REGIONAL INSTITUTE FOR POPULATION STUDIES (RIPS) UNIVERSITY OF GHANA EDUCATION AND FERTILITY IN URBAN POOR COMMUNITIES IN ACCRA, GHANA BY EMMA AKWELEY AMARH (10221973) THIS THESIS/DISSERTATION IS SUBMITTED TO THE UNIVERSITY OF GHANA, LEGON IN PARTIAL FULFILLMENT OF THE REQUIREMENT FOR THE AWARD OF MA POPULATION STUDIES DEGREE. JULY, 2014 University of Ghana http://ugspace.ug.edu.gh i ACCEPTANCE Accepted by the Faculty of Social Sciences, University of Ghana, Legon in partial fulfilment of the requirements for the degree of MA (Population Studies). ………………………………. PROF. A.F. ARYEE (SUPERVISOR) Date …………………………… University of Ghana http://ugspace.ug.edu.gh ii DECLARATION I hereby declare that, except for references to other studies which have been duly acknowledged, this dissertation is my own original work and that it has not been presented in whole or in part anywhere for another degree. ….…………………………… EMMA AKWELEY AMARH (Student) Date….……………………... University of Ghana http://ugspace.ug.edu.gh iii DEDICATION I dedicate this piece of work first to the Almighty God for his constant love, mercies and protection upon me throughout my period of study. Secondly, to my parents Mr and Mrs Ashitey Amarh for their support throughout my education, I say God richly bless you. University of Ghana http://ugspace.ug.edu.gh iv ACKNOWLEDGEMENTS I would like to wholeheartedly express my deepest thanks to first the Almighty God, the source of every good gift and perfect present for his loving care towards me in the preparation of this project. Secondly I would like to register my sincerest gratitude and profound appreciation to Prof. A.F. Aryee for his advice, criticisms, contributions; guidelines and instructions he invested into making this work a success. I really appreciate your efforts and may the good Lord bless you for that wonderful gesture you showed towards me. I also wish to acknowledge the staff of the Regional Institute for Population Studies (RIPS) for the opportunity to be part of their loving family. Furthermore, I wish to thank all the PhD students at RIPS especially Ernest, Adriana, Tobi, Reuben and Desmond for their assistance and support. Finally I would like to thank my parents once again, my entire family members and all my friends especially Charles, Moses, Charlotte, Akua, Abigail, Margaret and Godfred for their support throughout the study. I say God richly bless you. University of Ghana http://ugspace.ug.edu.gh v ABSTRACT This study examined the relationship between female education and fertility in urban poor communities, namely James Town, Usher Town and Agbogboloshie. Using the second round of the 2011 Urban and Health Project data conducted by the Regional Institute for Population Studies (RIPS), fertility levels for different educational categories were estimated. The unit of analysis for the study was women in their reproductive ages (15–49). A sample of five hundred and forty three (543) women in their reproductive ages was used for the study. Children ever born was used to measure fertility. Three methods of analysis, namely univariate, bivariate and multivariate were used for the study. The mean age of the women was 30.0 years. With regards to their highest level of educational attainment , more than half of the respondents (70.9%) had basic education (primary and middle/JSS) level of education and about one fifth (20.8%) of the women had secondary and higher level of education. Bivariate analysis showed that women with secondary and higher education had the lowest MCEB (0.81) compared to women with no education (2.96). The MCEB tends to decrease as the level of education of the women increases. The multivariate analysis revealed that age and marital status were the most significant variables influencing fertility behaviour in urban poor communities while the intermediate variables (contraceptive use and abortion experiences) as well as the other variables (locality, religion, occupation, ethnicity and household wealth) were not significant. Based on the findings of the study it is therefore recommended that, the Ministry of Education, non- governmental organizations (NGOs) and other stakeholders must make efforts to increase the levels of female educational attainment in order to reduce their risk of exposure to childbearing. University of Ghana http://ugspace.ug.edu.gh vi TABLE OF CONTENTS Title Page ACCEPTANCE ................................................................................................................. i DECLARATION .............................................................................................................. ii DEDICATION ................................................................................................................ iii ACKNOWLEDGEMENTS ............................................................................................. iv ABSTRACT ...................................................................................................................... v TABLE OF CONTENTS ................................................................................................. vi LIST OF TABLES ........................................................................................................... xi LIST OF FIGURES ...................................................................................................... xiii CHAPTER ONE ............................................................................................................... 1 INTRODUCTION ............................................................................................................ 1 1.0 Background of the Study .................................................................................... 1 1.1 Statement of the Problem ................................................................................... 4 1.2 Rationale of the Study ........................................................................................ 6 1.3 Objectives of the Study ...................................................................................... 7 1.4 Research Questions ............................................................................................ 7 1.5 Definition of Concepts ....................................................................................... 7 1.6 Organization of the Study .................................................................................. 8 CHAPTER TWO .............................................................................................................. 9 LITERATURE REVIEW ................................................................................................. 9 2.0 Introduction ........................................................................................................ 9 University of Ghana http://ugspace.ug.edu.gh vii 2.1 Education and fertility ...................................................................................... 11 2.1.0 Education and Fertility in Developed Countries .......................................... 11 2.1.1 Education and Fertility in Developing Countries ......................................... 12 2.1.2 Education and Fertility in Ghana .................................................................. 13 2.2 Fertility in Urban Poor Communities ............................................................... 15 2.3 Proximate Determinants of Fertility................................................................. 17 2.4 Socio-Demographic Determinants of Fertility ................................................. 19 2.5 Effects of Education on Fertility ...................................................................... 20 2.6 Education and Proximate Determinants of Fertility ......................................... 21 2.7 Conclusion........................................................................................................ 23 2.8 Conceptual Framework ......................................................................................... 23 2.9 Hypothesis ........................................................................................................ 26 CHAPTER THREE ........................................................................................................ 27 RESEARCH METHODOLOGY ................................................................................... 27 3.0 Introduction ...................................................................................................... 27 3.1. Data Source ...................................................................................................... 27 3.2 Sampling Design .............................................................................................. 27 3.3 Categorization of variables .............................................................................. 28 3.4 Data Analysis ................................................................................................... 29 3.5 Data Limitations ............................................................................................... 30 CHAPTER FOUR ........................................................................................................... 31 University of Ghana http://ugspace.ug.edu.gh viii Profile of the Study Area and Background Characteristics ............................................ 31 4.0 Introduction ...................................................................................................... 31 4.1 Profile of the Study Area .................................................................................. 31 4.2 Background Characteristics of the Respondents .............................................. 35 4.2.0 Age of the Respondents ............................................................................ 35 4.2.1 Educational Attainment of the Respondents ............................................. 36 4.2.2 Locality of Residence ............................................................................... 37 4.2.3 Occupation of Respondent ........................................................................ 37 4.2.4 Religion of Respondent ............................................................................ 38 4.2.5 Wealth Status ............................................................................................ 39 4.2.6 Respondents Ethnicity .............................................................................. 40 4.2.7 Marital Status ............................................................................................ 40 4.3 Intermediate Variables ..................................................................................... 41 4.3.0 Contraceptive Use ..................................................................................... 41 4.3.1 Induced Abortion ...................................................................................... 42 4.4 Dependent Variable .......................................................................................... 43 4.4.0 Children Ever Born ................................................................................... 43 CHAPTER FIVE ............................................................................................................ 44 Association between Background Characteristics and Fertility ..................................... 44 5.0 Introduction ...................................................................................................... 44 5.1 Education and Mean Children Ever Born (MCEB) ......................................... 44 University of Ghana http://ugspace.ug.edu.gh ix 5.2 Contraceptives and Mean Children Ever Born ................................................ 45 5.3 Locality of Residence and Mean Children ever Born ...................................... 46 5.4 Age and Mean Children Ever Born .................................................................. 46 5.5 Marital Status and Mean Children Ever Born .................................................. 47 5.6 Ethnicity and Mean Children Ever Born .......................................................... 48 5.7 Occupation and Mean Children Ever Born ...................................................... 49 5.8 Religion and Mean Children Ever Born ........................................................... 49 5.9 Wealth Status and Mean Children Ever Born .................................................. 50 5.10 Association between Education and Contraceptive Use .............................. 51 5.11 Association between Education and Abortion ............................................. 52 5.12 Abortion and Children Ever Born ................................................................ 53 CHAPTER SIX ............................................................................................................... 54 THE RELATIONSHIP BETWEEN EDUCATION AND FERTILITY ........................ 54 6.0 Introduction ...................................................................................................... 54 6.1 The effect of education on children ever born ................................................. 54 6.2 The effect of education and intermediate variables on children ever born ...... 55 6.3 Other determinants of children ever born among women in urban poor communities in Accra. ................................................................................................ 57 6.4 The overall effect of education and other factors on children ever born ......... 60 CHAPTER SEVEN ........................................................................................................ 66 SUMMARY, CONCLUSIONS AND RECOMMENDATIONS .................................. 66 7.0 Introduction ...................................................................................................... 66 University of Ghana http://ugspace.ug.edu.gh x 7.1 Summary of the Study ...................................................................................... 66 7.2 Conclusion........................................................................................................ 69 7.3 Recommendation(s) ......................................................................................... 69 REFERENCES ............................................................................................................... 70 University of Ghana http://ugspace.ug.edu.gh xi LIST OF TABLES Table 4. 1 Percentage distribution of respondents by Educational Attainment. ............. 36 Table 4. 2: Percentage distribution of respondents by Occupation. ............................... 38 Table 4. 3: Percentage distribution of respondents by Religious Affiliation. ................. 39 Table 4. 4: Percentage distribution of respondents by Wealth Status ............................ 39 Table 4. 5: Percentage distribution of respondents by Ethnicity .................................... 40 Table 4. 6: Percentage distribution of respondents by Induced Abortion ...................... 42 Table 4. 7: Percentage distribution of respondents by their Fertility Level. .................. 43 Table 5. 1: Education and Mean Children Ever Born (MCEB). ..................................... 45 Table 5. 2: Contraceptive Use and Mean Children Ever Born. ...................................... 46 Table 5. 3: Locality and Mean Children Ever Born. ...................................................... 46 Table 5. 4: Age and Mean Children Ever Born. ............................................................. 47 Table 5. 5: Marital Status and Children Ever Born. ....................................................... 48 Table 5. 6: Ethnicity and Children Ever Born. ............................................................... 48 Table 5. 7: Occupation and Mean Children Ever Born. ................................................. 49 Table 5. 8: Religion and Mean Children Ever Born. ...................................................... 50 Table 5. 9: Wealth and Mean Children Ever Born. ........................................................ 51 Table 5. 10: Percentage Distribution of Respondents by their Education Level and Contraceptive Use. .......................................................................................................... 52 Table 5. 11: Percentage Distribution of Respondents by their Education Level and Abortion. ......................................................................................................................... 52 Table 5. 12: Abortion and Children Ever Born. ............................................................. 53 Table 6. 1: Results of Linear Regression analysis of Education and Children Ever Born. ........................................................................................................................................ 55 University of Ghana http://ugspace.ug.edu.gh xii Table 6. 2: Multiple Linear Regression on the effect of education and intermediate variables on Children Ever Born. ................................................................................... 