factors of the choice of household cooking fuel, Bansal, Saini & Khatod (2013) and Nlom & Karimov (2015) also discover that household income is a contributory factor in the choice of cooking fuel in rural India and Northern Cameroon respectively. A household with a higher level of income will prefer modern cooking fuels like LPG and electricity whereas a low- income household would choose traditional fuels such as firewood and charcoal for cooking (Stephen 2011). In Ghana, similar studies by Karimu (2015) and Alhassan (2018) reveal that University of Ghana http://ugspace.ug.edu.gh 18 income determines whether a household should use traditional, intermediate, or modern cooking fuels. In addition to income, education also influences the preference for a particular type of household cooking fuel. For example, Pandey & Chaubal (2011) find that in rural India, several educated females between the age bracket of 10 and 50 years use clean cooking fuels as compared to their counterparts with no level of education. Using data from Kisumu in rural Kenya, Pundo & Fraser (2010) also assert that though firewood is by far the preferred cooking fuel for many households in the study area, the level of education of the wife determines the choice of cooking fuel in a household. The authors claim that it is possible for an educated wife in a household to use charcoal or kerosene for cooking instead of firewood because household heads who are educated are well- informed about the side effects of cooking fuels. Additionally, Adeyemi & Adereleye (2016) realize that in Nigeria, a household head who is educated is more likely to use liquid fuel instead of solid fuel for cooking. This, they believe education informs people about fuel characteristics, thereby increasing the preference for better cooking fuels. Many households in Sub-Saharan Africa also consider price when choosing a particular cooking fuel. It has been noted that households prefer cooking fuels that are less expensive and tend to use the alternative of expensive fuels (Ahiekpor, Antwi, Bensah & Ribeiro, 2015; Nlom & Karimov, 2015). Therefore, households that perceive LPG to be expensive prefer charcoal for cooking. According to Tettey (2014), since the price of cooking fuel is one of the major determinants of cooking fuel preference, the majority of households in Ghana use relatively cheaper cooking fuels such as charcoal. The occupation of a household head is also found to be a factor in the choice of cooking fuel. Thus, household heads who are employed in white-collar jobs prefer LPG while University of Ghana http://ugspace.ug.edu.gh 19 those employed in blue-collar jobs prefer solid fuels like firewood and charcoal for cooking (Heltberg, 2005; Eakins, 2013). 2.4.2 Environmental Factors It is widely known that environmental factors such as the availability of freely gathered fuel, accessibility, cleanliness of cooking fuel, and cooking location, determine the choice of cooking fuel (Rehfuess, Mehta & Prüss-Üstün, 2006). For instance, a study by Nyankone & Waithera (2016) in Kenya reveals that the majority of Thuti residents have trees on their farms from which they source their firewood and those without trees on their farms can buy from those that grow them. Thus, getting firewood for energy is not difficult in Thuti-Kenya. The accessibility of cooking fuels also contributes to their preference. In Nigeria, Nielsen (2017) asserts that most households purchase cooking fuel in bits and so prefer to use the ones that are easily accessible in their neighborhoods. Karimu (2015) also asserts that in Ghana, access to points of sale where cooking fuel can be purchased influences a household’s choice. In the study, he discovers that LPG usage was popular among households that had easy accessibility to LPG within their communities. Preference of any of the cooking fuels may also depend on the type of dwelling unit of a household and cooking location. Karimu (2015) remarks that in Ghana, households living in modern apartments rarely use solid fuels but prefer modern fuel types such as LPG for cooking. Again, households whose kitchens are separated from their apartments prefer solid fuels such as charcoal but those with their kitchens located in their apartments prefer to use modern fuels like LPG to reduce the amount of indoor pollution. University of Ghana http://ugspace.ug.edu.gh 20 Community physiognomies are also considered important in cooking fuel preference among households. For instance, Suliman (2013) asserts that residents in an urban area may have a preference for LPG instead of charcoal or firewood for their cooking needs. Moreover, studies by Kwakwa, Wiafe, & Alhassan (2013) and Francis, Geofffrey & Gemma (2014) reveal that residence in a rural community increases the usage of solid fuels and kerosene for cooking but reduces the use of LPG and electricity for cooking and heating. They also discover that households in urban centers use modern cooking fuels such as LPG and sometimes electricity, while those in rural areas prefer traditional and intermediate fuels like biomass and kerosene for cooking. Correspondingly, families who live in modern infrastructure prefer LPG for cooking than households that live in other structures (Barnes, Krutilla & Hyde, 2005). In Ghana, Karimu (2015) draws a similar conclusion that households who have access to electricity usually use non-solid fuels such as LPG and electricity for cooking and warming their foods. Consequently, household units are also proven to influence the choice of cooking fuel. For example, Olugbire, Aremu, Opute, Ojedokun, Olawale & Adisa (2016) reveal that in Nigeria, the choice of cooking fuel is dependent