i REGIONAL INSTITUTE FOR POPULATION STUDIES UNIVERSITY OF GHANA AGE AT FIRST SEXUAL INTERCOURSE AND PARITY AMONG WOMEN IN URBAN POOR COMMUNITIES ACCRA GHANA BY MARIAN YAA ABOAGYEWA DONKOR SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF M.A. (POPULATION STUDIES) IN THE FACULTY OF SOCIAL SCIENCE, UNIVERSITY OF GHANA, LEGON JULY, 2014 University of Ghana http://ugspace.ug.edu.gh ii ACCEPTANCE Accepted by the Faculty of Social Science, University of Ghana, Legon in partial fulfilment of the requirements for the degree of M.A. (Population Studies) Supervisor of Thesis...................................................................................................... Dr. Naa Dodua Dodoo Date:...................................... University of Ghana http://ugspace.ug.edu.gh iii DECLARATION I hereby declare that, except for references to other people’s work which have been duly acknowledged, this work is the result of my own research and that it has not been presented elsewhere either in part or in whole for another degree. Signed.................................................... Marian Yaa Aboagyewa Donkor Date...................................................... University of Ghana http://ugspace.ug.edu.gh iv DEDICATION To my family, my nieces and nephews. University of Ghana http://ugspace.ug.edu.gh v ACKNOWLEDGEMENTS I am grateful to the almighty God for granting me the grace and strength to undertake this study. My sincere thanks go to my family and aunt for their moral support, encouragement and prayers throughout my study. My utmost appreciation goes to my supervisor, Dr. Naa Dodua Dodoo for her invaluable comments, constructive criticisms and suggestions which made this study complete. I am also thankful to Adriana Biney and Tobi Sanuade. University of Ghana http://ugspace.ug.edu.gh vi ABSTRACT The study of age at first sexual intercourse and parity examined the relationship between age at first sexual intercourse and parity among women in three selected urban poor communities in Accra, Ghana. The study investigated the average age at first sexual intercourse among women aged 15-49 years and their implications on fertility. The unit of analysis was individual women aged 15-49 years and the data set that was used for the analysis was the EDULINK 2011 wave two. Statistical measures such as percentages and means were used to describe the relationship between the background characteristics and the mean parities. Simple linear and multiple regression analyses were also used to analyze the effects of age at first sexual intercourse on parity and the effect of the background characteristics on parity. It was found out through the analyses that the average age at first sexual intercourse among women within the three selected urban poor communities was 17.96 years while the average parity was 2.17 births per woman. The bivariate and multivariate analyses, showed an inverse relationship between age at first sexual intercourse and parity, in that a lower age at first sexual intercourse increased the parity of a woman, hence confirming the hypothesis. Based on the findings, it was recommended that laws in Ghana governing age at first sex should be made stricter by raising the age. Extensive campaign on the knowledge and use of contraception should be embarked on by government and other stakeholders to educate women. Female education should be encouraged, also religious bodies should encourage young women to postpone first sex until they are ready. University of Ghana http://ugspace.ug.edu.gh vii TABLE OF CONTENTS CHAPTER ONE ........................................................................................................................................... 1 INTRODUCTION ......................................................................................................................................... 1 1.1 Background of the study ........................................................................................................... 1 1.2 Statement of the Problem .......................................................................................................... 3 1.3 Objectives of the Study ....................................................................................................................... 6 1.4 Rationale ............................................................................................................................................. 7 1.5 Research Question .............................................................................................................................. 9 1.6 Organization of Study ......................................................................................................................... 9 CHAPTER TWO ........................................................................................................................................ 10 LITERATURE REVIEW ............................................................................................................................... 11 2.0 Introduction ................................................................................................................................... 11 2.1 Determinants of early sexual initiation ......................................................................................... 12 2.2 Religion ......................................................................................................................................... 12 2.3 Media exposure ............................................................................................................................. 13 2.4 Knowledge and/use of Contraception ........................................................................................... 14 2.5 Parental supervision ...................................................................................................................... 14 2.6 Age at menarche ........................................................................................................................... 15 2.7 Effects of early age at sexual initiation ......................................................................................... 16 2.8 Unwanted pregnancy..................................................................................................................... 16 2.9 Sexually transmitted infections ..................................................................................................... 17 2.10 Maternal deaths ........................................................................................................................... 17 2.11 School dropouts .......................................................................................................................... 17 2.12 Early sexual initiation within urban poor communities .............................................................. 18 2.13 Determinants of fertility .............................................................................................................. 19 2.14 Age of woman ............................................................................................................................. 19 2.15 Religion ....................................................................................................................................... 20 2.16 Education .................................................................................................................................... 21 2.17 Contraceptive use ........................................................................................................................ 21 2.19 Place of residence/locality .......................................................................................................... 23 2.20 Migration ..................................................................................................................................... 23 University of Ghana http://ugspace.ug.edu.gh viii 2.21 Wealth Status .............................................................................................................................. 24 2.22 Marriage ...................................................................................................................................... 24 2.23 Pregnancy loss (Abortions, Still births, Miscarriages) ................................................................ 25 2.21 Conceptual framework ................................................................................................................ 27 2.22 Hypothesis ................................................................................................................................... 29 CHAPTER THREE .................................................................................................................................... 30 RESEARCH METHODOLOGY .................................................................................................................... 30 3.1 Introduction ................................................................................................................................... 