57 Table 6. 3: Linear regression of Socio-demographic characteristics and CEB. ............. 59 Table 6.4: Linear Regression of Children Ever Born, Education and other variables. .. 64 University of Ghana http://ugspace.ug.edu.gh xiii LIST OF FIGURES Figure 2. 1: Framework for Analysing Education and Fertility ..................................... 25 Figure 4. 1: A Map Showing Profile of the Study Area ................................................. 34 Figure 4. 2: Percentage distribution of respondents by Age. .......................................... 35 Figure 4. 3: Percentage distribution of respondents by Locality of Residence. ............. 37 Figure 4. 4: Percentage distribution of respondents by their Marital Status. .................. 41 Figure 4. 5: Percentage distribution of respondents by Contraceptive Use .................... 42 University of Ghana http://ugspace.ug.edu.gh 1 CHAPTER ONE INTRODUCTION 1.0 Background of the Study Education has been an important aspect of societal development. It is the process of acquiring knowledge, skills, values and attitudes to fully develop individual capacity to societal well-being (Cunningham, 2006). There is a relationship between education, human resource development and economic growth (United Nations Development Programme, 2011). Countries therefore place emphasis on educational policies in designing their plans to accelerate development. Education is generally held to be the foremost determinant of fertility. Highly educated women have more knowledge about their reproductive and health seeking behaviour than illiterate women. A great deal of research has been carried out to find out the linkages between female education and fertility. It has been argued that female education empowers women mainly through four ways. These are: their involvement in family decision – making, autonomy and control over household resources, knowledge and awareness of the modern world and inter spousal communication (Jejeebhoy, 1999). This research focused on two main domains of women: education and fertility. These two main domains are of special importance to women, as they have a strong and lasting impact on the roles women occupy over the course of their lives. In the past decades, Western societies have witnessed profound changes in the educational careers and in the childbearing behaviour of women (Blossfeld, 1995). In addition, there has been a considerable improvement in women’s educational qualifications in the Sub-Saharan Africa. Globally, males have higher education University of Ghana http://ugspace.ug.edu.gh 2 compared to females (Cullen & Luna, 1993). This is as a result of social roles and values placed on women and this has a potential impact on their fertility. These roles are likely to increase the fertility of women. In comparison to previous generations, women today spend more time in the educational system, achieve higher levels of education, and are increasingly able to leverage their higher education to get better jobs (Blossfeld, 1995). At the same time there has been a pronounced postponement of entry into motherhood. Due to the simultaneous increase in women’s educational participation and in women’s ages at first birth, the interplay between these two life domains has been an important topic of discussion in the scientific literature (Tesching, 2011). The importance of childbearing cannot be overemphasized as it helps to replace a dying population. It is also crucial in determining not only the size but also the age structure of a population. High fertility produces a young age structure, whereas low fertility produces an older age structure. Issues of fertility are a major concern all over the world both in developed and developing nations including Sub-Saharan Africa (Weeks, 2008). Fertility behaviour in Sub-Saharan Africa, like other parts of the world, is determined by biological and social factors (Weeks, 2012). Several factors have contributed to sustain relatively high levels of fertility in most of Sub-Saharan Africa. These factors include high levels of infant and child mortality, early and universal marriage, early child bearing as well as child bearing within much of the reproductive life span, low use of contraception and high social value placed on child bearing. In the face of perceived high infant and child mortality, the fear of extinction encouraged high procreation with the hope that some of the births would survive to carry on family lineage (Bigombe&Khadiagala, 2003). University of Ghana http://ugspace.ug.edu.gh 3 Over the last thirty years (1979 to 2009), Ghana has experienced fertility decline. This fertility decline has been from seven children per woman to four children (Ghana Statistical Service, 2009). The high fertility coupled with declining mortality, has contributed to an intercensal growth rate of around 2.5 percent per annum since 1960. High fertility exposes women of childbearing age to greater risk of morbidity and mortality (UNFPA, 2012). Though the current TFR for Ghana is 4.2, there are some significant regional variations. The lowest total fertility rates of 2.53 and 2.56 in 2000 and 2010 respectively were observed in the Greater Accra Region. A plausible explanation for the situation in the Greater Accra Region is a stall in the fertility decline. All the regions, except Greater Accra Region, experienced decline in TFR between 2000 and 2010, ranging from 0.04 percent in the Volta Region to 32.6 percent in the Ashanti Region (Ghana Statistical Service, 2013). Education of women persistently emerges as the single most powerful predictor of their demographic behaviour (Cleland, 2002). Research consistently shows that women who are empowered through education tend to have fewer children and have them later. If and when they become mothers, they tend to be healthier and raise healthier children, who then also stay in school longer. They earn more money with which to support their families, and contribute more to their communities’ economic growth. Indeed, educating girls can transform whole communities (Fitzgerald, 2011). According to Elo (1990), Education and fertility of females works through the improved survivorship chances of their children. The health seeking behaviour of educated mothers for the use of maternal and child health care services are better than that of the health behaviour of illiterate mothers. The effect of female education works independently regardless of the socio-economic status of the household. Improved child survivorship affects both the age at marriage and fertility. University of Ghana http://ugspace.ug.edu.gh 4 A study by Shapiro and Gebreselassie (2008, 2010) shows that work on education and fertility has been done at the national, regional and urban-rural areas within Sub- Saharan Africa. In Ghana studies have been conducted to estimate fertility levels, patterns and trends as well as to examine factors affecting fertility behaviour (Gaisie, 1981; Nyarko, 2005; Ghana Health Service and ICF Macro, 2009; Tawiah, 1984). However, little research has been done on education and fertility at the micro-level. This research attempts to find out the role of education on fertility in selected urban poor communities in Accra, Ghana. 1.1 Statement of the Problem The high fertility rates and the rapidly growing populations in Africa including Ghana are becoming serious problem as far as socio-economic development is concerned (Weeks, 2008). Fertility being a major component of population growth has for several years, been a subject of discussion especially the causes of high fertility in relation to population growth. As population keeps increasing, issues of population pressure, carrying capacity and environmental degradation have emerged. Some of these concerns are consistent with Malthusian theory which postulates that if population is not controlled, the population will outstrip the number of resources hence resulting in poverty and its related adverse effect (Weeks, 2008). The proportion of people living in urban poor slums and informal settlements has been increasing rapidly worldwide especially in the developing world (World Urban Forum (WUF), 2004). Two-thirds of Africa’s urban population lives in informal settlements (World Bank 2000; HABITAT 2001 and 2003; APHRC 2002a). According to UN- HABITAT (2012), the proportion of urban population who live in slums is very high in University of Ghana http://ugspace.ug.edu.gh 5 Sub-Saharan Africa with about 61.7% compared with North Africa which recorded 13.3% of the proportion of the population who live in slums. The urban population is growing at a rate of about 4.4% per annum, whereas the rural population growth rate has fallen below 2.0% (UN, 2000). This rapid growth has been attributed primarily to rapid rural-urban migration, high unemployment, poverty, informal economy, poor planning, natural disasters, social conflicts and high fertility rate (UN-HABITAT, 2012). High fertility rate in urban poor communities in Accra results in increase in unemployment rate, poor health services, retards socio-economic development as well as places constraints on land for development and education. As the economy of Ghana experienced decades of economic stagnation and poor governance, the unprecedented growth of urban populations created a new face of poverty associated with significant numbers of people living in informal settlements (Brockerhoff& Brennan, 1998). As the population of Ghana increases, majority of its population tend to live in urban areas. According to Population and Housing Census (2010) more than half of Ghana’s population live in urban areas. It further explained that the proportion of urban population increased from 23.1 in 1960 to 50.9 in 2010. There were increases in the proportions of urban communities in all regions from 1960 to 2010 (Ghana Statistical Service, 2013). Informal urban settlements will increasingly play a dominant role in defining progress toward national and international development agendas, including national policies and the Millennium Development Goals (MDGs). The Government of Ghana does not have polices targeted at reducing fertility in urban poor communities. This presents a challenge as to whether Ghana can reach its target of reducing TFR to 3.0 by the year 2020. Understanding the unique characteristics and University of Ghana http://ugspace.ug.edu.gh 6 determinants of the fertility of the urban poor will help to identify efficient fertility reduction strategies. The context specific determinants of fertility in urban poor communities have not been well understood. For instance, the dynamics of female education and fertility reduction is not clear. Fertility keeps stalling and even increasing in some communities though educational attainment is on the rise. 1.2 Rationale of the Study Almost all populations are increasing all the time (PRB, 2012). High rate of population growth is generally considered an impediment to socio-economic development. Increase in population puts pressure on the scares resources which are used for development (Weeks, 2012). Several studies have reported that female education leads to a significant decline in fertility which eventually leads to a slowing down of the rate of population growth. Evidence relating to the timing of the secular decline in fertility in the Western countries has concluded that, increased education for women leads to lower fertility (Skirbekk& Samir, 2012). Studies by Bongaarts (2010), Weeks (2012) among others can also attest to the fact that education has the tendency of reducing fertility in both the developed and developing world. Increase in women’s education has helped in fertility reduction because they might stay longer in school which delays their age at marriage and eventually lower their parity. In Ghana, though level of education is on the ascendancy in many parts of the country, the Total Fertility Rate (TFR) keeps stalling. There have been limited studies on the effect of maternal education on fertility in Ghana, specifically in urban poor communities which are characterized by high fertility. Thus, there remains a University of Ghana http://ugspace.ug.edu.gh 7 discrepancy in the increasing female enrolment at all level of education and the stalling fertility in Ghana. In most urban poor communities, fertility still remains high, though there is an increase in female enrollment. This study will thus clarify the linkage between female education and fertility. This will help in designing policies aimed at fertility decline. 1.3 Objectives of the Study The main objective of this research is to determine the relationship between the level of maternal education and fertility in urban poor communities in Accra, Ghana. The specific objectives are: i. To determine the level of education in urban poor communities in Accra, Ghana. ii. To determine the level of fertility in urban poor communities in Accra, Ghana. iii. To determine the relationship between other background characteristics of women and fertility in urban poor communities in Accra, Ghana. 1.4 Research Questions The research will answer the following questions i. What is the fertility level of women in urban poor communities in Accra, Ghana? ii. What is the relationship between maternal education and fertility in urban poor communities in Accra, Ghana? 1.5 Definition of Concepts Fertility can be defined as the actual birth performance of a woman (Weeks, 2008). Total fertility Rate (TFR) is the total number of births a woman would have by the end of her childbearing period if she were to pass through those years bearing children University of Ghana http://ugspace.ug.edu.gh 8 at the currently observed rates of age-specific fertility. For the purpose of the study, fertility would be measured as children ever born. (CEB) 1.6 Organization of the Study The study consists of seven chapters. Chapter One includes background information, statement of the problem, rationale of the study, objectives, research questions and definition of concepts. Chapter Two is made up of the various literature reviews on maternal education and fertility, conceptual framework and statement of hypothesis. Chapter Three involves the methodology used for the study. Chapter four describes the profile of the study area and uses descriptive statistics to examine the background characteristics of the respondents. Chapter Five discusses the association between background characteristics and fertility. Chapter Six uses linear regression to examine the relationship between education and fertility. Chapter Seven deals with the major findings of the study, conclusions drawn from investigations and recommendations made. University of Ghana http://ugspace.ug.edu.gh 9 CHAPTER TWO LITERATURE REVIEW This section reviews literature on education and fertility in developed countries, Sub- Saharan Africa and Ghana. It also examines fertility in urban poor communities and the proximate determinants of fertility. It further investigates socio-demographic determinants of fertility, the effects of education on fertility and education and the proximate determinants of fertility. 2.0 Introduction Population dynamics has been demonstrated in literature as a key factor that contributes towards sustainable development and poverty alleviation. It covers areas such as trends and changes in population growth, age structure, sex structure, family composition, urbanization and population density (Balk et al., 2009). The three principal components of population dynamics, fertility, mortality and migration have a way of affecting the demographic character of the population. Of these three main components of population dynamics, fertility can be said to be the major dynamic element (Loter, 1988). Between the year 1970 and 1975 to the year 1995-2000, the world’s total fertility has declined by 37%. Thus there has been a decline in the birth per woman from the level of 4.5 between the year 1970 to 1975 to 2.8 per birth within the year 1995 to 2000 (United Nations, 2001). This decline reflects different changes in reproductive behaviour especially in populous countries. Some findings (Betzig, 1986; Razi, 1980) suggest that before the onset of the fertility decline, a high social status was associated with high fertility, thus individuals of higher social standing were found to have more children compared to individuals of lower social standing. However, with a decline in fertility levels, high social standing has often been found to be associated with relatively low fertility (Coale& Watkins, 1986; Cochrane, 1979; Haines, 1992) and yet Fieder et al. University of Ghana http://ugspace.ug.edu.gh 10 (2005) have argued that the relationship between fertility and status still remain positive even after the decline in fertility. In the developing world, fertility rate as a whole declined by an estimated thirty six percent from 6.0 births per woman to 3.8 births per woman (United Nations 1995). This decline was recorded between the early years of 1960 to 1980. The reason for fertility decline in developing countries can be attributed primarily to increase in contraceptive use (Feyisetan & Casterline, 2000). There is little dispute surrounding this finding but considerable debate is raised about the causes of the increase in the contraceptive use. The causal contribution of changes in fertility desires seems to be an unresolved issue. As at date, the factors that have been theorized to be the cause of the decline can be grouped into demand and supply factors. The demand factors are concerned with factors that affect the demand of contraceptives. Thus among these factors is the desire, either permanently or temporarily to avoid pregnancy. Another aspect of these demand factors is the social, psychological and cultural variables. The other factors, supply factors are concerned with the supply of contraceptives accessible to individuals. In the same way these factors are determined by geographical accessibility of family planning services (Feyisetan & Casterline, 2000). There are a considerable number of literatures on the relationship between education and fertility. Some other few literatures also focus on the causes of the decline of fertility in developing countries. Despite this, there seem not to be enough study on the analysis of fertility by status measures such as education, occupation, income or wealth. This study thus focuses on a measure of status education, and how it impact on fertility, contributing toward bridging the knowledge gap in the area. University of Ghana http://ugspace.ug.edu.gh 11 2.1 Education and fertility This section talks about education and fertility in the developed countries, developing countries including Sub-Saharan Africa and Ghana. 