on the nature of a household’s dwelling place. Households with single-unit dwellings usually use charcoal or kerosene because there is not enough space for a cylinder and a cooker at the same time. However, LPG is more suitable for households living in spacious apartments. In some cases, households consider the cleanliness of cooking fuels when making a choice. A study by Selvakkumaran & Ahlgren (2019) reveals that urban households prefer clean cooking fuels that reduce pollution and are environmentally friendly. Thus, many households in this area prefer cooking fuels that emit no or less smoke. University of Ghana http://ugspace.ug.edu.gh 21 2.4.3. Cultural and Behavioural Factors Food tastes and cooking practices are among the factors that influence household energy choices (Heltberg, 2005; IEA, 2006). As proof, a study by Masera et al. (2000) in rural Mexico revealed that although people can afford modern fuels, they are still using traditional biofuels to cook dishes because of the convenience of cooking traditional dishes over direct flames. Correspondingly, IEA (2006) found that Indian households prefer to use traditional wood stoves to bake bread. Heltberg (2005), and Guptaa & Kohlin (2006) both revealed in their studies that cultural and lifestyle choices influence household fuel decisions. For instance, findings from Heltberg (2005) showed that indigenous ethinic tribes in Guatemala have unique energy portfolios due to cultural differences. In addition, other studies, for example, Masera et al. (2000) and Xiaohua & Zhenmin (2003) indicated that households make decision on cooking fuel choice based on their status, thus, a higher income household is most likely to opt for a modern trend of cooking fuel source. 2.4.4. Policies and Market Regulation characteristics Government policies on energy distribution and regulation ultimately affects households decision on the type of energy to use (Barnes et al. 2005). Kowsari & Zerriffi (2011) mention that government intervention policies have mostly increased the availability of contemporary fuels. Interventions such as government energy subsidies have made it possible for poorer households to afford modern energy fuels ( ESMAP, 2000, 2004). While the intent of government subsidies is to provide a cushion for the poor, research has revealed that most government subsidies on energy end up benefitting the middle-class and the rich rather than the poor (ESMAP, 2000). Aside subsidy having a conflicting outcome in some cases, it is found to be a drain on government’s coffers hence, most governments now resort to other market strategies such as regulation rather than subsidization policies (ESMAP, 2000; Elias et al. 2005; Victor & David, 2009). University of Ghana http://ugspace.ug.edu.gh 22 Regulatory policies on production and distribution affects household choices on energy, thus, the availability and access to energy choices to households are influenced by regulation. For example, in China, Jiang & O’Neill (2004), recorded that respondents chose coal as the main source of fue because there is a ban on traditional biofuel. Further, Barnes et al. (2005) found in a study that government’s policies has facilitated households choice cooking fuel from woodfuel to LPG. 2.5 Theoretical and Conceptual Framework The energy ladder theory and fuel stacking theory are considered for this work. The conceptual framework, therefore, is based on the assumptions of these theories. 2.5.1 The Energy Ladder Theory The Energy Ladder theory explains household cooking fuel preferences in developing countries (Muller & Yan, 2018). The theory explains that as income increases, households tend to switch from traditional cooking fuels to transition fuels such as kerosene, and then to modern fuels such as LPG and electricity (Heltberg, 2004; Muller & Yan, 2016). This concept mimics the typical income effect of consumer economic theory that explains that as income rises, consumers substitute inferior goods for necessities and luxury goods. This means that inferior goods in this model refer to biomass cooking fuel and luxury goods refer to LPG and electricity. Besides, this model assumes that households are faced with an array of cooking fuels in increasing technological sophistication, hence the need to make a choice based on their socio-economic status (Chambwera & Folmer, 2007; Link, Axinn & Ghimire, 2012). University of Ghana http://ugspace.ug.edu.gh 23 Figure 2.4: The Energy Ladder Source: (Schlag, N. & Zuzarte, 2008) In this theory, cooking fuels are ordered on a ladder according to a family’s preference based on characteristics such as efficiency, cleanliness, and speed (Hiemstra- van der Horst & Hovorka, 2008). Although not universally applicable, this theory explains the relationship between income and the choice of household cooking fuel in developing countries. In Sub-Saharan Africa, many households prefer traditional and intermediate fuels to modern fuels (Njiru & Letema, 2018). Thus, households with lower incomes cannot afford modern cooking fuels like electricity and LPG. The energy ladder theory is used in this work to show how households’ income affects their choice of cooking fuel. In essence, while the energy ladder theory postulates that moving up the ladder denotes an abandonment of lower level fuels, other studies (Hosier et al. 1987; Masera et al. 2000; Leiwen & O’Neill, 2003; Heltberg, 2004) showed that energy choices are not linear. This indicates that the energy ladder theory is limited in reality. For example, Kowsari & Zerriffi (2011) pointed out that the energy ladder is efficient in explaining energy choice, particularly in urban University of Ghana http://ugspace.ug.edu.gh 24 areas, thus, there is a limitation in the application of this theory in rural, and peri-urban communities. 