30 3.2 Data source .................................................................................................................................... 30 3.3 Sampling design ............................................................................................................................ 31 3.4 Unit of analysis ............................................................................................................................. 31 3.5 Methods of analysis ...................................................................................................................... 31 3.5 Operationalisation, measurement and coding of variables............................................................ 32 3.7 Data limitation .............................................................................................................................. 35 CHAPTER FOUR ....................................................................................................................................... 37 PROFILE OF STUDY AREA, SOCIO-ECONOMIC AND DEMOGRAPHIC CHARACTERISTICS OF RESPONDENTS ........................................................................................................................................ 37 4.1Introduction .................................................................................................................................... 37 4.2 Background characteristics of the study population ..................................................................... 38 4.3 Parity (Number of Children Ever Born)........................................................................................ 40 4.4 Age at first sexual intercourse ....................................................................................................... 40 4.5 Average age at first sex among women within the various age groups ........................................ 41 4.6 Current age of respondent ............................................................................................................. 42 4.7 Education ...................................................................................................................................... 42 4.8 Marital Status……………………………………………………………………………42 4.9 Religious affiliation....................................................................................................................... 43 4.10 Ethnicity ...................................................................................................................................... 43 4.11 Locality of residence ................................................................................................................... 43 4.12 Migration ..................................................................................................................................... 44 4.13 Use of contraception ................................................................................................................... 44 4.14 Pregnancy loss ............................................................................................................................ 45 4.15 Wealth status ............................................................................................................................... 45 University of Ghana http://ugspace.ug.edu.gh ix CHAPTER FIVE ........................................................................................................................................ 47 AGE AT FIRST SEXUAL INTERCOURSE AND SOCIO-DEMOGRAPHIC DETERMINANTS OF PARITY ........... 47 5.0 Introduction ................................................................................................................................... 47 5.1 Parity levels ................................................................................................................................... 48 CHAPTER SIX ........................................................................................................................................... 55 RELATIONSHIP BETWEEN AGE AT FIRST SEXUAL INTERCOURSE AND DEMOGRAPHIC AND SOCIO- ECONOMIC FACTORS AND PARITY .......................................................................................................... 55 6.1 Introduction ................................................................................................................................... 55 6.2 Discussion ..................................................................................................................................... 59 CHAPTER SEVEN .................................................................................................................................... 62 SUMMARY, CONCLUSION AND RECOMMENDATIONS ........................................................................... 62 7.1 Summary ....................................................................................................................................... 62 7.2 Conclusion .................................................................................................................................... 64 7.3 Recommendation .......................................................................................................................... 65 University of Ghana http://ugspace.ug.edu.gh x LIST OF TABLES Table 4.1 Descriptive statistics of the dependent, independent and control variables (N=478) ... 39 Table 4.2 Descriptive statistics of parity of women……………………………………………..45 Table 5.1 An Anova table showing the relationship between background characteristics and parity ....................................................................................................................................................... 48 Table 5.2 An independent t-test result showing the relationship between migration status, pregnancy loss and use of contraception and parity ..................................................................... 52 Table 6.1 A simple linear regression analysis of the relationship between age at first sexual intercourse and parity .................................................................................................................... 55 Table 6.2 A multiple regression analysis table showing the influence of the independent variable and control variables on parity ...................................................................................................... 56 University of Ghana http://ugspace.ug.edu.gh xi LIST OF FIGURES Figure 1: A conceptual framework of age at first sexual intercourse and parity ........................................ 27 University of Ghana http://ugspace.ug.edu.gh 1 CHAPTER ONE INTRODUCTION 1.1 Background of the study The world’s total population as of 2013 stood at 7.169 billion according to the United States Census Bureau (USCB ,2013). Africa is the biggest contributor to population growth. According to the Population Reference Bureau, with a projected growth of 1.3 billion between 2013 and 2050, Africa will add more population than any other region in the world. Virtually, all of that growth will be in the 51 countries of Sub-Saharan Africa, the region’s poorest according to the Population Reference Bureau (PRB, 2013). The increase in population in various countries across the globe has partly led to the emergence of large cities. These cities are fast growing, according to the World Health Organisation,(WHO, 2011) by 2030, 6 out of every 10 people will live in a city and by 2050 this proportion will increase to 7 out of 10 people. Currently, around half of all urban dwellers live in cities with between 100,000 to 500,000 people. By 2050, urban populations in developing countries are expected to grow to about 5.2 billion (WHO, 2014). Ghana’s population alone has increased from 6,700,000 in 1960 to 24,658,823 in 2010. The two major cities in Ghana, Kumasi and Accra, recorded the highest population (GSS, 2010). Accra recorded a total population of 491,817 in 1960 to 2,905,726 in 2000 with a growth rate of 4.4 percent (GSS, 2005). Kumasi recorded a total population of 4,780,380, representing 19.4 percent of the total population whiles Accra recorded a total population of 4,010,054 representing 16.3 percent of the total population in 2010. Accra is the capital city of Ghana, and it is the centre of many important facilities for development such as schools, hospitals and tourists sites. If its population is not controlled, the rapid growth in University of Ghana http://ugspace.ug.edu.gh 2 population will lead to the emergence of many urban poor communities competing for limited urban resources. Accra’s total fertility rate of 2.9 from the 2003 Ghana Demographic and Health survey report was the lowest in the country. The high population growth rate in the region is thus a mixture of natural increase and rapid migration into the urban parts of the region from all over the country. Observation by the Ghana Statistical Service (2005) suggests that a fair number of the migrants are unskilled rural migrants moving into the city to look for non-existent jobs and ending up in the pool of