2.1.0 Education and Fertility in Developed Countries Most developed countries such as Canada, Japan, Australia, the United States of America and New Zealand have been characterized with zero growth and declining growth (PRB, 2013). The slow or declining population in the developed world is attributable to high rate of female education at all levels; maternal and occupational role conflicts and or strains that encourage adoption of small family size; high prevalence rate of modern contraceptives; secularization of social reproduction and individualism and relativism with respect to fertility or reproductive behaviour. This has resulted in low fertility rate which is well below the replacement level of 2.1 births per woman. Over the past two decades from 1950s and early 1960s, fertility decline has continued but at a much slower pace, and in a few countries fertility has turned upward slightly, for example, in Denmark, Finland, Norway, and the United States. In the four decades from the late 1950s to the late 1990s the TFR of the developed world dropped by 44%, from 2.82 to 1.57 births per woman, with more than two-thirds of this decline occurring before the late 1970s (Bongaarts, 2002). Though the developed world has achieved low or decline fertility rate, it has significant social and economic consequences. These implications include aging of population and population stagnation or decline; old age/pension social security burden; health and burden of disease, particularly non-communicable diseases among the old-old (70+) population (Martin & Zücher, 2008; Weeks, 1999). University of Ghana http://ugspace.ug.edu.gh 12 2.1.1 Education and Fertility in Developing Countries Most growth of the world’s population happens in the developing world. Developing countries are countries with a lower standard of living, underdeveloped industrial base and low human development index relative to other countries. Sub-Saharan Africa is a region marked with rapid population growth. Countries in Sub- Saharan Africa have been well known for high fertility levels relative to the rest of the world (Indongo & Pazvakawambwa, 2012). The high rate of population growth in Sub- Saharan Africa, stemming primarily from declining infant or child mortality and sustained high fertility, has been a major impediment to the social and economic development of this region. Rapid population growth has contributed to economic stagnation, environmental degradation, and poverty for the masses of Africans (Kollehlon, 2003). The high fertility rate observed in Sub-Saharan Africa (SSA) and its potential adverse effects on the region’s development efforts led to substantial research, policy and action focused on identifying and addressing the various social, cultural and economic factors that served to maintain fertility at high levels in the region (Shapiro &Gebreselassie2008; Bongaarts & Sinding 2009). This policy include the implementation of voluntary family planning programmes that provided information about, and access to contraceptives that permitted women and men to control their reproductive lives and reduce unwanted childbearing (Bongaarts & Sinding, 2009). Fertility transition in Sub-Saharan Africa has been characterized by stalling in a number of countries to a greater degree than elsewhere in the third world (Bongaarts 2005, 2008; Westoff & Cross 2005; Shapiro & Gebreselassie 2010; Shapiro et al., 2013). Stalling is an interruption period in an ongoing fertility transition before a country University of Ghana http://ugspace.ug.edu.gh 13 reaches the end of a transition (Bong, 2006; Moultrie et al., 2008).For instance, in both Kenya and Tanzania fertility decline has stalled at the national level and continued to decline among the most educated women. In Zimbabwe, although fertility has continued to decline at the national level, stalling is observed among women with less than secondary education. These variations are as a result of changes in socio-economic variables and family planning programme (Ezeh et al., 2009). Women’s Education as a means of achieving lower fertility has become embedded in the ideology of major international organizations such as the World Bank and the United Nations Population Fund. It also emerged as one of the major themes of the 1994 International Conference on Population and Development held in Cairo. Thus education has long been recognized as a crucial factor influencing women's fertility (Bongaarts, 2003; Weeks et al., 2010). DeRose et al., (2002) acknowledge schooling to be the most promising means of reducing fertility in the developing world. Thus higher educational attainment raises a woman’s income thus lowering fertility and it also stimulates aspirations for higher standard of living and increased investments. 2.1.2 Education and Fertility in Ghana Fertility decline in Ghana was observed in the early 1990s. According to Caldwell et al., (1992) TFR in the early 1990s was 5.5. Ghana’s fertility issues received significant attention when the country was able to organize the foremost population census on the continent as early as 1960. The implementation of the 1969 Ghana’s population policy on family planning further created more awareness for fertility issues to receive significant attention (Locoh, 2002). Fertility decline has been rapid and phenomenal in Ghana. From 1993 to 1998 TFR dropped from 5.5 to 4.6. University of Ghana http://ugspace.ug.edu.gh 14 Fertility behaviour varies across space and time in Ghana. Fertility is influenced by a number of individual social factors. Even though population scientists are far from agreed on the importance of each factor and its respective contribution to fertility behaviour, the different fertility patterns observed suggest uneven development and variations in institutional and cultural contexts within the country. Admittedly, socio- economic development and modernisation, which accompanied urbanisation, schooling, child survival, the status of women, family planning programmes and fertility desires, have played an important role in these declines (Agyei-Mensah et al., 2005). In the 1992 constitution of Ghana, education has been recognized as one of the basic human rights. This is also supported by the Millennium Development Goal 2 (MDG 2) which is “to achieve universal primary education by the year 2015”. In Ghana, prior to independence there were few schools as compared to the present, but the number was relatively high in relation to other African countries. The schools were established by the colonial government. According to Antwi (1992), Kwame Nkrumah the first president of Ghana’s developmental plan for education in the year 1952 made quantitative progress in the development of education in the country. The educational act of Ghana makes it compulsory by providing that every child of school going age should attend a course of instruction in a school recognized for the purpose. The expansion in educational services is reflected in the reductions in literacy rates over the years. For example in 2000, the percentage of the Ghanaian population aged 6 and above who were literate was 43%, a decline from the 57% recorded during the 1970 census (Ghana Statistical Service, 2000, 1970). Thus, the decline in fertility has occurred during the same period as gains in education. In the 2010 population and housing census (PHC, 2010) there has also been an increase in the literacy rate. Seventy four percent of the population aged 11 years and older was literate. The proportion for University of Ghana http://ugspace.ug.edu.gh 15 males was 80.2% and that of females 68.5%. For urban and rural areas, the proportions literate were 84.1% and 62.8% respectively. Among the regions, Greater Accra reported the highest literacy rate of 89.3%, followed by Ashanti (82.6%) and Eastern region (81.0%). For the Northern, Upper east and Upper West regions, the proportions literate were lower than those illiterate. For instance, literacy rate in the Northern region was 37.2%, the lowest in the country. The percent of males who were literate was higher than those of females in all the regions. Two-thirds of the 11 years and older population is either literate in English only (20.1%) or English and a Ghanaian Language (45.8%). The population that is literate in a Ghanaian Language only is just 7.0%, while only 0.3% reported that they could speak and write in English and French. In the Ashanti Region, 57.6%, the highest proportion, were literate in both English and a Ghanaian language followed by the Eastern Region (53.4%). The lowest rate was in the Upper East region where only 14.0% of the population was literate in both English and a Ghanaian language. Although English is the official language in Ghana, only 20.1% of the population can read and write in English only, with the highest proportion of 34.9% in the Greater Accra region followed by the Upper East Region (32.0%). Volta Region had the least proportion literate in English only (10.8%) but comparatively the highest literacy rate (12.0%) in a Ghanaian Language only and Upper East the least (1.3%). Literacy in English and French is very low with less than one percent being literate in both languages in any region (Ghana Statistical Service, 2013). 2.2 Fertility in Urban Poor Communities Between 2010 and 2050, the world’s estimated population increase of 2.3 billion people is expected to be absorbed by urban areas in less developed regions, with Africa’s urban population projected to increase by 0.8 billion (United Nations Population University of Ghana http://ugspace.ug.edu.gh 16 Division,2010). Slums are places characterized by overcrowding, marginalization, harmful environmental exposure, poverty, social disadvantage, insecurity, lack of access to amenities, and poor health. These features are associated with increased risk of and vulnerability to natural and man-made hazards, particularly decreased health and well-being (Pelling, 2003; Davis, 2006). In other words, slums are universally considered to be vulnerable places. As the world’s population becomes increasingly urban, there is a need to examine the fertility and family planning needs of urban poor populations. It is often assumed that urban residents are better off than their rural counterparts. However, the reality is that the urban poor are equally disadvantaged because of overcrowding, high demand for limited resources, increased cost of services in urban settings, and lack of access to clean water and sanitation (Montgomery, 2009). With increased urbanization, urban centres are becoming over‐burdened by people seeking employment, housing, and a better way of living. As of 2010, about half (50.5%) of the world’s population was living in urban areas (United Nations, 2010). While Asia and Africa are less urban (40% and 42%, respectively), they are both projected to attain populations that are two‐thirds urban by 2050 (United Nations, 2010). Fertility levels tend to be negatively related to levels of urbanization and on average, most fertility indicators are better in urban areas than in rural areas. Increasingly, the vulnerability of the urban poor to undesirable health including reproductive health outcomes is being documented in different settings (Fotso, 2007, 2009a, and 2009b; Van de Poel et al., 2007; Montgomery, 2009). University of Ghana http://ugspace.ug.edu.gh 17 2.3 Proximate Determinants of Fertility Numerous academic scholars have contributed towards explaining why fertility has been declining in Sub-Saharan Africa. However, no single factor can wholly explain the changes in fertility in the Sub-regions (White et al., 2002). According to Bongaarts and Potter (1983) fertility is directly determined by a set of biological and behavioural variables called the proximate determinants. These proximate determinants include Age at First Marriage, Induced Abortion, Contraceptive Usage and Postpartum Infecundability. For instance, If an intermediate fertility variable, such as the prevalence of contraception changes, then fertility necessarily changes also (assuming the other intermediate fertility variables remain constant), while this is not necessarily the case for an indirect determinant such as income or education. Thus, fertility differs among populations and trends in fertility over time can always be traced to variations in one or more of the intermediate fertility variables. Each of the intermediate fertility variables may be described briefly as follows: Proportion of married people: This variable tends to measure the proportion of women of reproductive age that engages in sexual intercourse regularly. All women in sexual unions should theoretically be included, but to avoid difficult measurement problems, the analysis deals with only childbearing women living in stable sexual unions, such as formal marriages and consensual union. However, many women spend substantial proportions of their potential reproductive years out of marriage; either before first marriage or after a marriage has ended due to divorce, separation, or death of the husband. In addition, some women never marry. Contraception: Is any deliberate parity-dependent practice-including abstention and sterilization-undertaken to reduce the risk of conception. Thus, the absence of contraception and induced abortion implies the existence of natural fertility. University of Ghana http://ugspace.ug.edu.gh 18 Contraceptive practice is the intermediate fertility variable primarily responsible for the levels of fertility within marriage. In developing countries the practice of contraception is rare or virtually absent and marital fertility is relatively high. However, in the developed nations, marital fertility is lowest, and over half the married women in the reproductive years are current users of contraception. Education of women makes them have better access to information about contraception as well as giving women greater willingness to tolerate various side effects and inconveniences associated with the use of methods of contraception. Induced abortion: Is a practice that deliberately interrupts the normal course of gestation. Though reliable measurements of the prevalence of induced abortion are often lacking, it is well known that induced abortion is practiced in many societies. Even in cases where good estimates are available, it has proven difficult to determine the reduction in fertility that is associated with the practice of induced abortion. Postpartum infecundability: When a woman is pregnant, she remains infecundable (i.e., unable to conceive) until the normal pattern of ovulation and menstruation is restored. The duration of the period of infecundity is a function of the duration and intensity of lactation. Natural marital fertility is highest in the developed countries which are explained by the differences in lactation practices among countries at various levels of development. In modern western populations lactation is generally short, and many women do not lactate at all. In traditional societies in Africa, Latin America, and Asia, lactation is usually long and often lasts until the next pregnancy occurs. Lactation has an inhibitory effect on ovulation and thus increases the birth interval and reduces natural fertility. University of Ghana http://ugspace.ug.edu.gh 19 2.4 Socio-Demographic Determinants