2.5.2 The Fuel Stacking Theory This theory is a critique of the energy ladder theory where Masera, Saatkamp & Kammen (2000) argue that the former cannot adequately represent the dynamics of household cooking fuel choice. This theory assumes that multiple interactive socioeconomic, cultural, and environmental factors determine the choice of cooking fuel (Masera, Saatkamp & Kammen, 2000). The proponents assert that fuel stacking is common in both urban and rural areas of developing countries and relates to many cooking fuel usages where households select a combination of fuels from both upper and lower levels of the ladder. Therefore, this study is adopting the fuel stacking theory to help explain how socioeconomic, cultural, and environmental factors affect the choice of household cooking fuel. University of Ghana http://ugspace.ug.edu.gh 25 Figure 2.5 : Energy Ladder to Fuel stacking Source: (Van der Kroon, Brouwer & Van Beukering, 2013) The fuel stacking theory posits that there are no substitutes for cooking fuels and a particular cooking fuel is often chosen for specific tasks. Therefore, instead of just switching between cooking fuels as income increases, households use different cooking fuels at the same time depending on needs, budget, and preferences (ESMAP, 2003; Leiwen & O’Neill, 2003; Kowsari & Zerriffi, 2011). Based on this notion, if a family switches to a modern type of cooking fuel, traditional fuels and devices are normally kept and households only partially switch to use them when need be (Kowsari & Zerriffi, 2011). Nonetheless, Victor (2002) counter the universal heirachcy concept of the stacking theory with the premise that household energy transition is not a step-like progress. University of Ghana http://ugspace.ug.edu.gh 26 2.6 Conceptual Framework A conceptual framework is an interrelated set of ideas (theories) about how a particular phenomenon operates or is related to its parts. The framework serves as the basis for understanding the causal or correlational patterns of interconnections across ideas, observations, concepts, knowledge, and interpretations (Svinicki, 2010). Thus, the conceptual framework influences how an event is interpreted. Also, the ideas that constitute a conceptual framework support one another, express their respective phenomenon and establish a framework-specific philosophy (Jabareen, 2009). Figure 2.6: Conceptual framework Source: (author’s construct, 2020) Household cooking fuel choice Firewood Charcoal Kerosene Biogas LPG Electricity Cultural factors Cooking habit Taste and Preference Beliefs Socioeconomic factors Income Education Price of cooking fuel Household size Sex of household head Age of household head Energy Policy LPG Promotion Program Environmental factors Availability of cooking fuel Accessibility of cooking fuel Community characteristics Type of dwelling unit Cleanliness of cooking fuel University of Ghana http://ugspace.ug.edu.gh 27 Based on findings from the literature, a conceptual framework is designed to help answer the research questions. The conceptual framework above demonstrates the elements of cooking fuel choice. The preference for a particular cooking fuel stems from socioeconomic factors such as education, income, Price of cooking fuel, et cetera. (Berhe, 2014; Karimu, 2015; Soltani, Rahmani, Pour, Ghaderpour, Ngah & Misnan, 2019). According to the literature, household heads who have at least secondary education prefer cleaner cooking fuels to traditional fuels. Cultural factors such as beliefs, cooking habits, and taste and preferences are also impelling factors in the selection of household cooking fuel. Some households choose a cooking fuel that would add taste to their meals and others make a decision based on their place of abode. Also, the conceptual framework has drawn a link between environmental factors such as availability and accessibility of cooking fuel, community characteristics, and cleanliness of cooking fuel and how they affect the choice of cooking fuel. Many low-income households have adopted their coping strategies. They opt for less expensive cooking fuels such as firewood and charcoal instead of expensive cooking fuels like LPG and electricity. However, the government has also designed some interventions to make LPG more accessible and less expensive to users by disseminating clean sets of LPG stoves and accessories to both rural and urban locations across the country. This is aimed at increasing LPG usage and also reducing the usage of biomass fuel. Therefore, this intervention strategy is aimed at achieving the usage of LPG among households without any barriers to usage. University of Ghana http://ugspace.ug.edu.gh 28 2.7 Conclusion The works of literature reviewed indicate that different socio-economic and environmental factors affect the choice of household cooking fuel. Ghana has recently realized an appreciable increase in the percentage of LPG users from 22.3 percent in 2014 to 24.5 percent in 2017 (Ghana Statistical Service, 2019). The reasons for the increase in the number of LPG users are unknown. Therefore, the fuel stacking theory is adopted for the study to explain the factors that influence cooking fuel preference among households. University of Ghana http://ugspace.ug.edu.gh 29 CHAPTER THREE RESEARCH METHODOLOGY 3.0 Introduction The research methodology is a way of systematically solving a research problem through the steps adopted by a researcher (Kothari, 2004). This chapter focuses on the research design, study area, sources of data, sampling procedure, instruments used for data collection, and analysis. 