urban poor. Dodoo et. al ( 2006), estimated that about 72 percent of urban dwellers in Africa lived in slums. An increasing growth of urban slums in Ghana, has led to increased urban poverty. Although such phenomenal growth is partly due to rural-urban migration, birth rates play an important role (Ezeh et. al, 2010). The fertility rates of a geographical location will determine its population over time. Fertility levels have remained high in most of Sub-Saharan Africa (PRB, 2013). Lower age at first sexual activity coupled with the low level of contraceptive usage has contributed to population growth in Sub-Saharan Africa for decades (Khan & Mishra, 2008), but this phenomenon has seen an increase in recent times (Speizer et. al, 2013). The earlier a woman initiates sexual activity the longer she stays in the reproductive years, hence increasing her fertility level. According to the Guttmacher Institute, (2014) the average age at first sexual activity for young people globally is 17 years, whiles in Sub-Saharan Africa it is 16 years. In Ghana, the average age of first sexual intercourse is between 15- 19 years (Guttmacher Institute, 2014). The Ghana Demographic and Health Survey report indicated that the contraceptive prevalence rate in Ghana was 24 percent for any method and 17 percent for modern methods in 2008 (GSS et. al, 2009). Cities may play important roles in national development, they can serve as central points for tourist attraction, and also serve as places for centres of higher learning and improved medical University of Ghana http://ugspace.ug.edu.gh 3 treatment. However the rate at which city populations are expanding does not match up with the current level of socio-economic development in Sub-Saharan Africa. There are issues of poor sanitation, poor housing, environmental pollution and inadequate health and educational facilities (Hardoy & Mitlin ,2014), all of which have implications for economic and sustainable development. Considering the fast rate at which cities are urbanizing in Africa, and the continent’s bleak economic future (Zulu et. al, 2002), it is important to understand the fertility behaviour of women in order to check population growth. Although several studies on factors that determine the fertility behaviour of women have been done in most parts of Africa and Sub-Saharan Africa, few studies have been done at the micro level, for example within urban poor communities (Speizer et. al, 2013). This study seeks to establish the relationship between age at first sexual intercourse and parity among women within the urban poor context 1.2 Statement of the Problem Early sexual activity among girls may lead to a prolonged exposure to childbearing. This prolonged exposure to childbearing may increase the parity levels of a woman. A high parity among women may lead to an increased risk to maternal mortality. In urban poor communities where health facilities may be very few, pregnant women may not have access to these facilities, this may lead to an increase in maternal mortality. The frequency of childbearing throughout the reproductive years may also expose a woman to maternal death. Each pregnancy is associated with the risk of death, and this risk rises with the age of the mother and the order of the pregnancy. In least developed countries, the life time risk of dying from pregnancy related causes is about 5 percent University of Ghana http://ugspace.ug.edu.gh 4 and many more women suffer related health problems or disabilities (Bongaarts, 2011).The maternal mortality rate of Ghana as of 2010 stood at 350 per 100,000 live births (Banchani & Tenkorang, 2014). As a development indicator, Ghana needs to reduce its current maternal mortality ratio by three quarters (75%) by the year 2015 in order to achieve the Millennium Development Goal Five (MDG, 2012). High Parity among women may lead to a youthful population structure that must be absolved into the labour force. About half the population of the least developed countries is under age 20 (UNFPA, 2012).In Ghana, the 2010 Population and Housing Census estimated that about 22.4 percent of the county’s population of 24,658,823 were young people of ages 10 to 19 years (Assampong et.al , 2013). In order to achieve the demographic dividend Ghana needs to make effective use of its youthful population structure by providing employment. However, unemployment is widespread because most economies are unable to provide jobs for the rapidly growing number of young people seeking to enter the labour force. The biggest weakness of the Ghanaian economy is the chronic joblessness of a vast proportion of people. Aryeetey (2011) was of the opinion that the extent of joblessness was evident in the huge number of youth who line the streets “selling things nobody will buy’. The magnitude of the problem of unemployment in Ghana cannot be underestimated considering the fact that according to Aryeetey (2011), it was estimated that about 50% of graduates who leave Ghanaian universities and polytechnics would not find jobs for two years after their national service. In urban poor communities where formal education may be low, due to either the limited number or absence of educational facilities, the problem of unemployment may be worst. Vigorous competition for limited number of jobs may lead to low wages which in turn contribute to poverty. The presence of large numbers of unemployed and University of Ghana http://ugspace.ug.edu.gh 5 frustrated males especially will contribute to socio-economic tensions, high crime rates and political instability. As populations expand further in urban areas, there is the problem of environmental degradation. The overcrowding in urban areas often leads to shortage of fresh water and arable land, water, air and soil pollution. Insanitary conditions such as poor drainage system, poorly ventilated houses and poor toilet facilities pose serious health hazards. In Sub-Saharan African cities where virtually no mechanisms are put in place by governments to ensure that the environment is protected, there are major health risks such as the spread of diarrhoeal diseases and respiratory tract infections (Hardoy & Mitlin ,2014).These environmental stresses build up as population increases. In addition, this increase in population may put pressure on the already limited and mostly deteriorated social amenities such as schools, hospitals and toilet facilities available in urban areas. The pressures in urban towns are likely to result in the development of illegal and informal settlements such as slums. The poor housing conditions in slums make residents vulnerable to acute health problems including malaria, tuberculosis and respiratory tract infections. High parities among women in urban poor areas where poverty is on the increase due to unemployment and low wages may lead to young mothers engaging in social vices such as prostitution in order to make enough money to cater for their children. Extreme deprivation associated with high unemployment and low wages of slums trap residents into engaging in risky sexual behaviour for economic survival. The problems are worsened by the inability of slum dwellers to access family planning services (Zulu et. al, 2002) In urban poor communities where because of poverty females engage in a lot of transactional sex (Ankomah et. al, 2011), young women in urban poor communities engage in multiple sexual University of Ghana http://ugspace.ug.edu.gh 6 partnerships most of which are concurrent or overlapping (Kabiru et. al, 2010). Such risky sexual behaviour may lead to an increase in the contraction of STI’s especially HIV/AIDS. In a region that accounted for almost 70 percent of all cases of HIV/AIDS in the world, approximately half of all new infections in Africa occurred among young people aged 15-24 ( Kabiru et. al, 2010). Sub- Saharan Africa recorded the largest HIV/AIDS related deaths in 2010 (WHO, 2011). A study conducted by Magadi ( 2013), confirmed that the poor in urban areas are more affected by HIV than the non-poor. Children born to women in slums suffer abuses such as sexual exploitation due to the fact that they may lack any form of social protection and social cohesion (Magadi, 2013). 