of Fertility Many studies (Skirbekk and Samir, 2012; Bongaarts, 2008) had been aimed at determining the factors associated with fertility. According to (Bongaarts, 2010) countries do not affect fertility change but rather, it is people classified according to social, economic and other ethnic characteristic that affect fertility. Some number of studies (Skirbekk and Samir, 2012; Bongaarts, 1983) link countries fertility with: current age, age at marriage, education, religion, ethnicity, occupation, contraception and abortion. Socio-demographic factors such as place of residence (Urban or Rural), occupation, religion and ethnicity has been found to have association or effect on the fertility (Weeks, 2008). Studies show that fertility levels are the highest in small towns and rural areas and the lowest in the capital city (Kulu, 2011). Urban residents have been discovered to have low fertility compared to rural residents, and this has been as a result of access to health facilities, modern contraceptive techniques and the harsh living conditions associated with urban residency (Weeks,2008). The influence of both religion and ethnicity on fertility had to do with norms, doctrines and traditions associated with being both a practitioner and a member of these categories (religion and ethnicity). Some of these norms are, desires for large families, acceptance and practice of family planning and contraceptive usage (Weeks, 2012). The occupation of women has a tendency of influencing their fertility. Compared to women who are not working, working women have on average fewer children. This could be because working women have limited flexibility in terms of maternity leave conditions, work commitments and this could negatively contribute to their fertility levels (Indongo and Pazvakawambwa, 2012). University of Ghana http://ugspace.ug.edu.gh 20 2.5 Effects of Education on Fertility The role of education in the decline of fertility has been a source of endless debate in the demographic literature. From a theoretical perspective, several causal channels have been emphasized. First, education raises a woman’s permanent income through earnings, tilting her optimal fertility choices toward fewer offspring of higher quality (Becker, 1960; Mincer, 1963; Becker and Lewis, 1973; Willis, 1973). Second, a woman’s education is causally connected to her mate’s education (Behrman and Rosenzweig, 2002), so that the effect of education on household permanent income is augmented through a multiplier effect. Third, education may improve an individual’s knowledge of, and ability to process information regarding, fertility options and healthy pregnancy behaviours (Grossman, 1972). Numerous studies have also documented the importance of increasing women's education as a key variable contributing to fertility decline in the developing world (Jejeebhoy, 1995; Rutstein, 2002; Bongaarts, 2008, 2010; Shapiro and Gebreselassie, 2008, 2010; Shapiro et al., 2013). Bongaarts (2010) has recently examined the causes of educational differences in fertility in Sub-Saharan Africa, and his analysis emphasised educational differences in desired family size as well as in demand for, use, and effectiveness of contraception. Improvements in women's educational attainment and reductions in infant and child mortality constitute important socio-economic changes. The growth in economic well- being is an additional form of socio-economic change. Shapiro et al.,(2013) found out that, contrary to the historical decline of fertility in the West, which was associated with increased levels of economic well-being, DHS data from the mid-1980s onward suggest that fertility decline in both Sub-Saharan Africa and in the developing world on the whole appears to be slower when economic growth is more rapid. This is attributable to University of Ghana http://ugspace.ug.edu.gh 21 education of women, urbanisation, decreased mortality and nuclearization of households (Locoh, 2002). Jeffery and Basu, (1996), Lutz and Goujon (2001) and Cleland (2002) suggest that the link between schooling and fertility is a transient phenomenon that could disappear when countries approach replacement fertility. In contrast to the view that fertility differentials by education initially widen and then shrink to low levels or nothing, John Bongaarts (2003), in his examination of data from 57 developing countries from all over the world, notes a tendency for the absolute difference in fertility between the lowest and highest education groups (no schooling and secondary and higher, respectively) to decline steadily as the transition proceeds. At the same time, he concludes that educational differentials in fertility remain substantial in the late and post- transitional stages and hence does not anticipate convergence as countries reach the end of their fertility transitions. 2.6 Education and Proximate Determinants of Fertility As at date, analysis of the determinants of fertility change is still on going. Evidence of some researchers suggests that fertility has no correlation with income or with changes in income (Garenne & Joseph, 2002). They explained that Kenya has undergone a steady growth of about fifty percent or even more in their income per capita in uniformity with their purchasing power within the period of 1960 to 1992. In contrast, Madagascar undergone a steady drop in their income per capita to about fifty percent or to a much lesser percentage yet still Kenya and Madagascar both has records of an early and steady decline in urban fertility. The same research by Garenne and Joseph (2002), suggests that there was no clear evidence of a relationship between a mother’s level of education and fertility level at the time of the onset of fertility decline. University of Ghana http://ugspace.ug.edu.gh 22 Garenne and Joseph (2002) also found that in Africa, there exist a correlation between contraceptive use and total fertility. Their findings reflect nothing new as it is often the case that the more individuals use contraceptives the lower will be their level of fertility. This finding strengthens the obvious reason why individuals use contraceptives to control their fertility level. Studies by Oppong (1983), Mahmood and Khan (1985) and Mahmood (1992) who analysed fertility in African countries found that fertility levels were systematically low. This means that in the countries they investigated, the cumulated fertility was much lower than could have been predicted from the level of contraceptive prevalence. This seems to suggest that induced abortion is playing a major role in fertility regulation in Sub-Saharan Africa. Some studies conducted by Caldwell (1981) and Sathar (1984) confirmed that there is a strong negative association between fertility preferences and education, and that the impact of wife’s education was stronger than that of the husband. They further explained that a substantial proportion of their respondents who were female (61%) and their husbands (31%) had never been to school. In contrast, more of their respondents who were educated were younger in age as compared to the older respondents most of whom had never been to school. Comparable conclusions could also be drawn with respect to the educational attainment of their husbands, indicating a possible upward trend in educational levels. However the initial three years of women’s schooling did not seem to have any significant impact on the fertility preferences and behaviour. But after “4–5” years of schooling there was a significant decline in the number of children ever born and living, and there was also a decline in the desired or ideal number of children the women would like to have. Other studies (Yusuf, 1988; Mahmood, 1992) have also found out that education of the husband was not a very good predictor of women’s fertility. University of Ghana http://ugspace.ug.edu.gh 23 2.7 Conclusion There are many contributing factors that affect fertility levels. Such factors include but are not limited to the literacy status of individuals, the age at which people first marry, the perceived ideal number of children one wants to have and wealth status. From the literature review, the above mentioned factors for instance, education (literacy status) has relationship with fertility. Another important aspect of the findings from literature seems to suggest that most women in developing countries who move to the urban sectors have high lifetime childbearing. Thus more birth is recorded for most women at the end of their years of staying in the urban sector. This suggests that urban life thus provides a very strong constraint that serve to change the fertility plan of women. Some previous studies have found that education and fertility are inversely related (Lexon, 2004). On the other hand, the effects of education on fertility have undergone serious revision. Furthermore, previous studies focused on women in the rural-urban areas and neglected the urban poor areas where education has a high tendency of affecting fertility. This research has attempted to give insight knowledge on education and fertility in the urban poor communities in Sub-Saharan African countries laying more emphasis on Ghana. 2.8 Conceptual Framework To improve the understanding of the factors of fertility change, it is necessary to analyse the mechanisms through which it occurs. A number of variables have been identified to have an effect on fertility. This cuts across a number of socio-economic variables through proximate determinants of fertility. Socio-economic variables are said to have a direct influence on the proximate determinants which then controls fertility. University of Ghana http://ugspace.ug.edu.gh 24 The socio-economic variables significant to the study and which was measured in the source of data are education, ethnicity, wealth quintile, religion, age, location, occupation and marital status. Bongaarts (1978) identified four intervening variables which are responsible for influencing fertility behaviour of any population. These are age at first marriage, contraceptives usage, induced abortion and postpartum amenorrhoea. In this study, education would be regarded as the main socio-economic variable which influences fertility. Education impacts on fertility through some intervening variables like contraceptive usage, induced abortion etc. Other independent socio-economic variables to be considered are ethnicity, wealth quintile, religion, age, occupation, location and marital status with children ever born as the dependent variable. University of Ghana http://ugspace.ug.edu.gh 25 Figure 2. 1: Framework for Analysing Education and Fertility Independent Intermediate Dependent Source: Modified Model Adapted from Bongaarts (1978) Education is expected to influence fertility by raising the age at first marriage. The higher a woman goes with constant education; marriage is postponed and hence lowers fertility mainly because the age at which she marries has been raised by education and consequently her reproductive life span is shortened. The higher the educational status of a woman, the greater her exposure to the awareness and use of contraceptive methods. The education thus helps her to choose the most appropriate and efficient method needed to reduce her chances of becoming pregnant. Fertility Children Ever Born Contraceptive Use Induced Abortion Level of Education No Education Primary Middle/JSS SSS& Higher Control Variables Age Religion Wealth quintile Ethnicity Marital Status Occupation Location/place of residence University of Ghana http://ugspace.ug.edu.gh 26 2.9 Hypothesis Educated women are more likely to have lower fertility than women with no education in urban poor communities in Accra, Ghana. University of Ghana http://ugspace.ug.edu.gh 27 CHAPTER THREE RESEARCH METHODOLOGY 3.0 Introduction This chapter discusses the research techniques and methods used in the study. It focuses on the following areas: data source, sampling design, operationalization and measurement of variables, data analysis and data limitations. 3.1. Data Source The study relied on the second round of the 2011 Urban Health and Poverty Project. This is a population-based survey by the Regional Institute for Population Studies (RIPS), University of Ghana, Legon. The survey was part of a project called the Population Training and Research Capacity for Development (PopTRCD), which was a collaborative project. The Project was a cross-sectional survey conducted in three urban poor communities in Accra namely James Town and Ussher Town which constitute Ga- Mashie and Agbogbloshie between 25th November and 22nd December 2011. The Urban Health and Poverty Survey seeks to contribute to knowledge on inequalities in health and human welfare between the urban poor and other sub-groups in Africa and to sensitize local and regional stakeholders on urban poverty and health issues. The survey brings together a variety of research interests in the areas of migration, community and environmental challenges, climate change, reproduction, contraception, child and maternal health, marriage, sexual health and behaviour, fertility preference, HIV/AIDS and other STIs, adolescent sexual and reproductive health, and nutrition. 3.2 Sampling Design The sampling design consists of selected 29 enumeration areas (EAs) in these localities and 40 households were sampled from each EA. After informed consent was obtained, University of Ghana http://ugspace.ug.edu.gh 28 household questionnaires were conducted with household members; individual questionnaires were also administered to respondents between the ages 15-49 (females) and 15-59 (males). However, this study focused on women in their reproductive ages 15-49 years. A sample of five hundred and forty three (543) women was filtered from the data set. 3.3 Categorization of variables This section dealt with how the independent, intermediate, control and the dependent variables were categorized and measured in the study. Education as the independent variable was measured using Level of Education and the question asked in the data was what the highest level of education is. Contraceptive use and induced abortion are the intermediate variables used for the study. Contraceptive was categorized into no contraceptive, modern methods, traditional methods and both methods (traditional and modern methods). The question used in categorizing contraceptive use into the following groups was “which methods have you or your partner ever used? Induced Abortion was measured by using the question in the data: “have you ever had an abortion”. Control variables used in the study were age (women in their reproductive ages 15-49) religion, marital status, ethnicity, occupation, locality and wealth status. Respondents were categorized into 5 year aged groups to help determine the effect of each age group on the dependent variable. Marital status was classified into married, living with a partner, not married, widowed, divorced and separated. Religion was measured as no religion, catholic, Protestants, Pentecostal/charismatic, other Christian, Islam and other religion which comprised of traditionalist/spiritualist and eastern religions). Respondents were categorized in relation to their locality of residence namely University of Ghana http://ugspace.ug.edu.gh 29 Agbogbloshie, JamesTown and Ussher Town. Ethnicity was measured in the following categories; Akan, Ga-dangme, Ewe, Guan, Mole-Dagbani and other ethnic groups which are made of the Gursi, the Mande and the Gurma. Occupation was categorized into no occupation, professional/managerial/clerical, sales and services, skilled and unskilled and other occupation. Wealth Status was categorized into poorest, poor, middle, rich and richest. Fertility is the dependent variable to be measured using children ever born and the questions asked in the data are: How many children are alive, How many sons dead and How many daughters dead. The addition of these three questions computed the dependent variable. 3.4 Data Analysis Three methods of analysis were used for the study. These are univariate, bivariate and multivariate. The unit of analysis for the study was women in their reproductive ages of 15-49. Frequencies and percentages were used to describe the socio-economic and demographic characteristics of the respondents at the univariate level. At the bivariate level, Analysis of Variance (ANOVA), Independent T- Test and Cross tabulations were used to compare the means of children ever born to the different educational levels of the respondents. Multivariate analysis used simple multiple regression to examine the relationship between the dependent variable (Children ever born) and various socio- economic characteristics like education, age of the respondents, marital status, religion, occupation, ethnicity and locality which were the independent variables used in the study. This was done to see the effect of education and other socio-economic variables on fertility. The same kind of regression was undertaken to determine the net effect of education on fertility by including the intermediate variables (contraceptive use and Level of Education No Education Primary Middle/JSS SSS& Higher Control Variables Age Religion Wealth Status Ethnicity Marital Status Occupation Location o Place of Residence University of Ghana http://ugspace.ug.edu.gh 30 induced abortion) into the model. All associations were carried out under 95% significance level. 