3.1 Research Design The study used a cross-sectional design to examine relationships between variables. The quantitative method of gathering data was also used. Quantitative research is concerned with quantities and measurements aimed at making scientific generalizations and predictions in research activity (Biggar, 2008). It is also an approach for testing objective theories by examining the relationship among variables (Creswell, 2014). These variables, in turn, can be measured, typically on instruments, so that numbered data can be analyzed using statistical procedures and results presented in the form of graphs and tables. The results derived from quantitative data are usually presented in tables, bar graphs, and pie charts. In essence, since the study is not focusing on a specific class/type of household, quantitative method was employed for the purposes of providing numerical representation as well as to capture the variations of households and their choice of cooking fuel. 3.2 Profile of the Study Area The study was conducted in the Nsawam-Adoagyiri Municipality in the Eastern Region of Ghana. The choice of the study area was partly due to its peri-urban nature and limited research works conducted on the choice of cooking fuel among households in that area. The Nsawam-Adoagyiri University of Ghana http://ugspace.ug.edu.gh 30 Municipality is about 24km from the national capital, Accra, and covers an area of approximately 175 square kilometers. The area is situated in the southeastern part of the Eastern Region of Ghana between latitude 5’.45 N and 5’.58 N and longitude 0.07’W and 0.27’ W (Ghana Statistical Service, 2014). Weather conditions in the Municipality are generally cool due to its location in the wet semiequatorial climate, coupled with the double maxima rainfall, which records an average annual rainfall of between 125cm and 200cm. The first rainy season is usually between May to June, with the heaviest rainfall experienced in June and a second rainy season between Septembers to October. This usually account for an entire year farming practice. The highest 2 temperatures averaging 30ْC are recorded between March and April. With the lowest average temperature of 26ْC recorded in August. The Nsawam-Adoagyiri Municipality has a population of 86,000 representing 3.3 percent of the Eastern Region’s total population (2,633,154) (Ghana Statistical Service, 2014a). Out of the total population, 49.7 percent are males and 50.3 percent are females, with more than half of the population living in urban communities. The area also has a household population of 79,402 representing 92.3 percent of the municipality’s total population of 86,000 with an average household size of 4.1 persons per household. In this area, about 64 percent of the total population can speak and understand English and a Ghanaian Language. The economy of the area is predominantly agrarian with other economic activities in the area including manufacturing, craftsmanship, forestry and fishery, and services and sales. The housing stock of the Nsawam Adoagyiri Municipality is 9,953 with the two main sources of lighting in dwelling units in the municipality being electricity (68.5%) and kerosene University of Ghana http://ugspace.ug.edu.gh 31 lamp (20.9%). The major sources of cooking fuel in the area are wood fuel, charcoal, and LPG (Ghana Statistical Service, 2019). Figure 3.1: Map of Study Area Source: (Ghana Statistical Service, 2014) 3.3 Study population The study population involved households in the Nsawam-Adoagyiri Municipality. The respondents of the study were members of households who are eighteen years or older and did the cooking in their homes. This was to guarantee the ability of participants to provide enough information about their cooking fuels. University of Ghana http://ugspace.ug.edu.gh 32 3.4 Sample Size and Sampling Technique According to the Ghana Statistical Service (2014), the Nsawam-Adoagyiri municipality has a total population of 86,000 and an urban population of 50,864. Since the unit of analysis of the research involves households in a peri-urban area, the research population is the total number of households in urban areas in the Nsawam-Adoagyiri Municipality, which is 12,406. In determining the sample size and sampling procedure, the population of the study area, settlement, and diversity of household characteristics were some important factors considered. The Sample size for the study was calculated using the Taro Yamane formula (Yamane, 1973) with a 95% confidence level. The sample size is given as: n = N 1+N (e) ² Where n = sample size, N = Total population, and e = Error tolerance (0.05). Therefore, n = 12406 = 388 sample size. 1+12,406 (0.05)² The sampling technique adopted for the study was systematic random sampling. The researcher divided the population size by the sample size to obtain the interval for the systematic sampling procedure. Sample interval = 12406 = 31.9 ~32. 