1.3 Objectives of the Study The main objective of this study is to examine the relationship between age at first sexual intercourse and parity among women aged 15-49 in urban poor communities in Accra. The study will specifically:  Describe age at first sexual intercourse and the background characteristics of women in urban poor communities  Determine the average age at first sexual intercourse  Determine the influence of age at first sexual intercourse on parity among women in urban poor communities  Suggest evidence based interventions for effective policy planning. University of Ghana http://ugspace.ug.edu.gh 7 1.4 Rationale The increasing high fertility rates and the rapidly growing populations in Africa including Ghana is becoming a 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 populations keep increasing, issues of population pressure, carrying capacity and environmental degradation have emerged. Some of these concerns are consistent with the Malthusian theory which postulates that if population is not controlled, the number of resources will outstrip the population hence resulting in poverty (Weeks, 2008). The increasing rate at which the city of Accra is fast urbanising (GSS, 2010) ,and its attendant emergence of informal settlements (slums) has become a major cause for concern for governments and policy makers. There are a large number of people already living in slums in Ghana. According to Ghana’s development policy blue-print, the Growth and Poverty Reduction Strategy, 2006-2009 (GPRS II), in 2001, the number of people who lived in urban slums in Ghana was about five million (representing about 58 percent of the total urban population then), and grew at the rate of 1.8 percent per annum (Owusu & Afutu-kotey, 2010).Target 11 of goal 7 of the Millennium Development Goals seeks to significantly reduce and improve the lives of urban dwellers by the year 2020 (Owusu & Afutu-kotey, 2010) .The rapid urbanization of Accra with its attendant upsurge of informal dwellings, coupled with massive environmental degradation, young people’s early exposure to sexual intercourse and its implications for socio-economic development calls for a study that will examine the relationship between age at first sexual intercourse and parity (children ever born) among women within the urban poor context. University of Ghana http://ugspace.ug.edu.gh 8 In Africa several studies have been done to study the fertility behaviour of women .These studies are usually large-scale but not community based (Weeks & Allan , 2013). Few studies have been done to understand the fertility behaviour of women in urban poor communities and to establish the relationship between age at first sexual intercourse and parity among women. According to Lockwood (2014), it is important to study micro societies in order to fully understand the context within which fertility behaviours exist. There is a gap in literature on the relationship between age at first sexual activity and parity especially among women in urban poor communities. This study seeks to contribute to filling this gap in literature. The influences of urban poverty on fertility outcomes are still poorly understood (Ezeh et. al, 2010). This situation is partially attributable to the assumption that urban residence is associated with extensive exposure to ideation regarding fertility regulation, a better way of raising children, and better access to reproductive and other health services and information. Considering the deprived conditions in slums, this might not be the case for the poor in most urban areas in Sub- Saharan Africa and Ghana. Some studies have showed that the levels of key fertility indicators among the urban poor are significantly worse than those of their better-off counterparts (Ezeh et. al, 2010) . For a deeper understanding of the relationship between age at first sexual activity and parity among women in urban poor communities, there will be a need for a community based study in order to understand the relationship between living in an urban poor community and a woman’s parity or fertility behaviour. This will help suggest evidence based interventions for people already living in urban poor communities in Accra. University of Ghana http://ugspace.ug.edu.gh 9 Attention to the details of the processes through which urban residence affects population growth is relatively new (Brockerhoff, 1995). Most researchers have used data on age specific fertility rates and total fertility rates to analyse fertility patterns, less attention has been paid to how place of residence, as well as socio-economic status affect fertility (Kulu, 2013).This study, will add to the body of knowledge on how place of residence, and poverty impact on fertility. 1.5 Research Question This study seeks to answer the research question: what is the relationship between age at first sexual intercourse and parity among women in urban poor communities? 1.6 Organization of Study This dissertation is organized in seven chapters as follows: Chapter One consists of the background to the study, the problem statement, the research question, the general and specific objectives, the rationale, and the organization of the study. Chapter Two is made up of the literature review, conceptual framework and hypothesis. Chapter Three is the methodology, which comprises of source of data, methods of analyses and data limitations. Chapter Four consists of profile of study area and respondents background characteristics. Chapter Five is an analysis chapter that describes the relationship between age at first sexual intercourse and parity among women. Chapter Six is an analysis chapter where multivariate analysis on age at first sexual intercourse and parity among women is done. University of Ghana http://ugspace.ug.edu.gh 10 Chapter Seven is the final chapter that is made up of summary, conclusions and recommendations. University of Ghana http://ugspace.ug.edu.gh 11 CHAPTER TWO LITERATURE REVIEW 2.0 Introduction High fertility is still persistent in Africa, even though recent studies have indicated that there have been some declines (Bongaarts, 2011). Sub-Saharan Africa, the region’s poorest still records the highest total fertility rates. As of 2013, Niger had the highest total fertility rate of 7.6 births per woman (PRB, 2013). The Ghana Demographic and Health Survey report (2008) indicated that the total fertility rate of Ghana was 4.0 births per woman, which is far above replacement levels (GSS et. al, 2009). This high fertility has implications for socio-economic development in Ghana and Sub-Saharan Africa as a whole. Sub-Saharan Africa is experiencing rapid urbanization (Owusu & Afutu-kotey, 2010), and so is Ghana, the increase in urban population size that is not matching up with current levels of development such as the provision of decent housing, hospitals, schools, water and sanitation facilities may increase the number of urban poor communities in Accra . There are many factors that affect fertility, among these are age at onset of sexual activity and contraceptive use. Lower age at sexual initiation is an important social issue, studies have shown that unplanned, mistimed and unwanted pregnancies are common among young women who initiate sexual activity earlier (Collins et .al, 2004). An early age at first sexual intercourse may also predispose a woman to a long period of childbearing hence possibly leading to an increase in her parity. The age at first sexual intercourse among young people varies across population sub- groups. A longitudinal study on the sexual behaviour of young people in two communities, a slum and non-slum urban community in Nairobi Kenya, revealed that the average age at first sex for both males and females differed from that of slum dwellers to that of non-slum dwellers. University of Ghana http://ugspace.ug.edu.gh 12 For females, the average age at first sexual activity was 18 and 15 years for non-slum and slum dwellers respectively. The males on the other hand recorded 17 years for non-slum dwellers and 15 years for their slum counterparts (Kabiru et. al, 2010). Early initiation of sexual activity among women may lead to high parities due to a prolonged exposure to childbearing (Mberu, 2011). 2.1 Determinants of early sexual initiation Age at first sexual intercourse among women is determined by a number of factors, among which are religion, media exposure, knowledge/use of contraception, parental supervision and age at menarche. 2.2 Religion Religious values are the source of moral prescriptions for many individuals, the teachings of some religious organizations are likely to play a role in the formation of individual attitudes, values and decisions. The extent to which religion influences individual attitudes, and behaviour however depends on specific doctrines and policies of faiths and on the degree of integration and commitment of individuals to their particular religious institutions (Odimegwu, 2005). The Christian religious groups have strong opposition to premarital sex, although such opposition is more radical among the Pentecostal and Evangelical religious movements. While the latter can sanction their members by excommunication, the former can tolerate the offending members with the hope that they will turn a new leaf. This differential institutional commitment to premarital sexual abstinence leads to the expectation that individual Catholics and Evangelical Protestants would be less accepting of premarital sex than Non-Fundamentalist Protestants. Those with no religious affiliation would be most likely to accept and engage in premarital intercourse (Green, 2001). University of Ghana http://ugspace.ug.edu.gh 13 2.3 Media exposure A number of studies that have been done in population sub groups such as that of young people have shown that the viewing of sexually explicit material on television is associated with a range of sexual outcomes and behaviours, amongst which are sexual arousal, transactional sex and lower age at first sexual intercourse (Parkes