3.5 Data Limitations Questions asked about the age of a woman may not be known by her. They may misreport on their exact ages. Again, women in old age tend to forget to include all the children they have given birth to. Children dead are consciously not mentioned due to the sad memory of the event and these can also affect the rates calculated. University of Ghana http://ugspace.ug.edu.gh 31 CHAPTER FOUR Profile of the Study Area and Background Characteristics 4.0 Introduction This chapter describes the profile area and examines the background characteristics of respondents used in the study. 4.1 Profile of the Study Area The study was conducted in three urban poor communities in Ghana namely James Town, Ussher Town and Agbogbloshie. James Town and Ussher Town constitute “Ga- Mashie” while Agbogbloshie is a migrant community. Ga-Mashie is located on the Atlantic Coast of the Greater Accra Region of Ghana. The area is referred to as Old Accra, due to the fact that it is where the original Gas first settled, thus making it the oldest community in Accra. Ga-Mashie covers an area of 100 hectares along the southwest coast of Accra. This division came about as a result of the influence of Europeans from the Netherlands, Britain and Denmark, who were allowed to build trading lodges on the coast in the 17th century (Mahama et al., 2011). Ga-Mashie is one of the major slum area located at the heart of Accra. It has a slum population of about 100,342 persons, with female population representing 51.8% and that of the males 48.2% (Based on 2010 Provisional census sex ratios for Greater Accra Region). Ga-Mashie is principally inhabited by the Gas, of the Ga-Adangbe tribe, although a considerable number of non-Gas reside in the community, including Akans, Ewes, Guans and Mossi-Dagomba, as well as other foreign groups. Most adults in this area are engaged in low skilled occupations which comprise petty trades, fishing, transport work, artisans and craftwork (www.wikipedia.com, Ghanadistrict.com). University of Ghana http://ugspace.ug.edu.gh 32 James Town and Ussher Town are located directly east of the Korle Lagoon and emerged as communities around the 17th century British James Fort and Ussher Fort on the Gulf of Guinea coast. These towns were developed by the end of the 19th century, and following the rapid growth of the city during the 20th century, they became areas of a dense mixture of commercial and residential use. Despite their central location in the capital city, the densely populated towns are characterised by poor housing conditions, inadequate social services and basic amenities, poor health outcomes, insecurity, and unstable incomes and livelihoods. James Town is a vibrant, smelly and noisy area, sound-tracked by the babble of radios and the roar of exhausts. This town was once the hub for a succession of Portuguese governments before the city became the capital of the British gold coast in 1877. The main sporting activity in James Town is boxing. There are more than twenty boxing schools in James Town. Some of these schools are Joshua “The Hitter’’Clottey, Ike “Bazooka” Quartey and Azumah “The Professor” Nelson. There is no other area in the world with this quantity of boxing schools, and there is no other place in the world that has produced so many world champion boxers in the last seventy five years (Mahama et al., 2011). Ussher Town is the centre of Accra which is linked to James Town along the high street. Ussher Fort which is one of the tourist site located in Jamestown is five hundred metres east of James Town Fort. James Town and Ussher Town are both fishing communities. They are popular tourist destinations for those seeking to see the remnants of Accra’s colonial past. Currently, there has been much emphasis on James and Ussher Town respectively in international research and help programs. The emphasis is mainly in the areas of health and socio- University of Ghana http://ugspace.ug.edu.gh 33 economic subjects of a dense urban population where the government regulation and infrastructure has been insufficient in providing for the requirements and safety of its people. Agbogbloshie, a former wetland and a suburb of Accra, is known for the destination of dumping of electronic waste from industrialized nations. The town covers approximately four acres and it is situated on the banks of the Korle-lagoon, north- west of Accra’s central business district. About forty thousand Ghanaians inhabit the area, most of who are migrants from rural areas (www.wikipedia.com). University of Ghana http://ugspace.ug.edu.gh 34 Figure 4. 1: A Map Showing Profile of the Study Area Source: Henry Rebecca and Clara Fayorsey (2002). Coping with Pregnancy: Experiences of Adolescents in Ga-Mashie Accra. Calverton, Maryland, USA: ORC Macro. University of Ghana http://ugspace.ug.edu.gh 35 4.2 Background Characteristics of the Respondents 4.2.0 Age of the respondents Age is considered an important factor which indirectly determines fertility due to its influences on the capacity of the women to conceive and the number of additional children she wants. The second round of the 2011 Urban Health and Poverty Project data captured age of the respondents in complete years. The age in complete years was categorized into five year age groups and is presented in Figure 4.2. Figure 4. 2: Percentage distribution of respondents by age. Source: Urban Health and Poverty Project Data, 2011. Figure 4.2 shows the respondents in the age groups 40–44 and 45–49 have the smallest proportion of mothers with 11% and 10.5% respectively. The largest proportions of mothers were in the age categories of 20–24 and 25–29 years constituting 19.9% and 17.1% respectively. The mean age of the study population was 30.0 years. Bongaarts 13.1 19.9 17.1 15.5 12.9 11 10.5 0 5 10 15 20 25 15–19 20–24 25–29 30–34 35–39 40–44 45–49 P er ce n ta g e d is tr ib u ti o n Categories of Age Mean age = 30 years University of Ghana http://ugspace.ug.edu.gh 36 and Potter (1983) observed in their analysis of fertility that the rate of childbearing varies with age of a woman. 4.2.1 Educational Attainment of the Respondents Education which is the main focus of this study is important in analysing fertility. Education affects fertility by influencing age at first marriage, fertility preference, knowledge and use of contraceptives and socio-cultural beliefs (Cleland, 2002; Jejebhoy, 1995). Education generally has a negative influence on fertility in that time spent in school leads to delay in marriage and therefore child bearing. Higher education a empowers woman in making decisions as to the number of children she may want to have, control her reproductive life through contraceptive use and spacing births. Table 4.1 shows the percentage distribution of respondents by education. From the table, 8.3% of the respondents had no education. More than half of the respondents (70.9%) had primary and middle/JSS level of education and 20.8% of the respondents had secondary/higher education. From the table, the percentage of respondents who have secondary and higher education is very low compared to respondents who have had middle/JSS education. However, about 92% of respondents interviewed had attained some degree of formal education. Table 4. 1Percentage distribution of respondents by Educational Attainment. Level of Education Frequency Percentage No Education 45 8.3 Primary 143 26.3 Middle/JSS 242 44.6 Secondary/Higher 113 20.8 Total 543 100.0 Source: Urban Health and Poverty Project Data, 2011. University of Ghana http://ugspace.ug.edu.gh 37 4.2.2 Locality of Residence Respondents were classified by the three communities: Agbogbloshie, James Town and Ussher Town. This variable is important in knowing the educational level and fertility dynamics among the women in their reproductive ages (15–49) of these areas in order to assist with the implementation of appropriate interventions. These localities have different geographical descriptions that may be influenced by their fertility behaviour. James Town and Usher Town are fishing communities in which settlements in the area are slightly clustered compared to Agbogbloshie. From Figure4.3, about 82% of the respondents were residing in Ussher Town and James Town and 18.2% resided in Agbogboloshie. Figure 4. 3: Percentage distribution of respondents by Locality of Residence. Source: Urban Health and Poverty Project Data, 2011. 4.2.3 Occupation of Respondent Fertility behaviour may be influenced by the type of occupation a woman engaged in. Women who are self-employed may have more children depending on their desire on family size and income, while women in professional and clerical occupations may Agbogbloshie, 18.2% James Town, 30.2% Ussher Town, 51.6% University of Ghana http://ugspace.ug.edu.gh 38 have fewer children because of the short periods of maternity leave and the nature of work (Shapiro and Tambashe, 1994). Table 4.2 shows the percentage distribution of respondents by occupation. 20.1% had no occupation. Of all the types of occupation, sales had the highest percent of women with 55.2% whiles the other types of occupation had the least with 24.7% of women. Table 4. 2: Percentage distribution of respondents by Occupation. Occupation Frequency Percentage No Occupation 109 20.1 Professional/clerical/Managerial 27 5.0 Sales/Services 300 55.2 Skilled/Unskilled 101 18.6 Other Occupation 6 1.1 Total 543 100.0 Source: Urban Health and Poverty Project Data, 2011. 4.2.4 Religion of Respondent Religion is one of the socio-economic variables in the education-fertility analysis. An individual’s sexual behaviour can be shaped by his or her religious beliefs and practices which invariably affect fertility. Various religions in Ghana have different doctrines with respect to their fertility behaviour (Nasir, 2012; Addai, 1999). Respondents in the 2011 Urban Health and Poverty Project data were asked about their religious affiliation. The results of the survey indicated that, majority of the respondents were Christians (82.1%), followed by Islam (Muslims) and other religion which comprised of traditionalist and eastern religion as shown in Table 4.3. University of Ghana http://ugspace.ug.edu.gh 39 Table 4. 3: Percentage distribution of respondents by Religious Affiliation. Religion Frequency Percentage No Religion 31 5.3 Catholic 29 5.3 Protestants 113 20.3 Pentecostal/Charismatic 253 46.6 Other Christian 51 9.4 Islam 55 10.1 Other Religion 11 2.0 Total 543 100.0 Source: Urban Health and Poverty Project Data, 2011. 4.2.5 Wealth Status Table 4.4 shows that a large proportion of the women interviewed were classified as belonging to the poorest and poor categories constituting 22.3% and 20.8% respectively. Also smaller proportions of mothers belong to the richer and richest categories 16.8% and 20.4% respectively. Table 4. 4: Percentage distribution of respondents by Wealth Status Wealth Status Frequency Percentage Poorest 121 22.3 Poor 113 20.8 Middle 107 19.7 Rich 91 16.8 Richest 111 20.4 Total 543 100.0 Source: Urban Health and Poverty Project Data, 2011. University of Ghana http://ugspace.ug.edu.gh 40 4.2.6 Respondents Ethnicity Ethnicity refers to the ethnic group that a person identifies him/herself with. Different ethnic groups have varied norms, beliefs and practices about marriage and childbearing (Brunette, 1996). Some ethnic groups are concentrated at different parts of the country. Due to their location, they might mostly be in urban or rural areas. Ethnic groups that find themselves located in urban areas have greater access to education and therefore might have lower fertility. Conversely, ethnic groups in the rural areas might tend to have higher fertility. The respondents were predominantly Ga-dangme and Akan in Table 4.5. The largest respondents recorded are the Ga-dangme constituting 53.6%. The second largest ethnic group is the Akan with the respondents adding up to 28.4%. The Ewe constituted 6.1% whiles other ethnic groups (Gurma, Gursi and Mande) constituted 7.7%. Table 4. 5: Percentage distribution of respondents by ethnicity Ethnicity Frequency Percentage Akan 154 28.4 Ga-dangme 291 53.6 Ewe 33 6.1 Guan 10 1.8 Mole-Dagbani 13 2.4 Other ethnic groups 42 7.7 Total 543 100.0 Source: Urban Health and Poverty Project Data, 2011. 4.2.7 Marital Status Marriage is important in fertility analysis. It is one of the basic institutions in every society. For maintenance and establishment of family life, marriage is the recognized institution (Nukunya, 2003). Women are exposed to frequent sexual intercourse when University of Ghana http://ugspace.ug.edu.gh 41 they marry which results in a number of births. This means that the possibility of raising the fertility level is high in marriage. Marital status of respondents was categorized into the following groups: married, living with a partner, not married, divorced, separated and divorced. Figure 4.4 shows the marital status of the respondents. 34.3% of the respondents were not married whiles 46.7% of the respondents constitute those who were married and those not married but living with a partner. Divorced, separated and widowed represent 19% of the respondents. Figure 4. 4: Percentage distribution of respondents by their Marital Status. Source: Urban Health and Poverty Project Data, 2011. 4.3 Intermediate Variables 4.3.0 Contraceptive Use Contraception is the temporary prevention of fertility. Effective contraceptive use reduces the number of unwanted pregnancies and in effect reduces fertility. Contraceptive use was categorized into no contraceptive, modern, traditional and both methods (Modern and Traditional methods).From Figure 4.5, 49% of the respondents had never used any method of contraception whiles 35.5% of the respondents had used 23.9 22.8 34.3 2.8 6.8 9.4 0 5 10 15 20 25 30 35 40 Married Living with a partner Not Married Widowed Divorced Separated P er ce n ta g e D is tr ib u ti o n Marital Status University of Ghana http://ugspace.ug.edu.gh 42 only modern methods of contraception. 7.4% of the respondents had used only traditional methods of contraception while 8.1% of the respondents had used both methods of contraception. Figure 4. 5: Percentage distribution of respondents by Contraceptive Use Source: Urban Health and Poverty Project Data, 2011. 4.3.1 Induced Abortion Induced abortion is any intended intervention that leads to the termination of a pregnancy not resulting in a live birth no matter the duration of the pregnancy. From Table 4.6, 71.1% of the respondents had never had induced abortion. 28.9% of the respondents had engaged in induced abortion. Table 4. 6: Percentage distribution of respondents by Induced Abortion Induced Abortion Frequency Percentage Yes 157 28.9 No 386 71.1 Total 543 100.0 Source: Urban Health and Poverty Project Data, 2011. 49 35.5 7.4 8.1 0 10 20 30 40 50 60 No Method Modern Methods Traditional Methods Both Methods P er ce n ta g e D is tr ib u ti o n Methods of Contraception University of Ghana http://ugspace.ug.edu.gh 43 4.4 Dependent Variable 4.4.0 Children Ever Born From Table 4.7 the mean children ever born to respondents are about 2. About one third of the respondents had no child. 18.6% of the respondents had a child. The smallest proportion of mothers who gave birth from 10 to 15 children represents 0.6%. Table 4. 7: Percentage distribution of respondents by their Fertility Level. Children ever born Frequency Percentage 0 175 32.2 1 101 18.6 2 82 15.1 3 76 14.0 4 41 7.6 5 32 5.9 6 15 2.8 7 8 1.5 8 10 1.8 10 1 0.2 13 1 0.2 15 1 0.2 Total 543 100.0 The mean of children ever born = 1.9908 Source: Urban Health and Poverty Project Data, 2011. University of Ghana http://ugspace.ug.edu.gh 44 CHAPTER FIVE Association between Background Characteristics and Fertility 5.0 Introduction In this chapter, the associations between the demographic and the socio-economic variables of interest (education, contraceptive use, induced abortion, age, religion, wealth quintile, occupation, locality of residence, marital status and ethnicity) and fertility are examined. The analysis was carried out because fertility is affected directly or indirectly by these characteristics of the woman. To assess the associations, analysis of variance (ANOVA), cross tabulations and independent t-tests were used to test whether there is a significant difference between the mean number of children ever born by the demographic and socio-economic characteristics. 