388 This means that every 32nd household in the population was chosen for the survey. Therefore, the researcher identified a starting point by randomly choosing number 26 which falls between 1 and University of Ghana http://ugspace.ug.edu.gh 33 the sample interval of 32. Hence, any 32nd household after household number 26 was selected for the survey. Subsequently, all other participating households were selected following the systematic random sampling procedure. This sampling method was appropriate for the study because it costs less and is simple to use. 3.5 Type and Sources of Data Two major sources of data were employed in the study: the primary source of data from the field and secondary data from books, journals, and newspapers were used to gather information. Primary data is a first-hand data source that a researcher uses to answer the study objectives and research questions. This type of data is usually obtained through interviews, surveys, observation or experiment, and focus group discussion (Salkind, 2010). Secondary data is also data that already exists and can be used by others other than the original collector and sometimes for a purpose other than it was originally intended (Salkind, 2010). This source of data can be accessed in books, newspapers, journals, and online. The primary data used for the research was obtained through the questionnaires administered to the participants. Secondary data was also obtained from the Institute of Statistical, Social and Economic Research and Regional Institute of Population Studies Library, journals, online sources, and books. 3.6 Data Collection Method In this research, data was collected using a semi-structured questionnaire (see Appendix 1) designed by the researcher which contained questions related to demography and also captured information about households such as the type of cooking fuel they use. The questionnaire was also used to elicit information such as factors influencing the type of cooking fuel used by households, their level of income, etc. To keep the flow of information, the questions were divided into four major parts. University of Ghana http://ugspace.ug.edu.gh 34 The first part was aimed at knowing the demographic background of the respondents, and the second part sought to gather information about households’ cooking fuel choice and kinds of fuels used, amount of fuel used for cooking, and monthly fuel expenditure. The third part of the questionnaire solicited information about households’ cooking practices. This was to enable the researcher to know about the cooking norms and local cultural practices of the households in the study area. Subsequently, the fourth part covered questions about households’ socio-economic and cultural characteristics. The questionnaire was pre-tested with selected households in the La Nkwantanang-Madina Municipality to assess the appropriateness of the design, clarity, and relevance of the questions before the data was collected. All necessary modifications were made after the pre-test to solicit relevant information related to the study objectives. 3.7 Data Analysis Data obtained from the field was coded and entered into STATA software for processing. Two types of analysis, descriptive and econometric analyses were used in analyzing the data. 3.7.1 Descriptive Analysis Descriptive statistical analyses were performed on the sample to obtain a clear understanding of the demography of the population. Measures of central tendency (means and other percentiles) were also computed and presented in tables to provide clear characteristics of the study population. Consequently, cross-tabulations were performed to test the level of association between the dependent variable (cooking fuel choice) and the independent variables (income, age, the gender of household decision-maker, and the type of dwelling unit of households). University of Ghana http://ugspace.ug.edu.gh 35 3.7.2 Econometric Analysis The data was analyzed using a Multinomial Logit model to achieve the objectives of the study. This model was chosen because it performs better with discrete choice studies (Lutkepohl, Judge, Hill, Griffiths & Lee, 1985). The Multinomial Logit Model was chosen because there was more than one cooking fuel alternative available to households. In the Nsawam-Adoagyiri Municipality, the popular cooking fuels available are firewood, charcoal, LPG, and electricity. It was assumed that households choose a particular cooking fuel to maximize their satisfaction. Therefore, if a household selects j at a time, then Y*ij is the maximum option utilized among the other sources of cooking fuel. The fuel category is also defined as a vector (Yi = [Yij]) of three dummy categories and the probability that j will be in the choice set of ‘i' is P (Y*>0). The model is also expressed as follows; 𝑃𝑟(𝑌𝑖=𝑗) = exp⁡(𝛽𝑖𝑗𝑋𝑖) ∑ exp⁡(𝛽𝑖𝑗𝑋𝑖) 𝑚 𝑗=0 (1) Where: Pr [Yi = j] represents the probability of choosing any of the cooking fuels like charcoal, kerosene, LPG, biomass, or electricity. With firewood being the reference cooking fuel set,  m is the number of cooking fuels in the choice set,  j = 0 (firewood)  xi is a vector of the exogenous variables  βi is also a vector of the estimated parameters. The above equation can be rearranged algebraically as follows; Pi = e (b0 + b1 x 1+…+b y x y) (2) 1 + e (b 0 + b1x1 +…+b y x y ) University of Ghana http://ugspace.ug.edu.gh 36 The following equation was also used to estimate the coefficient; ln[ 𝑃𝑖 1−𝑃𝑖 ]⁡= b0 + b1 x1 +...+ bv xv (3) The quantity Pi/(1-Pi) in the above equation (3) represents the odds ratio. The P-value also determines if a change in the predictor variable has a significant effect on the logit at the level of acceptance. Also, if the P-value is higher than the accepted confidence level, then there is no enough evidence to support that a change in the predictor variable affects the choice of cooking fuel. The dependent variable in the model is the choice of cooking fuel (firewood, charcoal, or LPG) with firewood being the reference choice. In the logit model, estimated coefficients measure the likely change in the predictor variable while holding all other predictor variables constant. A positive estimated coefficient indicates an increase in the possibility of a household choosing an alternative fuel. A negative estimated coefficient also specifies that the possibility that