et. al, 2013). Adolescents and emerging adults watch a great deal of television and research shows that they are exposed to a steady stream of messages about sex and sexuality, the vast majority of which do not contain important messages about safe and healthy sexual behaviour. According to Farrar (2012), young people learn information about sex from watching television and thought that their attitudes about sexual topics, including sexual health, can be influenced. Watching sex on television is linked to both initiation of and advancement in sexual behaviours. However not all materials that are displayed on television are harmful, some are beneficial. Some of the television shows stress delaying sexual debut as a healthy choice and also provide information on the correct use of contraceptives for young people who become sexually active, whereas others are abstinence-based, which advocate delaying sexual initiation until marriage (Erkut et. al, 2012). A study conducted within the Lagos metropolis of Nigeria between 2009 and 2010 (Collins et. al, 2004) found out that accurate portrayal of sexuality in the media could contribute immensely to improving public health in the metropolis. Television is not the only media through which young people explore sexuality, other media such as the internet is also accessed by young people. A study conducted in Dar es Salaam and Mtwara, two urban towns in Tanzania among adolescents aged 15-19 years indicated that young people accessed the internet mainly through mobile phones. Facebook is by far the most popular internet University of Ghana http://ugspace.ug.edu.gh 14 site. In that study, adolescents highlighted their interest in reproductive and sexual health messages and updates being delivered through humorous posts, links and clips, as well as by youth role models like music stars and actors that are entertaining and reflect up-to-date trends of modern youth culture (Pfeiffer et. al, 2014). 2.4 Knowledge and/use of Contraception Early sexual initiation has been associated with increased sexual risk behaviours including unprotected intercourse, multiple sexual partners, and unintended pregnancy among young people (Nield et. al, 2014). For most first timers, no contraception was used (Akintola et. al, 2012). In a study conducted by Henry and Fayorsey (2002) in an urban poor community in Accra it was found out that sexually active teenagers whose first sexual activity was below age 16 had considerable knowledge in contraception and were using them although they did not use any form of contraception during their first sexual activity. The most common contraceptive that was used was the male condom. 2.5 Parental supervision Parental supervision is necessary for the development of young people. The vulnerability and resilience of young people is to a large extent determined by parental ties, communication and the mere presence of a parent at home (Ngom et. al, 2003). For many parents living in urban areas, there is the belief that their adolescents are sexually active and so are their friends (Asampong et. al, 2013). Asampong and his colleagues (2013) in their study in four Sub-Saharan African countries including Ghana found out that young people who stated that they had a good relationship with their parents, initiated sexual activity at a higher age, the same applied to young people who lived with both parents .However, the impact of parental University of Ghana http://ugspace.ug.edu.gh 15 supervision on young people may differ based on which of the parents’ lived with the children at home. In a study conducted in an urban slum in Nairobi, (Ngom et. al, 2003) it was found out that father’s presence instead of mother’s presence was strongly associated with resilience among adolescents, this may have an impact on the age at first sexual activity of young people. Parents’ communications with their children play a vital role in shaping adolescent reproductive health behaviour. Parents who frequently communicated with their children about sexual behaviours recorded better reproductive health outcomes of their children (Mullen, 2001). Adolescents that are encouraged by their parents to open up about their sexual concerns are more likely to delay age at first sexual activity, compared to their counterparts who do not. Asampong and his colleagues (2013) writing on adolescents and parents perception on the best time for sex indicated that maternal opposition toward engaging in sex was associated with lower probability of engaging in sex by an adolescent. Adolescent girls in particular were advised by their mothers mostly to abstain from sex until they were much older (Henry & Fayorsey, 2002). 2.6 Age at menarche Age at menarche marks the onset of reproductive capability and is a commonly used milestone for pubertal development (Marino et. al, 2014). Early menarche provides an opportunity for education to aid young women to make decisions concerning sexual initiation and sexual partnering that are healthy for them (Nield et. al, 2014).Some studies suggest that age at menarche are affected by biological factors, one being that the age of a mother at menarche might be the same for her daughter, and if she initiated sexual activity earlier, the daughter is likely to do same (Mott et. al 1996). University of Ghana http://ugspace.ug.edu.gh 16 In Ghana, the transition from childhood to adulthood is a major stage in a person’s life. It is accompanied by special ceremonies that teach initiates about, cultural, social and religious values. For females, initiation takes place after they have had their first menstrual period (Amos, 2006). Dipo is an initiation rite done in Ghana among the Krobos, it is a rite of passage that is intended to postpone the onset of sexual intercourse among girls until marriage. However a study done in Kroboland on Dipo by Asampong and his colleagues (2013) suggested that a reduction in age of eligibility may increase the likelihood of sexual debut before the ceremony. Participants who received the rite as toddlers had a greater length of time between the dipo and adulthood to become sexually active post initiation than females who were initiated during their late teens or early twenties. 2.7 Effects of early age at sexual initiation There are several negative effects of early sexual initiation among young people, these include unwanted pregnancy, the contraction of STIs (Kaestle et. al, 2005) unsafe abortions (Henry & Fayorsey, 2002) leading to maternal deaths and school dropouts. 2.8 Unwanted pregnancy First sexual initiations are mostly done without the use of a condom or any other method of contraception (Henry & Fayorsey, 2002) . This may be due to young people’s lack of knowledge on contraception (Prata et. al, 2013). Because early sex is usually experienced by adolescents who may still be in school and therefore unprepared, this may lead to mistimed and unwanted pregnancies. A lower age at first intercourse may also lead to a prolonged exposure to childbearing that may increase the parity of a woman. University of Ghana http://ugspace.ug.edu.gh 17 2.9 Sexually transmitted infections Early sexual initiation among adolescent girls has been associated with increased risk of STI infection, this is due to the exposure of an immature cervix to likely infections when young girls have sexual intercourse with an infected male partner (Kaestle et. al, 2005).Studies have found that people who initiate sexual activity earlier are likely to have multiple partners in the future (Kabiru et. al, 2010).This increase in sexual partnership predisposes an individual to the risk of contracting an STI or HIV. 2.10 Maternal deaths Young women, especially adolescents may suffer adverse pregnancy outcomes such as haemorrhages and ectopic pregnancies due to the fact that they may not access prenatal facilities (Fraser et. al, 1995), because that they may feel shy to visit prenatal clinics due to the fact that they may not be in a marital union. Young women may resort to unsafe abortions which may put them at risk of maternal death, when their male partners refuse responsibility for the pregnancy (Schwandt et. al, 2013). 