5.1 Education and Mean Children Ever Born (MCEB) From the ANOVA, Table 5.1 shows that the education level of women in an urban poor community has a significant association with the mean children ever born (p=0.000), which means that a woman’s level of education has an influence on her fertility level. MCEB to a woman decreases as the educational level rises. Women with no education had a MCEB of 2.96 while those of secondary and higher had a MCEB of 0.81. The findings from this study are comparable to what was reported by the Measurement, Learning and Evaluation project carried out in Kenya (MLE, 2012) where those with no education had 2.7 more children than those with secondary or higher education. University of Ghana http://ugspace.ug.edu.gh 45 Table 5. 1: Education and Mean Children Ever Born (MCEB). Level of Education MCEB Standard Deviation No Education 2.96 2.10 Primary 2.96 2.46 Middle/JSS 1.95 2.04 Secondary/Higher 0.81 1.16 Total 1.99 2.14 F = 22.054 p-value = 0.000 Source: Urban Health and Poverty Project Data, 2011. 5.2 Contraceptives and Mean Children Ever Born Contraceptive usage also has an influence on the number of children ever born to a woman. Table 5.2 shows a significant association with MCEB and contraceptive use (p value of 0.003). Women who had used both modern and traditional contraceptives had a higher MCEB (2.41) and those who had used only traditional methods had the lowest MCEB (1.18). Robey et al., (1992) reported that contraceptive use has become more common in developing countries and much of this increase has been in the form of modern methods (voluntary sterilization, oral contraceptives, intrauterine devices (IUD), condoms, injectables and vaginal methods) of fertility control. According to Trussel and Kost (1987) modern contraceptive methods are generally more effective in preventing pregnancy than traditional methods, although effectiveness varies with the quality of practice. However, findings from this study indicate that women in urban poor areas who had used modern methods of contraception had a higher MCEB (2.41) compared to women who had used only traditional methods. This may be due to the fact modern methods of contraception may be expensive for the women in urban poor communities so they might prefer the traditional methods of contraception which is easier to use and less expensive to control their fertility. University of Ghana http://ugspace.ug.edu.gh 46 Table 5. 2: Contraceptive Use and Mean Children Ever Born. Contraceptive Use MCEB Standard Deviation No Contraceptives 1.81 2.17 Modern Methods 2.31 2.07 Traditional Methods 1.18 1.50 Both Methods 2.41 2.44 Total 1.99 2.14 F = 4.723 p-value = 0.003 Source: Urban Health and Poverty Project Data, 2011. 5.3 Locality of Residence and Mean Children Ever Born The respondent’s locality of residence was significantly associated with her number of children ever born (p=0.042), which means that there are differences by the location. For instance, the mean children ever born to a woman in Agbogbloshie was higher (2.28) compared to a woman in Ussher Town (2.08). James Town residents had lowest MCEB (1.66). Table 5. 3: Locality and Mean Children Ever Born. Locality MCEB Standard Deviation Agbogboloshie 12.2 2.31 James Town 1.66 1.88 Ussher Town 2.08 2.20 Total 1.99 2.14 F = 3.178 p-value = 0.042 Source: Urban Health and Poverty Project Data, 2011. 5.4 Age and Mean Children Ever Born The overall age of the women in their reproductive ages was found to be significantly associated with mean children ever born. Generally MCEB to a woman increases as the University of Ghana http://ugspace.ug.edu.gh 47 age of the woman also increases. As expected, younger women had fewer MCEB than older women. Women in the age category 15-19 had a mean number of children ever born of 0.15 while women in the category 45-49 had mean children ever born of 4.00. In a study by Adhikari (2010), similar findings showed that older women (45-49) have a significantly higher MCEB compared to younger women (15-19). Table 5. 4: Age and Mean Children Ever Born. Age MCEB Standard Deviation 15-19 0.15 0.40 20-24 0.63 0.90 25-29 1.52 1.40 30-34 2.34 1.47 35-39 2.96 1.84 40-44 3.82 2.89 45-49 4.00 2.43 Total 1.99 2.14 F = 56.927 p-value = 0.000 Source: Urban Health and Poverty Project Data, 2011. 5.5 Marital Status and Mean Children Ever Born Marital Status and Mean Children Ever Born are significantly related (p=0.00). Marriage is one of the basic institution in which childbearing is legally recognized in every society because women are exposed to frequent sexual intercourse. From Table 5.5, women who were either married or had ever married (widowed, separated, and divorced) had higher MCEB compared to those who had never married. Widowed women had a MCEB of 4.60 while those not married had a MCEB of 0.45. University of Ghana http://ugspace.ug.edu.gh 48 Table 5. 5: Marital Status and Children Ever Born. Marital Status MCEB Standard Deviation Not married 0.45 0.97 Married 3.24 2.32 Living with a partner 2.33 1.88 Widowed 4.60 2.75 Divorced 2.57 2.15 Separated 2.43 1.58 Total 1.99 2.14 F = 49.630 p-value = 0.000 Source: Urban Health and Poverty Project Data, 2011. 5.6 Ethnicity and Mean Children Ever Born The association between ethnicity and mean children ever born was not significant (0.966). However, Guan women had the highest MCEB (2.50) compared to Ga- Dangme women (1.99) who form the highest proportion in the study area. Table 5. 6: Ethnicity and Children Ever Born. Ethnicity MCEB Standard Deviation Akan 1.93 2.32 Ga-dangme 1.99 2.12 Ewe 1.91 1.81 Guan 2.50 2.51 Mole-Dagbani 2.23 1.96 Other Ethnic Groups 2.07 1.99 Total 1.99 2.14 F = 0.192 p-value = 0.966 Source: Urban Health and Poverty Project Data, 2011. University of Ghana http://ugspace.ug.edu.gh 49 5.7 Occupation and Mean Children Ever Born The influence of women’s occupation on the mean number of children ever born was significant with a p value of 0.00. Women who had ‘other’ occupation had the lowest MCEB of 0.83 while those who engaged in sales or services had the highest MCEB of 2.57. Table 5. 7: Occupation and Mean Children Ever Born. Occupation MCEB Standard Deviation No Occupation 1.04 1079 Prof/Clerical/Managerial 1.03 1.42 Sales/Services 2.75 2.22 Skilled/Unskilled 1.62 1.86 Other Occupation 0.83 0.98 Total 1.99 2.14 F = 14.796 p-value = 0.000 Source: Urban Health and Poverty Project Data, 2011. 5.8 Religion and Mean Children Ever Born From Table 5.8, religious affiliation has no significant association with MCEB (p value of 0.36). Women belonging to other religion (Traditionalist and Eastern Religion) had the highest MCEB of 3.36 while women belonging to the Pentecostal/charismatic religion had the lowest MCEB of 1.85. Caldwell et al., (1992) found out that among those who practice the African traditional religion high fertility is morally right and correct due to the fact ancestors are honoured and their spirits are appeased through the bearing of children as descendants. University of Ghana http://ugspace.ug.edu.gh 50 Table 5. 8: Religion and Mean Children Ever Born. Religion MCEB Standard Deviation No Religion 2.22 2.43 Catholic 2.28 2.20 Protestants 2.00 2.30 Pentecostal/Charismatic 1.85 2.00 Other Christian 2.06 1.67 Islam 2.00 2.05 Other Religion 3.3 4.15 Total 1.99 2.14 F = 1.096 p-value = 0.364 Source: Urban Health and Poverty Project Data, 2011. 5.9 Wealth Status and Mean Children Ever Born Generally, there is a positive relationship between poverty and fertility. Poor people are more likely to have more children than rich people. The reason could be that since they cannot lose both, one should compensate for the other. For poor people, children become a source of social security. On the other hand, people with more money have a greater ability to cater for more children and therefore most likely to have more children. Whatever way, one looks at it; there is some form of interrelationship. Thus wealth quintile is an important variable to consider in this study. From Table 5.9 women belonging to the poorest quintile had a higher MCEB (2.35) while those in the richest quintile had the least MCEB (1.65). University of Ghana http://ugspace.ug.edu.gh 51 Table 5. 9: Wealth and Mean Children Ever Born. Wealth MCEB Standard Deviation Poorest 2.35 2.47 Poor 2.01 1.97 Middle 1.89 2.21 Rich 2.02 1.84 Richest 1.65 2.05 Total 1.99 2.14 F = 1.664 p-value = 0.157 Source: Urban Health and Poverty Project Data, 2011. 5.10 Association between Education and Contraceptive Use The association between education and contraceptive use was marginally significant (p=0.050) Therefore, the educational status of urban poor women had an influence on their contraceptive use. As indicated in Table 5.10, a little above one half (51.1%) of women with no education had never used any method of contraception while 43.4% of women with secondary or higher education had never used any method of contraception. Education is closely linked to the use of contraception. Educated women are more likely to use family planning methods. Shapiro and Tambashe (2002) found out that one’s level of education directly affects the level of knowledge and use as well as the choice of contraceptive methods. University of Ghana http://ugspace.ug.edu.gh 52 Table 5. 10: Percentage Distribution of Respondents by their Education Level and Contraceptive Use. Educational Status Percentage in Type of Contraceptive use No Methods Modern Contraceptives Traditional Contraceptives Both Methods Total (%) Total (Freq.) No Education Primary Middle/JSS Secondary/Higher Total 51.1 53.1 48.8 43.4 49.0 33.3 33.6 38.0 33.6 35.5 6.7 4.2 5.4 15. 7.4 8.9 9.1 7.9 7.1 8.1 100 100 100 100 100 45 143 242 113 543 χ2=16.938 Degrees of freedom =9 p value=0.050 Source: Urban Health and Poverty Project Data, 2011. 5.11 Association between Education and Abortion The association between education and abortion was not significant (p=0.184); therefore the educational status of urban poor women is not related to their abortion experience. Studies suggest that women with high educational attainment may be more prone to abortion experiences as they terminate pregnancies in order to fulfil their educational goals (Adanu & Tweneboah, 2004). Table 5. 11: Percentage Distribution of Respondents by their Education Level and Abortion. Education Level Abortion Percentage Total Yes No No Education Primary Middle/JSS Secondary/Higher Total 17.8 31.5 31.4 24.8 28.9 82.2 68.5 68.6 75.2 71.1 45 143 242 113 543 χ2value =4.840 Degrees of freedom=3 p-value=0.184 Source: Urban Health and Poverty Project Data, 2011. University of Ghana http://ugspace.ug.edu.gh 53 5.12 Abortion and Children Ever Born The association between abortion and children ever born was not significant with a p value of 0.542. Women who have ever experienced an induced abortion had MCEB of 2.30 while women who had never experienced an abortion had a MCEB of 1.87. This finding is contrary to other research works who found out that women who have abortions had lower MCEB compared to women who have never had an abortion. Table 5. 12: Abortion and Children Ever Born. Abortion MCEB Standard Deviation Yes 2.30 2.13 No 1.87 2.13 F = 0.372 p-value = 0.542 Source: Urban Health and Poverty Project Data, 2011. University of Ghana http://ugspace.ug.edu.gh 54 CHAPTER SIX THE RELATIONSHIP BETWEEN EDUCATION AND FERTILITY 6.0 Introduction This chapter discusses the relationship between education and fertility using the linear regression analysis to determine the effect of education and other socio-economic variables on fertility. Multiple linear regression analysis is a statistical analysis tool used for studying the relationship between a continuous dependent variable and two or more independent variables (whether continuous or categorical). Dummy variables were created for all categorical variables. This was done by coding the category of interest as “1” while the others assume the value of “0”. By doing this, a variable of three categories may end up with two dummies with the other set as a reference category. The variables are then put into the regression model except for the one used as the reference category. Four multiple linear regression models were run. In the first model, the education variable was regressed on the children ever born (CEB), while the second model considered education, the intermediate variables (induced abortion and contraceptive use) and CEB. For the third model, education and the control variables were regressed on CEB, and the fourth model considered all the variables of interest. 6.1 The effect of Education on children ever born Model 1 in Table 6.1, displays the effect of education on the MCEB to a woman in her reproductive ages 15-49 in urban poor communities in Accra. The model shows that education alone explains only 10.4 percent of the variation in the mean number of children ever born to a woman in urban poor communities. Thus, about 89.6 percent of the variation is due to other variables. The level of education has a strong significant University of Ghana http://ugspace.ug.edu.gh 55 (p<0.05) association with the mean number of children ever born to a woman. The results show negative associations between Middle/JHS and Secondary/Higher educated women and their CEB; hence, women with higher levels of education experienced fewer births than those with no education. This finding is consistent with that of Bongaarts (2010) who examined the causes of educational differences in fertility in Sub-Saharan Africa and stated that the higher a woman’s educational level, the less number of children she would have. Table 6. 1: Results of Linear Regression analysis of Education and Children ever born. Source: Urban Health and Poverty Project Data, 2011. RC: Reference Category p<0.05 6.2 The effect of education and intermediate variables on children ever born Table 6.2 describes model 2 which displays the effect of education and the intermediate variables on children ever born. The model indicates that when education, contraceptive use and abortion were regressed on children ever born the association was significant (p value<0.05). The adjusted R2 suggests that about 12 percent of the variation in the MCEB can be explained by the level of education, contraceptive use and abortion. This model compared to model 1, shows children ever born to a woman decreased for women with primary and middle level of education when contraceptive use and abortion was introduced. Women with high levels of education are more likely to use Variable Model 1 Children Ever Born Adjusted R2 = 0.104 Beta(β) Std. Error Sig. 95% Confidence Level Intercept 2.956 0.302 0.000 2.362 3.549 Education No Education (RC) Primary -0.263 0.346 0.447 -0.943 0.417 Middle/JSS -1.005 0.329 0.002 -1.651 -0.359 Secondary & Higher -2.150 0.357 0.000 -2.852 -1.449 University of Ghana http://ugspace.ug.edu.gh 56 contraception and use them effectively, thus influencing their MCEB. Tawiah (1997) revealed that there was a positive effect of secondary and higher levels of education on contraceptive use among women. The results from Model 2 indicate that modern contraceptive use is significant with a p value= 0.010. However, with women who had ever used modern contraception there was an average increase in 0.501 children ever born compared to those who had never used any method of contraception. This is however contrary to the findings in the literature. Even though traditional methods were not significant with a p value of 0.539, it indicates a decrease of 0.212 number of children ever born. This finding is similar to that of Bertrand et al., (1993) who found out in his survey of some African countries that women with secondary/higher levels of education are more likely to use traditional methods of contraception because they are better informed or they have a greater sensitivity to side effects of modern methods of contraception. Abortion status was not significantly associated with CEB in this study. University of Ghana http://ugspace.ug.edu.gh 57 Table 6. 