a household will change to alternative fuel is low. Table 1, therefore, represents a matrix of the explanatory variables which are described in table 2. Due to time constraint in the collection of data, other explanatory variables captured in the conceptual framework including energy policy were left out in the analysis. University of Ghana http://ugspace.ug.edu.gh 37 Table 3.1: Description of dependent variables Dependent variable/cooking fuel Variable description Liquified Petroleum Gas (LPG) A relatively cleaner cooking fuel as per the energy ladder to fuel stacking theory. It is the preferred output for this work since it is relatively safe. Charcoal This is produced from the harvesting of live trees and wastes generated from logging and sawmills. It produces less smoke as compared to firewood. Firewood This is woody biomass energy that is mostly collected on farms and burned in open stoves, usually a traditional three-stone stove and mud stove to produce heat for cooking. It produces lots of smoke in the process of burning. Indiscriminate tree cutting for use as firewood also leads to deforestation. Quality of housing Description Traditional A type of house whose designs differ in origin and Sometimes built to portray the culture of a household. This type of house is usually built with wood, clay, and brick. Modest A type of house with similar designs around its location. It is usually built with cement blocks and has simple designs. Modern This type of house usually has a confiscated design made by an architect. Such a house is usually built using the latest designs and quality building materials. University of Ghana http://ugspace.ug.edu.gh 38 Table 3.2: Explanatory variables used in the Multinomial Logit Regression and the expected effect on the choice of cooking fuel Variable Variable description Observations Expected effect on Mean (N) choice of cooking fuel logHh (Log) of household head (proportion 245 + 0.779 of women to men) Age age of household head 245 + 37.8 Fsize Family size 246 + 3.82 Lev_educ Highest level of education of household head 246 + 1.26 Dec_fel Decision maker of the type of cooking fuel 245 + 2.17 Hh_income log per average monthly 246 + 3.21 Household income Housn_qulty Quality of housing 246 + 1.04 Fel_cost Cost of fuel 246 + 59.66 Kit_typ Type of kitchen 246 - 0.26 Tas_pref Taste and preference 245 - 0.31 Fel_clen Cleanliness of fuel 245 + 0.318 3.8 Ethical Consideration Approval was obtained from the Municipal executives of the Nsawam-Adoagyiri Municipality to research the area. The researcher also sought the consent of the participants and made them aware of their right to refuse to participate, assured them of anonymity and confidentiality, and the use of the information being collected. The objective and drive of the study were explained to the University of Ghana http://ugspace.ug.edu.gh 39 responders (mainly household heads) in a clear language and their cooperation was also sought without any financial or material reward for participating. University of Ghana http://ugspace.ug.edu.gh 40 CHAPTER FOUR RESULTS AND DISCUSSION 4.1. Introduction This chapter concentrates on data analysis, interpretation, and presentation of results by utilizing the multinomial logit regression model to draw the relationship that exists between the choice of cooking fuel and some household characteristics. The survey instructed the respondents to specify the type of fuel used by their households for cooking. The listed types of cooking fuel included firewood, charcoal, kerosene, LPG, biogas, and electricity. However, kerosene, biogas, and electricity were dropped from the analysis because none of the households interviewed used them. 4.2 Demographic Characteristics of Sample A total of 388 respondents were sampled for the survey. However, 246 responded, representing a response rate of 63.92%. This rate of response was achieved due to the strategy of visiting respondents on weekends and evenings on weekdays when household heads or their spouses are available. Table 4.1 below describes the demographic characteristics of the respondents while Table 4.2 presents the frequencies and percentages of the explanatory variables. 4.2.1 Sex of Respondents A majority of the respondents were females (77%). This was expected since females are traditionally more involved in cooking decisions at home. Males constituted 23% of the total number of respondents. University of Ghana http://ugspace.ug.edu.gh 41 Table 4.1: Descriptive Characteristics Variable Frequency % Cooking Fuel Firewood Charcoal LPG 9 97 140 3.66 39.43 56.91 Sex Male Female 57 189 23 77 Highest level of Education Tertiary Secondary Primary No Formal Education 62 64 112 8 25.20 26.02 45.53 3.25 Household Average Income over the past 12 months Less than GH500 GH500-GH1000 GH1001-GH2000 GH2001-GH3000 GH3001-GH4000 GH4001-GH5000 More than GH5000 38 33 24 42 25 21 63 15.25 13.41 9.76 17.07 10.16 8.54 25.61 Quality of Housing Traditional Modest Modern 22 193 31 8.94 78.46 12.60 Type of Kitchen Private Shared No Kitchen 208 11 27 84.55 4.47 10.98 University of Ghana http://ugspace.ug.edu.gh 42 Decision maker for the type of fuel to purchase Wife Husband Joint Daughter Son Others 67 73 72 17 11 5 27.35 29.80 29.39 6.94 4.49 2.04 Source: Author’s field data, September 2020 Table 1.2 : Mean values for the explanatory variables Variable Description Mean Values Cooking Fuel Firewood = 0 Charcoal = 1 LPG=2 1.53 Sex Male= 1 Female = 0 0.779 Age 18-83 years 37.8 Family Size 1-11 members 3.82 Highest level of Education Tertiary= 0 Secondary=1 Primary =2 No Formal Education=3 1.26 Household Average Income over the Past 12 months GHS5000=6 Quality of Housing Traditional=0 Modest=1 Modern=2 1.04 Type of Kitchen Private=0 Shared=1 No Kitchen=2 0.26 Decision maker for the type of fuel to purchase Price Wife=0 Husband=1 Joint=2 Daughter=3 Son=4 Others=5 GH0 – GH200 2.07 GH59.66 Source: Author’s field data, September 2020 4.2.2 Age of Respondents Out of the 246 households surveyed, the minimum age among the respondents was 18 years, while the maximum age was 83 years, and the mean age of respondents being 38 years. 