2.11 School dropouts Young women who are still in school but become pregnant because of lack of contraceptive use may drop out of school. This may be due to the fact that they may suffer stigmatisation due to the premarital pregnancies (Levandowski et al., 2012). Because these young women drop out of school early and may not continue their education ( Grant & Hallman, 2008), their employment may be limited to petty trading, dressmaking and hairdressing and their earnings may not be enough to cater for themselves whiles pregnant and the baby that will be born in future (Akuffo, 1987). University of Ghana http://ugspace.ug.edu.gh 18 2.12 Early sexual initiation within urban poor communities Young people who initiate sexual activity earlier usually come from families with low income (Blum et. al, 2000). Rapid urbanization characterized by the emergence of slums and other informal settlements is a common feature of developing countries. It has been argued that developing country cities are harbouring an increasing proportion of the poor and destitute (Owusu & Afutu-kotey, 2010). Sub-Saharan Africa's ongoing urbanization is occurring amid the worst economic circumstances of any world region (Dodoo et. al, 2006). Economic hardship is acknowledged to compound women's sexual vulnerability, and is associated with early onset of sexual activity, and multiple sexual partnerships. Since there is a growing vulnerability of an increasingly marginalized and an increasing slum population that merits attention, apparently because poverty and its attendant lack of access to basic family planning facilities have implications for urban residents (Dodoo et. al, 2006). A study conducted by Greif and his colleagues ( 2010), among slum dwellers and non slum dwellers in five cities, Accra (Ghana), Das es Salaam (Tanzania), Harare (Zimbabwe), Kampala (Uganda) and Kenya (Nairobi) indicated that slum dwellers exhibited riskier sexual behaviour partly due to poverty than non-slum dwellers which included early sexual initiation. Economic stresses associated with low wages and unemployment presumably forces many women to use sex to generate income for basic needs provoking early initiation of sexual activity and multiple sexual partners .This early initiation of sex and multiple partnerships may lead to unwanted pregnancies. Henry and Fayorsey (2002) in a study conducted in Ga Mashie, Ghana on how adolescents coped with pregnancy found out that the majority of the respondents became sexually active after dropping out of school, when their parents could no longer support them financially. Thereafter they took to engaging in petty trading to assist their parents financially The boys with whom they University of Ghana http://ugspace.ug.edu.gh 19 had sexual intercourse would have to provide ‘chop money’ contribution in order to remain in the relationship ,the longer they remained in the sexual relationship the higher their chance of getting pregnant and having unwanted births. 2.13 Determinants of fertility Fertility in Sub-Saharan Africa stood at 5.1births per woman in 2005, ( Bongaarts & Casterline, 2013).This was more than twice the replacement level, this high fertility together with the declining mortality had led to population growth of about 2.5 percent per year. According to the Ghana Demographic Health and Survey report 2008, the total fertility rate of Ghana stood at 4.0 births per woman, which is also above replacement levels (GSS et. al, 2009). Several factors are associated with fertility of a woman, amongst which are age, religion, education, contraceptive use, ethnicity, place of residence/locality, migration, wealth status, marriage and pregnancy loss. 2.14 Age of woman Tawiah, (1984) found that fertility differentials were more pronounced for older women in Ghana. Among older women, the differentials were larger for rural women than urban women. At first marriage was found to be inversely related to cumulative fertility. Urban women who married before age 15 years had 1.5 more children than women who married after 30 years of age. Many women in Sub-Saharan Africa are giving birth to children that they are not prepared to cater for. A study by Okumu and Chege , (1994) revealed that 20 to 30 percent of all deliveries in maternal institutions in Nairobi were born to teen mothers, in Pumwani maternity hospital, 28 percent of deliveries were to girls under the age of 16 years. University of Ghana http://ugspace.ug.edu.gh 20 2.15 Religion There are differences in the fertility behaviour of women across various religions. Among Christians, the fertility of Catholics is higher than that of Non-Catholics (Sander, 1992),this may be due to their non use of modern contraception, however, (Addai et. al, 2014), was of the view that the impact of religion on contraceptive use may be shaped by urban versus rural differentials in accessibility to health and family planning services and facilities . For example, religious groups that live in mostly rural areas may have more limited access to health and family planning facilities and services than their counterparts residing mainly in urban centres. Also groups that dominate urban and other modernized geographical areas may be more exposed to outside ideas and more receptive to family limitation ideas and practices. Therefore, rural-urban and regional disparity may significantly shape the tendency to use or not to use contraception among the religious groups. A religion whose doctrines are pronatalist will have members whose contraceptive use is lower and fertility higher than religious groups whose doctrines are not pronatalist and are liberal towards contraceptive use (Addai et. al, 2014).Contraceptive use differences between the various religious groups are due to religious norms about family size and norms regarding the proximate determinants of fertility. The Catholic Church for instance, forbids most forms of contraception and abortion, other Christian groups have moved towards a more relaxed view on the use of contraception but maintain a strong opposition to abortion (Simons, 2003) According to Kollehlon, (1994) ,Muslim women have higher fertility than Christian women, and also in societies where Muslim populations were high, fertility decline has been very slow, however some researchers are of a different view, for instance, Simons ( 2003), said that Islam is generally supportive of family limitation although opposition to some forms of contraception especially sterilization has been expressed by many Islamic leaders. University of Ghana http://ugspace.ug.edu.gh 21 2.16 Education Female education can reduce fertility and at the same time increase fertility. The reduction in fertility due to education mediate through factors such as women’s autonomy and they having control over their own lives (Basu, 2002).Education provides women with the tools to make informed choices about household matters and their fertility. Female education increases the opportunity cost of childbearing and rearing among educated women, it may also lower fertility through improvement in child health and reduced rates of child mortality as women need to have fewer births to yield the same desired family size. Also female schooling may affect fertility due to the fact that an educated female may have more knowledge and use of contraceptive methods (Osili & Long, 2008). However ,Bledsoe and her colleagues (Bledsoe et. al, 1999) were of the opinion that other factors such as individual choice and household and community resources may affect both schooling and fertility decision (Bledsoe et.al,1999). Bongaarts ( 2003), in his study on fertility transition in the developing world, found out that, women with no education have fertility above the replacement level of two, whiles those with primary education have fertility higher than those with secondary plus education. He concluded that, if developing countries seek to see a decline in their fertility level, then the female population must be educated. 2.17 Contraceptive use Among populations today, contraceptive practice is the intermediate fertility variable primarily responsible for the wide range in the levels of fertility within marriage. In some countries induced abortion is more important. In the traditional developing countries, the knowledge and practice of contraception is relatively low (Bongaarts, 1983). In Sub-Saharan Africa, it is estimated that many young people had their first sexual intercourse around the ages 15- 19 years (Blanc & Way, 2014), these first sexual experiences were usually done without any form of contraception. Hence young University of Ghana http://ugspace.ug.edu.gh 22 women who engaged in sexual intercourse early had pregnancies that were unwanted. None use of any method of contraception may expose many women in their reproductive years to the bearing of many children. 2.18 Ethnicity Ethnicity is an influential factor in Africa that affects all aspects of life. African societies are structured along ethnic lines from which individuals derive their identities regarding themselves and their roles in life. Fertility differs across various ethnic groups due to the fact that there are different norms and behaviour regarding fertility. A study conducted in Nigeria among Yoruba, Ibo and Hausa-Fulani women showed that the fertility of Hausa-Fulani women was lower than that of the Yoruba and Ibo ( Kollehlon, 2003). This is because Hausa-Fulani women married earlier at about age 12 or immediately after their menses. Once they were married, they were supposed to live within their husbands’ compound. When friends visited they were not supposed to escort them past their husbands’ compound. Few Hausa-Fulani women did any paid work outside the home. Unlike Hausa-Fulani women, Yoruba and Ibo women who worked outside the home, were more autonomous and suffered no seclusion except those who were Muslim. Bride wealth payment among certain ethnic groups also has an influence on fertility. According to Imoagene (1990) , bride wealth which is very expensive among the Ibo, may delay the age at first marriage and thus childbearing among the Ibo women. Another important cultural norm that exists within certain tribes that determine fertility is the celebration of women who have had their tenth child .Among the Ibo (Imoagene,1990),women who give birth to the tenth child is highly celebrated by her husband. Among the Akan of Ghana, University of Ghana http://ugspace.ug.edu.gh 23 high fertility is very much cherished. Denzer and Ebeling, (2002) wrote that due to the pronatalistic nature of the Akan, with every birth a special name is given to the child. For instance the first born of a woman is usually, called Abakan or literally speaking meaning the opener, whiles the tenth born is called Badu meaning, the tenth child. A special ram is given to the woman by her husband as a gift whiles her in laws present her with ten basins of foodstuffs, for making her husband a hero. The woman is smeared with white clay as a sign of victory. 