2: Multiple Linear Regression on the effect of education and intermediate variables on children ever born. Source: Urban Health and Poverty Project Data, 2011. RC: Reference Category p<0.05 6.3 Other determinants of children ever born among women in urban poor communities in Accra. In model 3, the effects of some background characteristics on CEB among women in Accra were estimated. Table 6.3 shows the influence of education and control variables considered in the study. The model shows that 49.2 percent of the variation in the number of children ever born is attributed to the variables. Comparing model 3 to the previous model, there is an influence of certain variables on children ever born. Therefore, it can be inferred that certain background characteristics apart from education also influence the number of children ever born to a woman in her reproductive ages. Variable Model 2 Children Ever Born Adjusted R2 = 0.120 Beta(β) Std. Error Sig. 95% Confidence Level Intercept 2.709 0.311 0.000 2.097 3.321 Education No Education (RC) Primary -0.306 0.344 0.374 -0.983 0.370 Middle/JSS -1.061 0.327 0.001 -1.704 -0.419 Secondary & Higher -2.141 0.356 0.000 -2.839 -1.442 Contraceptive Use No Contraception (RC) Modern Methods 0.501 0.195 0.010 0.118 0.884 Traditional Methods -0.212 0.346 0.539 -0.891 0.466 Both Methods 0.531 0.329 0.107 -0.116 1.178 Abortion Yes 0.261 0.197 0.186 -0.126 0.648 No(RC) University of Ghana http://ugspace.ug.edu.gh 58 With education, there was a decrease in the average number of children ever born to women in all the educational categories (β=-0.148, β=-0.611, β=-1.042) compared to women with no education. Results for the marital status of the respondents suggest a strong association among women who were married, living with a partner and widowed with children ever born (p<0.05). There was an increase in the CEB to woman who were married, living with a partner and widowed (β= 1.521, β=1.054, β=1.885) compared to never married women. Married women are exposed to frequent sexual intercourse which is likely to result in a higher number of births. Similarly, the widowed may have completed their reproductive years as fertility increases by the age of the women. Despite being significant at the bivariate level of analysis, locality, and occupation were not significantly associated with the number of children ever born to women after controlling for other. In addition, household wealth, ethnicity and religion remained as not significant predictors of women’s average number of births. University of Ghana http://ugspace.ug.edu.gh 59 Table 6. 3: Linear regression of socio-demographic characteristics and CEB. Source: Urban Health and Poverty Project Data, 2011. RC: Reference Category p<0.05 Variable Model 3 Children Ever Born Adjusted R2 = 0.492 Beta(β) Std. Error Sig. 95% Confidence Level Intercept 1.044 0.446 0.020 0.169 1.920 Education No Education (RC) Primary -0.148 0.278 0.593 -0.694 0.397 Middle/JSS -0.611 0.277 0.028 -1.156 -0.066 Secondary & Higher -1.042 0.313 0.001 -1.657 -0.428 Wealth Status Poorest (RC) Poor -0.138 0.209 0.510 -0.550 0.273 Middle -0.313 0.212 0.141 -0.729 0.103 Rich -0.332 0.226 0.142 -0.776 0.112 Richest -0.273 0.225 0.226 -0.715 0.169 Locality Agbogboloshie (RC) James Town -0.203 0.228 0.373 -0.650 0.244 Usher Town 0.024 0.224 0.916 -0.416 0.463 Marital Status Not Married (RC) Married 1.521 0.220 0.000 1.088 1.954 Living with a partner 1.054 0.203 0.000 0.656 1.452 Widowed 1.885 0.451 0.000 1.000 2.771 Divorced 0.538 0.316 0.089 -0.082 1.158 Separated 0.601 0.274 0.029 0.063 1.139 Ethnicity Akan (RC) Ga-dangme 0.031 0.171 0.859 -0.306 0.367 Ewe -0.131 0.304 0.667 -0.728 0.467 Guan 0.549 0.526 0.297 -0.485 1.582 Mole-Dagbani -0.480 0.517 0.354 -1.496 0.536 Other Ethnic Group -0.503 0.323 0.120 -1.138 0.132 University of Ghana http://ugspace.ug.edu.gh 60 Table 6.3 continued. Source: Urban Health and Poverty Project Data, 2011. RC: Reference Category p<0.05 6.4 The overall effect of education and other factors on children ever born From the conceptual framework controlling for all the socio-economic variables, Education is associated with the fertility of the woman. Results for the fourth and final model are displayed in Table 6.4. The adjusted R squared value of 0.500 indicates that model 4 is a good fit. This means that 50% of the variation in fertility can be explained by all the variables used in the model. Variable Model 3 Children Ever Born Adjusted R2 = 0.492 Beta(β) Std. Error Sig. 95% Confidence Level Religion No Religion (RC) Catholic -0.154 0.414 0.710 -0.969 0.660 Protestant -0.209 0.318 0.511 -0.834 0.416 Pentecostal/Charismatic -0.251 0.298 0.400 -0.837 0.335 Other Christian -0.249 0.356 0.484 -0.947 0.450 Islam 0.046 0.387 0.906 -0.715 0.807 Other Religion 0.549 0.551 0.319 -0.533 1.631 Age 15-19 (RC) 20-24 0.269 0.258 0.298 -0.238 0.776 25-29 0.738 0.287 0.011 0.173 1.302 30-34 1.334 0.301 0.000 0.743 1.926 35-39 1.716 0.319 0.000 1.089 2.343 40-44 2.567 0.321 0.000 1.936 3.198 45-49 2.800 0.337 0.000 2.138 3.462 Occupation No Occupation (RC) Professional/Clerical -0.565 0.353 0.110 -1.258 -0.129 Sales/Services -1.155 0.197 0.431 -0.231 0.541 Skilled/Unskilled -0.256 0.230 0.262 -0.710 0.193 Other Occupation -0.957 0.659 0.147 2.253 0.338 University of Ghana http://ugspace.ug.edu.gh 61 In model 4, the results indicate that number of children ever born to a woman reduces with increasing level of education. Secondary and higher educated women had a strong significant (p<0.05) negative association with children ever born. They had a decrease of 1.017 in the average number of children ever born compared to those with no education. This indicates that women with higher education had lower mean children ever born (MCEB) levels. This is probably because of them spending more years in school which can increase their age at first birth or age at first marriage. This finding is consistent with Bledsoe et al., (1999) who found out that a woman with secondary or higher education stays longer in school, postpones the age at first childbirth and thereby lowers her fertility. The current age of the women shows a significant relationship with the mean children ever born (MCEB) except for those aged 20-24. Fertility increased with increasing age .From Table 6.4, it can be seen that women in the age categories 15-19 and 20-24 had fewer mean children compared to women in the older age categories. Using the age group 15-19 as the reference, Table 6.4 shows that older women had positive significant association with the mean children ever born. Women in age categories of 30-34, 35-39, 40-44, and 45-49 had a strong significant (p<0.05) association with higher mean children ever born (β=1.207, β=1.605, β=2.471, β=2.684) respectively compared to younger women. It was found that religion was not significantly related to fertility. Catholic women, Protestants, Pentecostals/Charismatics, and other Christians had a decline in the mean children ever born compared with those with no religion. Islamic women and those of other religious affiliations tend to have more children ever born compared to the various religious affiliations. University of Ghana http://ugspace.ug.edu.gh 62 At the bivariate level, locality was significantly associated with the dependent variable but when all the other variables were considered in the model it was not significant. Abortion experiences showed an increase of 0.123 units children ever born compared to women who had never encountered an induced abortion. Using no contraception as the reference category, there was an increase in 0.391 units of children ever born with women who had ever used modern contraceptives. Women who had ever used traditional methods of contraception, there was a decrease of 0.295 units of children ever born compared to women who had never used any method of contraception. Among women who had ever used both methods of contraception, there was an increase of 0.264 units of children ever born compared to women who had never used any method of contraception. Occupation was not significantly associated with the dependent variable. For the different types of occupations women engaged in, there was a decrease of β= -0.589, β= -0.345, β=-0.939 in children ever born among women who engaged in the professional/clerical, skilled/unskilled and other occupation. Among women who engaged in the sales/services type of occupation, the association was also not significant (p=0.466). Ethnicity and wealth status were also not significant with children ever born when other variables were considered in the model. With regard to their marital status, there was a significant relationship (p<0.05) between women who were married and their number of births compared to women who were not married. Strong associations existed with children ever born among those who were married, living with a partner and widowed. These women displayed an increase in the number of children ever born compared to their never married counterparts. There was no significant level among women in the category of divorced and separated. The University of Ghana http://ugspace.ug.edu.gh 63 significant relationship between women who were married compared to those who were not married can be explained by the fact that child bearing in the Ghanaian society is socially accepted in marriage thus women who have never married have very few or no children. Women who were married had an increase of 1.528 average children ever born compared to those who were not married. University of Ghana http://ugspace.ug.edu.gh 64 Table 6.4: Linear Regression of Children Ever Born, Education and other variables. Source: Urban Health and Poverty Project Data, 2011. RC: Reference Category p<0.05 Variable Model 4 Children Ever Born Adjusted R2 = 0.500 Beta(β) Std. Error Sig. 95% Confidence Level Intercept 1.041 0.443 0.019 0.171 1.912 Education No Education (RC) Primary -0.175 0.276 0.526 -0.718 0.367 Middle/JSS -0.639 0.276 0.021 -1.181 -0.097 Secondary & Higher -1.017 0.311 0.001 -1.629 -0.406 Age 15-19 (RC) 20-24 0.148 0.260 0.570 -0.363 0.659 25-29 0.599 0.292 0.041 0.025 1.173 30-34 1.207 0.304 0.000 0.610 1.803 35-39 1.605 0.320 0.000 0.977 2.234 40-44 2.471 0.323 0.000 1.835 3.106 45-49 2.684 0.339 0.000 2.019 3.349 Religion No Religion (RC) Catholic -0.136 0.414 0.742 -0.949 0.676 Protestants -0.196 0.317 0.537 -0.819 0.427 Pentecostal/Charismatic -0.223 0.298 0.455 -0.809 0.363 Other Christian -0.232 0.355 0.514 -0.929 0.465 Islam 0.062 0.385 0.873 -0.696 0.819 Other Religion 0.634 0.547 0.247 -0.441 1.710 Locality Agbogboloshie (RC) James Town -0.240 0.228 0.292 -0.687 0.207 Usher Town -0.014 0.223 0.950 -0.452 0.424 Abortion Yes 0.123 0.159 0.439 -0.189 0.434 No (RC) University of Ghana http://ugspace.ug.edu.gh 65 Table 6.4 continued. Source: Urban Health and Poverty Project Data, 2011. RC: Reference Category p<0.05 Variable Model 4 Children Ever Born Adjusted R2 = 0.500 Beta(β) Std. Error Sig. 95% Confidence Level Contraceptive Use No Contraceptive (RC) Modern Methods 0.391 0.154 0.011 0.089 0.692 Traditional Methods -0.295 0.270 0.274 -0.825 0.235 Both Methods 0.264 0.261 0.311 -0.248 0.777 Occupation No Occupation (RC) Professional/Clerical -0.589 0.351 0.094 -1.279 0.101 Sales/Services 0.143 0.196 0.466 -0.241 0.527 Skilled/Unskilled -0.345 0.231 0.135 -0.799 0.108 Other Occupation -0.939 0.656 0.153 -2.229 0.350 Wealth Status Poorest (RC) Poor -0.131 0.208 0.529 -0.540 0.278 Middle -0.300 0.210 0.155 -0.713 0.113 Rich -0.328 0.225 0.146 -0.770 0.114 Richest -0.265 0.224 0.237 -0.705 0.175 Marital Status Married 1.528 0.222 0.000 1.091 1.964 Living with a Partner 0.980 0.203 0.000 0.581 1.379 Not Married(RC) Widowed 1.881 0.452 0.000 0.993 2.769 Divorced 0.533 0.315 0.091 -0.086 1.152 Separated 0.578 0.272 0.034 0.044 1.112 Ethnicity Akan (RC) Ga-dangme 0.013 0.171 0.940 -0.323 0.348 Ewe -0.084 0.303 0.781 -0.679 0.511 Guan 0.463 0.523 0.376 -0.565 1.492 Mole-Dagbani -0.489 0.515 0.343 -1.500 0.522 Other Ethnic Groups -0.489 0.325 0.133 -1.128 0.149 University of Ghana http://ugspace.ug.edu.gh 66 CHAPTER SEVEN SUMMARY, CONCLUSIONS AND RECOMMENDATIONS 7.0 Introduction This chapter presents the summary of the study in which the educational attainment and fertility among women in urban poor communities in Accra, Ghana were assessed. Conclusions drawn from the major findings in the study are discussed in the chapter. In addition, recommendations are offered to various policy audiences. 7.1 Summary of the Study The study sought to determine the relationship between the level of women’s education and fertility in urban poor communities in Accra, Ghana. The urban poor communities in which the study was carried out are James Town, Ussher Town and Agbogbloshie. The data used for the study was drawn from the second round of the 2011 Urban Health and Poverty Project conducted by the Regional Institute for Population Studies (RIPS). The unit of analysis was women in their reproductive ages 15-49. A sample of five hundred and forty three (543) women in their reproductive ages was used for the study. The main objective of the study was to determine the relationship between the level of maternal education and fertility in urban poor communities in Accra, Ghana. To do this the following were carried out: i. Determining the level of education in the three urban poor communities in Accra, Ghana. ii. Examining the level of fertility in the three urban poor communities in Accra, Ghana. iii. Determining the relationship between other background characteristics of the women and fertility in these urban poor communities in Accra, Ghana. University of Ghana http://ugspace.ug.edu.gh 67 The sampling design for the study consisted of a two-fold sampling approach. First, 29 selected enumeration areas (EAs) in the three urban communities were sampled, and then 40 households were sampled from each of the EAs. Three methods of analysis, namely univariate, bivariate and multivariate were used for the study. Univariate and bivariate methods were employed to describe the background characteristics of the respondents and also examine the associations between the dependent and the independent and intermediate variables respectively. At the multivariate level, multiple linear regression analysis was used to ascertain the contribution of education and other socio-demographic variables to fertility. Children ever born was the dependent variable used to measure fertility. Findings shows that one fifth (21.5%) of the women formed the least proportion of mothers and they were in the older age group. The mean age of the women was 30.0 years. With regards to their highest level of educational attainment, more than half of the respondents (70.9%) had basic education (primary and middle/JSS) level of education and about one fifth (20.8%) of the women had secondary and higher level of education. A little above one half (51.6%) of the women lived in Ussher Town and the sales and services occupation represented the highest proportion (52.2%) of the women. Four- Fifth (82.1%) were Christians and slightly over half (53.6%) of the women belong to the Ga-Dangme ethnic group. With reference to marital status, a little above one third (34.3%) were not married. Almost half of the respondents (49.0%) had never used any method of contraception.71.1% of the respondents had never had any abortion experience. University of Ghana http://ugspace.ug.edu.gh 68 One third of the women had no child and the smallest proportion of mothers with 10 and above births represents 0.6% of respondents. The mean number of children ever born to women was about two. At the Bivariate level, it was observed that women with secondary and higher education had the lowest MCEB (0.81) compared to women with no education (2.96). The MCEB tends to decrease as the level of education of the mother increases. The study also revealed that the MCEB increases with age. The hypothesis of an inverse relationship between education and fertility was confirmed. Socio-economic variables that had a significant association (p<0.05) with the dependent variable are age of the respondent, contraceptives use, marital status, occupation and locality. However, ethnicity, religious affiliation, induced abortion and wealth status (p=0.966, p=0.364, p=0.542, p=0.157 respectively) were not significantly associated with fertility. Four models were run at the multivariate level. In the first model, the education variable was regressed on the children ever born (CEB), while the second model considered education, the intermediate variables and the CEB. For the third model, education and the control variables were regressed on CEB and the fourth model considered all the variables. Multivariate analysis showed that although education is important in influencing fertility behaviour, its effect is slightly reduced by other socio-economic factors. At the multivariate level, age and marital status were the most significant variables influencing fertility behaviour in urban poor communities while the intermediate variables (contraceptive use and abortion experiences) as well as the other variables (locality, religion, occupation, ethnicity and household wealth) were not significant. University of Ghana http://ugspace.ug.edu.gh 69 7.2 Conclusion It can therefore be concluded that female education especially secondary and higher education plays a very significant role in reducing fertility in these urban poor communities. Female education is likely to delay marriage, increase the age at first birth and thus reduce fertility in a population. Also educated women have more control over their reproductive rights and are in a better place to manage their households. This finding also confirms other studies carried out elsewhere in developing countries. 