4.2.3 Family Size of Household From the survey, households had a mean family size of 4. The minimum size was 1 and the maximum was 11 members in a household. University of Ghana http://ugspace.ug.edu.gh 44 4.2.4 Price of Various Cooking fuels According to the survey, Table 4 shows that the mean price for all cooking fuels examined was GH¢59.66. This means that on average, many households spend about GH¢60 on cooking fuel every month. 4.2.5 Type of Kitchen The type of kitchens surveyed included private kitchens where the kitchen was utilized exclusively only by members of the household and shared kitchens where households living in compound houses would collectively share a common kitchen. The survey indicates that most households use a private kitchen. The majority of households, representing 84.55%, use private kitchens whilst 4.47% and 10.98% share and use no kitchens respectively. 4.2.6 Household Decision Maker for Cooking Fuel The decision-maker for the choice of fuel used in the household includes either the husband or wife or both husband and wife or daughter or son. The result from Table 4 indicates that the average household decision-maker for the choice of fuel was 2.07 which represents joint, thus both husband and wife decide on the type of cooking fuel. 4.2.7 Level of Education of Household Head Out of the total number of households surveyed, 45% had attained basic education as their highest level of education whiles 3.5% had no formal education. This indicates that close to 67% of the households were literates. This is in line with the statistics provided by the Ghana Statistical Service (2014), which indicates that about 86 percent of the population aged 11 years and older in the municipality are literate with about two thirds (63.7%) of the literate population being literate in English and a Ghanaian language and 24 percent being literate in the English language. University of Ghana http://ugspace.ug.edu.gh 45 4.2.8 Average Income of Households Out of the total sample, 29% of the households had their collective average income between GHS 500 and GHS1000, 27% had their income between GHS1001 to GHS 3000, 19% of households had their average income between GH3001 to GHS5000, and 25% had an average monthly income more than GH5000 over the past 12 months. This is illustrated in Figure 4.1. Figure 4.1: Distribution of the average monthly income of households Source: Author’s field data, September 2020 4.2.9 Quality of Housing Approximately 79% of the households lived in modest homes, 12.1% lived in modern homes, while 8.94% lived in a traditional setting. In chapter 3, a modest home was described as a house that is not large nor expensive. Also, modern homes were described as houses with complicated architectural designs made by an architect. Such houses are usually built using the latest designs and quality building materials. Lastly, traditional settings are homes constructed with clay, bricks, and wood. 29% 27% 19% 25% GHS 5000 University of Ghana http://ugspace.ug.edu.gh 46 4.2.10 Preference for Cooking Fuel From Figure 4.2, it can be seen that 57% of the total number of households prefer LPG as their cooking fuel, 39% of the households prefer Charcoal as their cooking fuel, and 4% use Firewood. This development is possible because of the rapid modernization in the area. Again, since more than 50% of the respondents have received some form of education, they are more likely to select LPG over charcoal and firewood as their source of cooking fuel owing to its efficiency. This is in agreement with Pandey & Chaubal (2011) and Adeyemi & Adereleye (2016) assertion that education can influence the choice of cooking fuel. Figure 4.2: Choice of Cooking Fuel Source: Author’s field data, September 2020 Firewood , 4% Charcoal, 39% LPG, 57% Firewood Charcoal LPG University of Ghana http://ugspace.ug.edu.gh 47 4.2.11 Sources of Cooking Fuel As depicted in Figure 4.3, there were diverse means by which households accessed cooking fuels. Specifically, about 88% of households purchased these cooking fuels, 11% self-collected, and 1% received cooking fuels as a gift. Figure 4.3: Source of Cooking Fuel Source: Author’s field data, September 2020 4.2.12 Household Cooking task and Fuel consumption As illustrated in Figure 4.4 below, about 71% of households indicated that cooking stew or soup consumed more fuel while another 27% observed that cooking solid foods consumed more fuel. Further, about 2% of households agreed that boiling water consumes more fuel. None of the households used the cooking fuel options to bake. University of Ghana http://ugspace.ug.edu.gh 48 Figure 4.4: Household cooking task and fuel consumption Source: Author’s field data, September 2020 4.2.13 Reasons for a Household’s Choice of Fuel Many households prefer a particular cooking fuel to others based on many reasons. From Figure 10 below, many households (47%) in the study area cited efficiency as the reason behind their cooking fuel choice. The other underlying reasons - cleanliness, safety, convenience, cheap price, and availability were represented in the following respective percentages 24%, 14%, 7%, 4%, and 3%. Furthermore, 1% of the households indicated that their choice of cooking