2.19 Place of residence/locality Certain conditions that may be present in places of residence may influence fertility. Fertility levels are said to be higher in rural areas than in urban areas (Kulu, 2013). This is because it is perceived that people who live in urban areas have better knowledge on reproductive health services and access to family planning services than their rural counterparts (Ezeh et al., 2010). 2.20 Migration Migration may be disruptive to fertility. According to Goldstein and Goldstein, (1983), non migrants may have the highest fertility rates and migrant women may have the lowest fertility rates, particularly when duration of residence is considered. Migrants may adopt the fertility norms of their destination areas when they have stayed there for long. However, women of lower parity may be more likely to migrate. Women, who moved between two specified events, for example, marriage and first birth or births of specific parity, experienced a longer interval between the events. Women who moved both between the ages of 20 and 29 and between the ages of 30 and 39 exhibited lower fertility than those who moved only during one of these age periods. According to Brockerhoff (1995), new arrivals experienced much lower fertility in their first few years in cities than long term residents of similar age and parity. This results from the initial University of Ghana http://ugspace.ug.edu.gh 24 unmarried status of most migrants, high levels of spousal separation among new arrivals who are married, dramatic increases in use of modern methods of contraception after two years in cities, and continuation of traditionally long durations of postpartum abstinence. 2.21 Wealth Status Fertility intentions differ for the poor and the rich. Poorer women may desire to have few or no children, but the unavailability of family planning resources may not allow them to attain that desire. At other times they may not be able to afford modern forms of contraception even though they may be available (Shapiro, 2014). There are economic and investment theories that surround an increase or decrease in fertility. These models examine the costs and benefits of investing in children. For example, Caldwell’s Wealth Flows approach (Caldwell, 1982) suggested that in traditional agricultural societies, children provided their parents with significant wealth through labour, favouring high fertility, whereas in modern economies children consumed wealth, resulting in low fertility. Other researchers have argued that children are always costly but that children’s work can subsidize parental reproduction, leading to higher fertility (Sear& Coall, 2011). In modern economies, it is argued that if child quality is a function of the investments made in children, then parents should raise fewer, high-quality children (Kaplan, 1996). 2.22 Marriage Marriage is universally referred to as an institution with many concepts which has a lot of influences on culture and socio-economic life. Some of its components include the age at first marriage. The age at first marriage for a woman is an important determinant of the number of children she would have. Age patterns of fertility vary considerably among regions, different groups and within countries. Although some childbearing occurs before marriage, the age at University of Ghana http://ugspace.ug.edu.gh 25 marriage often represents the beginning of regular sexual activity. The age at which a woman first experiences sexual intercourse and is thus at risk of pregnancy and childbearing also have an important effect on fertility. The older her age, the lower her potential lifetime fertility ((Agyei- Mensah, 2006) and vice versa. Previously, childbearing was only acceptable within the context of marriage, but now, non marital childbearing is increasingly becoming acceptable (Smock, 2010).According to Shapiro, (2014), there is a rising age at marriage that has contributed to fertility decline in Africa, the delay is more pronounced in urban areas than it is in rural areas. This is because, the expensive housing and adverse economic conditions have led to delay in marriage leading to decline in fertility. Even within the marital union, there have been delays in the onset of childbearing due to economic hardships in urban areas (Romanuik, 2011). 2.23 Pregnancy loss (Abortions, Still births, Miscarriages) Pregnancy loss is also one of the most important determinants of fertility, pregnancies are lost through induced abortions, stillbirths and miscarriages. Worldwide a high proportion of women become pregnant unintentionally, in both developed and developing countries. In the United States and in some Eastern European countries for which data are available, about one-half to three-fifths of all pregnancies are unintended, and a large proportion of these are resolved through induced abortion. And in many developing countries, the proportion of recent births that are unintended exceeds 40%, even in regions where most couples still want large families, 10-20% of births are unplanned (Bankole et.al, 1998). Another form of pregnancy outcome is still birth. Research has shown that about 3 million third –trimester stillbirths occur worldwide with 98 percent arising in medium and low income countries (Goldenberg, McClure, & Bhutta, 2011). According to Frøen University of Ghana http://ugspace.ug.edu.gh 26 and Cacciatore ( 2011), many women who had stillbirths actually wanted the pregnancy but lost it, however there are some misconceptions about stillbirths such as the fact that evil women eat up their babies in the womb. Miscarriages, another form of pregnancy outcome accounted for about 9 percent of all pregnancy losses in Ghana in the last five years that preceded the Maternal Health Survey conducted in 2007 (GSS et. al, 2009). Bongaarts (1983) proposed a framework for the study of fertility behaviour of women. His framework explained the extent to which biological and behavioural factors affect fertility through the influence of social, economic and cultural factors. Bongaarts further explained that 96 percent of the variation in fertility is explained by 4 main variables, he continued to name them the Principal Proximate Determinants of Fertility. They are.  Age at first marriage  Use or non-use of contraception  Induced abortion  Lactational infecundability For instance, in pronatalist societies where women may get married at an early age women exposed to early sexual activity could have implications for fertility levels. Moreover in societies where knowledge of contraceptive use is almost non-existent or minimal, fertility levels can be high. Where lactation and post-partum abstinence are practiced extensively, it can have a negative effect on fertility. University of Ghana http://ugspace.ug.edu.gh 27 2.21 Conceptual framework In this study, the conceptual framework examines the influence that age at first sexual intercourse has on parity, which is the number of children ever born. In spite of the proposed relationship between age at first sexual intercourse and fertility, there are certain socio-economic variables that may influence fertility as well, these variables are based on literature. These are: age, education, religion, wealth status, locality, ethnicity, migration, contraceptive use, marital status, and pregnancy loss. Figure 1: A conceptual framework of age at first sexual intercourse and parity Source: Author’s construct 2014 Parity (Number of children ever born) Age at first sexual intercourse Age Education Religion Wealth status Locality Ethnicity Migration Contraceptive use Marital status Pregnancy loss University of Ghana http://ugspace.ug.edu.gh 28 An early age at sexual intercourse will lead to a prolonged exposure to childbearing by a woman. However other factors may also predict parity. Fertility increases with age, older women tend to have higher fertility than younger women. Women with higher education are more likely to have lower parity due to the fact that they may have better knowledge of contraception and may use them, they may also be able to make informed decisions regarding their fertility. Women who are affiliated to a particular religion may have