7.3 Recommendation(s) Based on the findings of the study, the following recommendations are presented: i. The Ministry of Education, non-governmental organizations (NGOs) and other stakeholders must make efforts to increase the levels of female educational attainment in order to reduce their risk of exposure to childbearing. This can be done by sustaining programmes that promote the enrolment of more females into school and encourage parents to send their female children to school. ii. Employment opportunities should be expanded to capture the already trained female labour force and hence encouraging enrolment. Employment here will have a double impact acting on fertility reduction and also acting as an encouragement for female enrolment in school. This recommendation was made largely based on findings from bivariate associations between certain types of occupation and fertility. iii. Strengthening family planning efforts in urban poor communities. If modern and traditional contraceptive use is encouraged early in their reproductive years it can result in a reduction in fertility levels. University of Ghana http://ugspace.ug.edu.gh 70 REFERENCES Adanu, R.M. and Tweneboah, E. (2004). Reasons, fears and emotions behind induced abortions in Accra, Ghana. Research Review, Vol. 20, No. 2, pp. 1-9. Addai, I. (1999).Does religion matter in contraceptive use among Ghanaian women? Review of Religious Research, Vol. 40, No. 3, pp. 259-277. Adhikari, R. (2010). Demographic, socio-economic, and cultural factors affecting fertility differentials in Nepal. Biomed Central, Vol.10, No.19. Agyei-Mensah, S. Casterline, J. B. and Agyeman, D. K. (2005). Reproductive change in Ghana: Recent patterns and future prospects. University of Ghana, Legon. Antwi, M. K. (1992). Education, society and development in Ghana. Unimax Publishers Limited. Balk, D. Montgomery, M. R. McGranahan, G. K. Mara, V. Todd, M. and Doreen, A. (2009). Mapping of urban settlements and the risk of climate change in Africa, Asia and South America, in J.M. Guzman et al. (Eds). Population Dynamics and Climate Change, UNFPA and IIED, New York and London. Becker, H. and Lewis, G. (1973). On the interaction between the quantity and quality of children, Journal of Political Economy, Part 2: New Economic Approaches to Fertility, Vol.81, No.2, pp. 279–288. Becker, Gary S. (1960). An economic analysis of fertility in the national bureau of economic research. (Eds).Demographic and Economic Change in Developed Countries, a conference of the Universities, National Bureau Committee for Economic Research, Princeton: Princeton University Press. pp. 209–240. Behrman, J. R. and Rosenzweig, M. R. (2002). Does increasing women’s schooling raise the schooling of the next generation? American Economic Review, Vol.92, No. 1, pp.323–334. Bertrand, J.Bauni, E.Lesthaeghe, R. Montgomery, M.Tambashe, O. and Wawer, M. (1993). Factors affecting contraceptive use in Sub-Saharan Africa. Washington, DC: National Research Council, National Academy Press. Betzig, L. (1986). Despotism and differential reproduction: A Darwinian View of History. Aldine: Hawthorne. Bigombe, B. and Khadiagala, G.M. (2003). Major trends affecting families in Sub- Saharan Africa. United Nations publications. Bledsoe, C.H. Casterline, J.B. Jennifer, A. Johnson-Kuhn, and Haaga, J.G. (1999). Critical perspectives on schooling and fertility in the developing world. (Eds.) Washington, DC: National Academy Press. Blossfeld, H. P. (1995). Changes in the process of family formation and women's growing economic independence: a comparison of nine countries. In: H. University of Ghana http://ugspace.ug.edu.gh 71 P. Blossfeld (Eds.). The new role of women. Family formation in modern societies, Boulder/Oxford, Westview Press, Vol. 3, pg-34. Bongaarts, J. and Potter, R. G. (1983). Fertility biology and behaviour, an analysis of proximate determinants. Academic Press, New York, pg-12. Bongaarts, J. and Sinding, S. W. (2009). A response to critics of family planning programs and international perspective. Sexual Reproductive Health, Vol. 35, pp. 39–44. Bongaarts, J. (1978). A framework for analysing the proximate determinants of fertility. Population and Development Review, Vol.4, No.1 pp.105-132. Bongaarts, J. (2002). The end of the fertility transition in the developing world. Population Development Review, Vol. 28, pp. 419-443. Bongaarts, J. (2003). Completing the fertility transition in the developing world: The role of educational differences and fertility preferences. Population Studies, Vol.57, pp.321-336. Bongaarts, J. (2005). The causes of stalling fertility transitions. Paper presented at the IUSSP International Population Conference, Tours, France. Bongaarts, J. (2006). The causes of stalling fertility transitions. Studies in Family Planning, Vol. 37, pg-16. Bongaarts, J. (2008). Fertility transitions in developing countries: Progress or Stagnation? Studies in Family Planning, Vol. 39, No. 2, pp. 105-110. Bongaarts, J. (2010). The Causes of Educational Differences in Fertility in Sub-Saharan Africa. Vienna Yearbook of Population Research, Vol.8, pp. 31-50. Brockerhoff, M. and Brennan, E. (1998). The Poverty of Cities in Developing Regions. Population and Development Review, Vol.24, No.1, pp. 75-114. Brunette, T.A. (1996). Ethnicity and fertility in West Africa. Demography, University of California, Berkely, Berkeley. Caldwell, P. Calwell, and Orubuloye, I. (1992). Fertility decline in Africa: A new type of Transition? Population and Development Review, Vol. 19, No. 2, pp. 225-262. Cleland, J. (2002). Education and future fertility trends, with special reference to mid- transitional countries. Completing the fertility transition.Pp.187-202. Cleland, J. and Chris, W. (1987). Demand theories of the fertility decline: An iconoclastic view. Population Studies, Vol. 41, pp.5-30. Coale, A. J. and Watkins, S.C. (1986). The Decline of fertility in Europe. Princeton, Princeton University Press. Cochrane, S. (1979). Fertility and Education. What do We Really Know? Baltimore: Johns Hopkins University Press. University of Ghana http://ugspace.ug.edu.gh 72 Cohen, B. (1998). The emerging fertility transition in Sub-Saharan Africa. World Development, Vol. 26, No.8, pp. 1431-1461. Cullen, D.L. and Luna, G. (1993). Women mentoring in academe: Addressing the gender gap in gender gap in higher education. Vol. 5, No. 2, pp. 123- 137. Cunningham, A. (2006). The broader societal benefits of higher education. Washington, DC: Solutions for our Future. Davis, K, and Blake, J. (1968). Social structure and fertility: An analytic framework. In: Nam, C.B. Population and society. New York: Houghton Mifflin Company. Davis, M. (2006). Planet of slums. New York: Verso. Day, L.H. (1983). Analysing population trends. New York: St Martin’s Press. DeRose, F. L. Dodoo, F. and Patil, V. (2002). Fertility desires and Perceptions of Power in Reproductive Conflict in Ghana. Gender and Society, Vol.16, pp.53- 73. Elo, I.T. (1990). Female education and the use of maternal and child health care services in Peru. Michigan, University Microfilms International, Ann Arbor. Ezeh, A. C. Mberu, B.U. and Emina, J.O. (2009). Stall in fertility decline in Eastern African countries: regional analysis of patterns, determinants and implications. Philosophical transactions of the royal society, Biological Sciences, Vol.364, pp. 2991-3007. Feyisetan, B, and Casterline, J. B. (2000). Fertility preferences and contraceptive change in developing countries. International Family Planning Perspectives, pp.100-109. Fieder, M. S. Huber, F. Bookstein, K. Iber, K. Schäfer, G. and Wallner, B. (2005). Status and reproduction in humans. New evidence for the validity of evolutionary explanations on basis of a university sample. Ethology, Vol. 111, pp. 940-950. Fitzgerald, B. (2011). Education Leads to Lower Fertility and increased Prosperity. The Earth Policy Institute Washington, DC. Fotso, J.C. Ezeh, A.C. and Essendi, H. (2009a). Maternal health in resource-poor urban settings: How does women’s autonomy influence the utilization of obstetric care services? Reproductive Health, Vol.6, pg-9. Fotso, J.C. Ezeh, A.C. Nyovani Madise, Abdhallah Ziraba, and Ogollah, R. (2009b). What does access to maternal care mean among the urban poor? Factors associated with use of appropriate maternal health services in the slum settlements of Nairobi, Kenya. Maternal and Child Health Journal, Vol.13, No.1, pp. 130–137. University of Ghana http://ugspace.ug.edu.gh 73 Fotso, J.C. (2007). Urban–rural differentials in child malnutrition: Trends and socioeconomic correlates in sub-Saharan Africa. Health and Place, Vol.13, No.1, pp. 205–223. Gaisie, S.K. (1981). Child spacing patterns and fertility differentials in Ghana. In Hilary Page and Ron Lesthaeghe (Eds.). Child spacing in Tropical Africa: traditions and change. Academic Press, pp. 237-253. Garenne, M. (2007). Situations of fertility stall in Sub-Saharan Africa. Paper presented at the 5th African Population Conference. UAPS, Arusha, Tanzania, pp.10–14. Ghana Statistical Service (GSS), Ghana Health Service (GHS) and ICF Macro (2009). Ghana Demographic and Health Survey 2008. Accra, Ghana: GSS, GHS and ICF Macro. Ghana Statistical Service. (1970). Ghana Population Census 1970, Accra- Ghana. Ghana Statistical Service. (2000). Ghana Population and Housing Census Report 2000, Accra-Ghana. Ghana Statistical Service. (2013). Population and Housing Census, National Analytical Report, Accra-Ghana. Grossman, M. (1972). On the concept of health capital and the demand for health. Journal of Political Economy, Vol.80, No.2, pp. 223–255. Haines, M. R. (1992). The European experience of declining fertility. The quiet revolution. In: Gillis, J. R. & Louise, A. (Eds). Tilly and David Levine. Cambridge, Massachusetts, Blackwell (Studies in Social Discontinuity):pp.193-226. Henry, R. and Fayorsey, C. (2002). Coping with Pregnancy: Experiences of adolescents in Ga-Mashie Accra. Calverton, Maryland, USA: ORC Macro. Indongo, N.and Pazvakawambwa, L. (2012).Determinants of fertility in Namibia. African Journal of Reproductive Health, Vol. 16, No. 4. Jeffery, Roger and Basu, M.K. (1996).Girls' schooling, Women's Autonomy and Fertility Change in South Asia. (Eds.).New Delhi: Sage. Jejeebhoy, S. J. (1995). Women’s Education, Autonomy and Reproduction behaviour: Experience from developing countries oxford, England. Kammeyer, K.C.W. (1971). An Introduction to Population. San Francisco: Chandler Publishing Company. Kulu, H. (2013). Why do fertility levels vary between urban and rural areas? Regional Studies, Vol. 47, No.6 pp. 895-912. Kollehlon, K. T. (2003). Ethnicity and fertility in Nigeria. Biodemography and Social Biology, Vol.50, pp. 3-4, 201-221. Leon, A. (2004). The Effect of Education on Fertility: Evidence from Compulsory Schooling Laws. Unpublished manuscript, University of Pittsburgh. University of Ghana http://ugspace.ug.edu.gh 74 Locoh, T. (2002). Fertility Decline and Family Changes in Sub-Saharan Africa. The Journal of African Policy Studies, Vol. 8, Nos. 2 and 3, pp.17-48. Lötter, J. M. (1988). Sociocultural factors and fertility. Southern African Journal of Demography, pp. 7-12. Lutz, W. and Goujon, A. (2001). The world's changing human capital stock: multi-state population projections by educational attainment. Population and Development Review, Vol.27, No.2, pp. 323-339. Mahama, A. S. Acheampong, A.T. Oti Boateng, P. and Baofo, Y. A. (2011).Preliminary Report for Ghana Mashie Urban Design Lab. Martin, P. and Zürcher, G. (2008). Managing Migration: The Global Challenge: Population Bulletin, Vol.63, No.1. Measurement, Learning and Evaluation Project for the Urban Reproductive Health Initiative in Kenya (2012). Mincer, J. (1963). Market Prices, Opportunity Costs, and Income Effects. In: Christ, C. (Eds.).Measurement in Economics: Studies in Mathematical Economics and Econometrics in Memory of Yehuda Grunfeld, Stanford: Stanford University Press. Montgomery, M.R. (2009). Urban poverty and health in developing countries. Population Bulletin, Vol. 64, No. 2. Moultrie, A. T. Hosegood, V. McGrath, N. Hill, C. Herbst, K. and Newell, M. L. (2008). Refining the criteria for stalled fertility declines: an application to rural KwaZulu-Natal, South Africa. Studies in family planning, Vol. 39, pp.39–48. Nasir, R. (2012). In search of reasons for the comparatively higher fertility among Muslim women in Delhi, India: Postulating a Cultural Hypothesis. Journal of Muslim Minority Affairs, Vol.32, No.4, pp. 557-575. Nukunya, G.K., (2003). Tradition and change in Ghana: An Introduction to Sociology, 2nd edition, Ghana Universities Press Publishing, Accra. University of Ghana http://ugspace.ug.edu.gh 75 Nyarko, P.E. (2005). Fertility levels, patterns and trends. Socio-economic and demographic trends, Vol.1, Ghana Statistical Service. Pelling, M. (2003. The vulnerability of cities: natural disasters and social resilience. London: Earths can Publications Limited. Population Reference Bureau (2012). Website: www.prb.org. Data Sheets, etc. Population Reference Bureau (2013). Website: www.prb.org. Data Sheets, etc. Razi, Z. V. I. (1980). Life, Marriage and Death in a Medieval Parish: Economy, Society and Demography in Halesowen, pp.1270-1400.Cambridge: Cambridge University Press. Robey, B.Rutstein, S.O. Morris, L. and Blackburn, R. (1992). The Reproductive Revolution: New Survey Findings. Population Report Series, No.11. Baltimore, John Hopkin University, Population Information Program. Rutstein, S.O. (2002). Fertility Levels, Trends, and Differentials 1995-1999.DHS Comparative Reports 3. Calverton, Maryland: ORC Macro. Schoumaker, B. (2004). Poverty and fertility in Sub-Saharan Africa: evidence from 25 countries. Paper presented at the 2004 annual meeting of the Population Association of America, Boston, and 1–3 April. Shapiro, D. and Tambashe, B.O. (2002). Fertility transition in urban and rural Sub- Saharan Africa: Preliminary Evidence of a three-stage process. Journal of African Policy Studies, Vol.8, Nos. 2-3, pp. 103-128. Shapiro, D. and Gebreselassie, T. (2008). Fertility Transition in Sub-Saharan Africa: Falling and stalling .African Population Studies, Vol. 23, No.1, pp. 3-23. Shapiro, D. and Gebreselassie, T. (2010). Falling and Stalling Fertility in Sub-Saharan Africa. Revised version of paper presented at the International Union for the Scientific Study of Population International Seminar on Human University of Ghana http://ugspace.ug.edu.gh 76 Fertility in Africa: Trends in the Last Decade and Prospects for Change, Cape Coast, Ghana, September 16-18, 2008. Shapiro, D. and Tambashe B. O. (1994). Education, Employment and Fertility in Kinshasa and prospects for change in reproduction behaviour. University Park Pennsylvania, Population Research Institute. Shapiro, D. Kreider, A. Varner, C. and Sinha, M. (2013). (Forthcoming). Stalling of fertility transitions and socio-economic change in the developing world: Evidence from the Demographic and Health Surveys. Skirbekk, V. and Samir, K.C. (2012). Fertility-reducing dynamics of women's social status and educational Attainment .Asian Population Studies, Vol. 8, No. 3, pp. 251-264. Tawiah, E. O. (1984). Determinants of cumulative fertility in Ghana. Demography, Vol.21, No.1, pp. 1-8. Tawiah, E.O. (1997). Factors affecting contraceptive use in Ghana. Journal of Biosocial Science, Vol.29, pp 141-149. Tesching, K. (2011). Dynamic Interrelations between Women’s Educational Level, Educational Field and Fertility in Sweden. In: Education and fertility. Print Centre US-AB. Trussel,J.Kost, K. (1987).Contraceptive failure in the United States: a critical review of the literature. Population Association of America. Annual Meeting Collected Papers. Vol. 9. UNDP (2011) Human Development Report 2011: Sustainability and Equity: A Better Future for All Palgrave MacMillan, New York. United Nations (1995). World Population Prospects: The 1994 Revision. New York. United Nations (2001). World Population Prospects: The 2000 Revision. Volume I: Comprehensive Tables. New York. United Nations Human Settlements Program (UN-Habitat), (2010). State of the world’s cities 2010/2011 – cities for all: bridging the urban divide. New York: United Nations. University of Ghana http://ugspace.ug.edu.gh 77 Van, D. P. Ellen, O. and Eddy, V. D. (2007). Are urban children really healthier? Evidence from 47 developing countries. Social Science and Medicine, Vol.65, No.10. Weeks, J. R. Getis, A.Hill, A. G. Agyei-Mensah, S. and Rain, D. (2010). Neighbourhoods and Fertility in Accra, Ghana: An AMOEBA-Based Approach. Weeks, J. (1999). Population: an introduction to issues and concepts. 7th Ed., California: Wadsworth Publishing. Weeks, J. (2008). Population: an introduction to issues and concepts. 10th Ed., California: Wadsworth Publishing. Weeks, J. (2012). Population: an introduction to issues and concepts. 12th Ed., California: Wadsworth Publishing. Westoff, C. F. and Cross, A. R. (2006). The stall in the fertility transition in Kenya. DHS Analytic Studies, No. 9, Calverton, NY: ORC Macro. Westoff, C.F. and Cross. A.R. (2005). The Stall in the Fertility Transition in Kenya. Paper presented at the IUSSP International Population Conference, Tours, France, July 2005. White, M.J. Tagoe, E. Stiff, C. Adazu, K. and Smith, D. (2002). Urbanization and the fertility in Ghana. Unpublished paper. Willis, R. J. (1973). A New Approach to the Economic Theory of Fertility. Journal of Political Economy, Part 2: New Economic Approaches to Fertility, Vol.81, No.2, pp. 14–64. http//:www.Ghanadistrict.com http//:www.wikipedia.com University of Ghana http://ugspace.ug.edu.gh