fuel was dependent on the cooking fuel’s contribution to the taste of food. Price of fuel and household income have been discovered to influence households’ choice of cooking fuel (Stephen, 2011; Bansal, Saini & Khatod, 2013; Berhe, 2014; Tettey, 2014; Karimu, University of Ghana http://ugspace.ug.edu.gh 49 2015; Nlom & Karimov, 2015; Alhassan, 2018; Soltani, Rahmani, Pour, Ghaderpour, Ngah & Misnan, 2019). Environmental factors such as efficiency, convenience, safety, and cleanliness have also been demonstrated to affect the choice of cooking fuel (Rehfuess, Mehta & Prüss-Üstün, 2006; Karimu, 2015; Nyankone & Waithera, 2016; Nielsen, 2017). Additionally, research shows that the effect cooking fuel has on the taste of food is relevant in the choice of fuel (Atanassov, 2010; Kwakwa, Wiafe, & Alhassan, 2013; Suliman, 2013; Malla & Timilsina, 2014; GACC & GAC 2017; Zhang & Hassen, 2017; Makonese, Ifegbesan & Rampedi, 2018). Figure 4.5: Reasons behind Households’ cooking fuel choice Source: Author’s field data, September 2020 University of Ghana http://ugspace.ug.edu.gh 50 4.3. Socio-Economic Factors Influencing the Choice of Cooking Fuel 4.3.1 Income level of a household Bansal, Saini & Khatod (2013) and Nlom & Karimov (2015) have shown that the level of household income is key in determining the choice of cooking fuel. In Ghana, the level of income of a household influences household cooking fuel choices. This means that the likelihood of a family choosing LPG or electricity over firewood and charcoal increases with a higher level of household income (Stephen, 2011; Alhassan, 2018). Since convenient, clean, and efficient fuel is relatively more expensive, its usage is expected to increase with higher income per the Energy Ladder and Fuel Stacking theories. According to the energy ladder and fuel stacking theories, the use of cleaner, convenient, and efficient fuels increases with income. For household income, the higher the family size, the greater the likelihood of a household choosing charcoal as a cooking fuel relative to LPG. This can be attributed to larger families having lower household income per head and yet requiring more cooking fuel for relatively more quantities of meals at home. Hence, large families will resort to cheaper cooking fuel to save costs. The Pearson chi-square test presented in Table 4.3 suggests a positive relationship between the income level of households and the choice of cooking fuel (chi-square = 42.1841, df = 12, p<0.05). This means that a household with a relatively higher income would choose LPG over charcoal and firewood. The results from the regression analysis also indicate that households who earn relatively higher incomes are likely to choose LPG over charcoal and firewood. University of Ghana http://ugspace.ug.edu.gh 51 Table 4.3: Cross-tabulation of average household income over the past 12 months and choice of cooking fuel Value Df Asymp. Sig. (2-sided) Pearson Chi-Square 42.1841 12 0.000 Likelihood Ratio 44.6847 12 0.000 Note DF = degree of freedom 4.3.2 Gender of Household Decision-maker and Choice of Fuel The gender of a family’s decision-maker impacts the choice of cooking fuel (Pundo & Fraser, 2006). The data revealed that male decision-makers are more likely to use LPG than charcoal as compared to female household decision-makers who prefer firewood and charcoal to LPG. Figure 12 depicts the distribution of gender roles in household decision-making and its influence on the choice of cooking fuel. Figure 4.6: Influence of gender role in household decision making on choice of Cooking fuel Source: Author’s field data, September 2020 1.70% 25.90% 72.41% 4.20% 43.61% 52.12% Firewood Charcoal LPG Male Female University of Ghana http://ugspace.ug.edu.gh 52 From Figure 4.6, it can be seen that male decision-makers prefer LPG as cooking fuel to charcoal and firewood. Female decision-makers on the other hand prefer charcoal and firewood to LPG. This could likely be attributed to low-income levels among females which makes them unable to afford LPG. A study by Kwasi Fredua (2012) indicated that women are poorer compared to men in Ghana, and this further shows that economic inequality between men and women will likely have a negative rippling effect on issues related to the environment as women have equally important roles in maintaining a clean environment. The results from the Pearson chi-square test presented in Table 4.4 indicate a positive correlation between the gender of a household decision-maker and the choice of cooking fuel (chi-square = 7.52, df = 2, p<0.05). The test reveals that male decision-makers are more likely to use LPG than charcoal as compared to female household heads who prefer firewood and charcoal to LPG. This is likely because a majority of men do not have the experience in lighting up charcoal or firewood for cooking unlike majority of women who possess such experience through upbringing. Table 4.4: Chi-Square Test result for Gender Role in Household Decision-making and Cooking Fuel Choice Value Df Asymp. Sig. (2-sided) Pearson Chi-Square 7.52 2 0.023 Likelihood Ratio 7.83 2 0.020 Note DF = degree of freedom University of Ghana http://ugspace.ug.edu.gh 53 4.3.3 The Age of Household Head The age of a household head influences the choice of cooking fuel. According to Truneh (2014), older household heads prefer solid fuel like charcoal for cooking as against LPG. It is further explained that aged household heads are more conversant with traditional fuels like firewood and cow dung and are keener to preserve norms and traditions. Older household heads prefer solid fuels like charcoal for cooking as against LPG. Figure 12 illustrates the correlatio