higher or lower parities, women who may be affiliated to the Catholic Church may have higher parities due to their non-use of contraception. Women who may have never used any form of contraception may have higher parities than those who may have used. Wealthy women may have lower parities than their poorer counterparts because they may have better access and may afford long term contraception. Women of particular ethnic groups may have higher parity than those of those of other ethnic groups. Migrants who may have been in a stable union at their place of origin may have a lower or no parity at their place of destination than migrants who were not in any stable union. Migrants who stay longer at their place of destination may assimilate the cultural norms of their host communities including their fertility behaviour, they may therefore have a higher or lower fertility based on the fertility norms of the host community. Married women and women in stable unions may have higher parities than those who are not. Women who may have suffered a pregnancy loss through a miscarriage or stillbirth may have their next pregnancy quite earlier. However, women who lost previous pregnancies through induced abortions may delay their next birth. University of Ghana http://ugspace.ug.edu.gh 29 2.22 Hypothesis This study hypothesized that lower age at first sexual intercourse increases the parity of a woman. In conclusion, the high fertility that is predominant in Africa is affected by many factors. These include age at first sexual intercourse, religion, age of woman, contraceptive use, marital status, ethnicity, locality of residence, migration status, pregnancy loss, wealth status and education. University of Ghana http://ugspace.ug.edu.gh 30 CHAPTER THREE RESEARCH METHODOLOGY 3.1 Introduction This chapter provides information on the source of data for the study, unit of analysis, the methods of data analyses, operationalization, measurement and coding of variables and the data limitation. 3.2 Data Source The study is based on secondary data from the 2011 Edulink Urban Health and Poverty project data wave 2. This Edulink Urban Health and Poverty Project data was collected by the Regional Institute for Population Studies (RIPS) in collaboration with faculties from the Universities of Southampton – United Kingdom, Cape Coast - Ghana, Ibadan – Nigeria, and Fourah Bay – Sierra Leone to integrate real life fieldwork into the teaching and learning of population sciences among students and staff in participating African and United Kingdom Higher Education Institutions (HEIs). The Project was conducted in three urban poor communities in Accra namely James Town, Ussher Town and Agbogbloshie. The Edulink 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. University of Ghana http://ugspace.ug.edu.gh 31 3.3 Sampling design The sample was drawn from 29 enumeration areas (EA’s), each with 20 households systematically chosen to make up a total sample size of 580 households distributed over the three localities. The number of EA’s and therefore, households in each locality was proportionate to the population size of that locality. There were 5 EA’s from Agbogbloshie, 8 from James Town and 16 from Ussher Town. The aim of this sampling procedure was to arrive at a survey with 500 households interviewed. Estimates from the Ghana Statistical Service indicated that the non-response rate in the Greater Accra Region is about 15%, and so adding an extra 15% of households to the 500 aimed at, led to the figure of 580. In each household chosen, every female between the ages of 15 and 49 and every male between the ages of 15 and 59 was eligible to be interviewed. For the purpose of this study, women from ages 15 to 49 who have ever had sexual intercourse have been selected so as to determine the impact of age at first sexual intercourse on their parity or number of children ever born. 3.4 Unit of analysis A total number of 478 women who have ever had sex were sampled out of the 548women population in the three localities. 3.5 Methods of analyses This study adopts a quantitative method of approach in its data analyses, the Statistical Package for the Social Sciences Software (SPSS) was used for the data analysis. 3.5.1 Univariate analysis: Frequencies were used to describe the various socio-demographic variables namely, age composition, religion, educational background, wealth status, migration University of Ghana http://ugspace.ug.edu.gh 32 status, locality, use of contraception, marital status, pregnancy loss and ethnicity. Means were used to describe the age at first sexual intercourse and parity (number of children ever born). 3.5.2 Bivariate analysis: A correlation was used to determine the effect of age at first sexual intercourse on parity at the bivariate level this is because both variables are continuous. A one way test of Analysis of Variance (ANOVA) was used to determine the influence of current age of respondents, marital status, locality of residence, wealth status, ethnicity, education and religion on the parity of women. This method of analysis was used because each of the variables has more than two categories and the dependent variable is continuous. An independent t-test was used to determine the influence of pregnancy loss, migration and contraceptive use on parity, this method was used because each of these variables are made up of only two categories and the dependent variable is continuous. 3.5.3 Multivariate analysis: A simple linear regression was used to determine the relationship between age at first sexual intercourse and parity. After that, a multiple linear regression was used to determine the combined effect of age at first sexual intercourse, background characteristics and parity. Usually a model would have been run with the background characteristics and the proximate determinants of fertility but these proximate determinants of fertility were not included in the models because the variable of main concern was the age at first sexual intercourse among women. 3.5 Operationalisation, measurement and coding of variables This section deals with how the independent variable, the control variables and the dependent variables were operationalised, measured and coded. University of Ghana http://ugspace.ug.edu.gh 33  The dependent variable Parity, (number of children ever born) was measured using the questions ‘Have you ever had a live birth?’ If the respondent answered yes then she is further asked ‘How many of these children are alive?’, and ‘How many sons are dead,’ ‘how many daughters are dead’? The answers put together were coded as Parity, and treated as a continuous variable.  The independent variable that is age at first sexual intercourse was operationalised based on how old women between ages 15-49 were when they first had sexual intercourse. The respondents were asked directly that ‘How old were you when you first had sexual intercourse?’. This variable was treated as a continuous variable. The control variables are locality, ethnicity, religion, age of respondent, educational attainment, wealth status, contraceptive use, migration, pregnancy loss and marital status.  The women used for this study were from three localities, James Town, Ussher Town and Agbogbloshie.  To measure ethnicity, the question asked was ‘What is your ethnicity?’ For this question, the options were [1=Akan] [2=Ga-Dangme] [3=Ewe] [4=Guan] [5=Gruma] [6=Mole- Dagbani] [7=Grusi] [8=Mande] [96=other]. Gruma, Mole-Dagbani, Grusi and Mande were collapsed into one ethnic group called Northern ethnic groups due to their similarities. Therefore, Akan was coded 1, Ga-Dangme was coded 2, Ewe was coded 3 Guan, Gruma, Mole-Dagbani Mande and Grusi were put together as Northern ethnic groups and coded 4 whiles other was coded 5.  Religion was measured using the question ‘What is your religion?’ The responses were [1=No religion] [2=Catholic] [3=Protestant] [4=Pentecostal/Charismatic] [5=Other Christian] [6=Islam] [7=Traditional/Spiritualist] [8=Eastern religions] [96=Other]. No University of Ghana http://ugspace.ug.edu.gh 34 religion and traditional/spiritualist was coded Other religion and 1 due to their small number. Catholic was coded 2, Protestant/Pentecost/Other Christian was coded 3, and Islam was coded as 4.  The ages of the women from 15-49 were categorised into 15-19,20-24,25-29,30-34,35- 39,40-44,45-49.  Educational attainment was measured using the question ‘What is the highest level of education you have attained?’ The options given were [0=Pre-school] [1=Primary] [2=Middle/SHS] [3=Secondary/SHS] [4=Higher]. No education was then coded 0, Primary coded 1, Middle code 2, SHS was added to Higher and coded 3.  With wealth status, poorest was coded 1, poorer was code 2, middle was coded 3, richer was coded 4, whiles richest was coded 5.This measure was used to determine the wealth status of the respondents.  Contraceptive use was measured by using the question ‘Which methods have you /your partner ever used?’ as a proxy and the responses were [a=Pill] [b=IUD] [c=Injections] [d=Implants] [e=Diaphragm/Foam/Jelly] [f=Male condom] [g=Female condom] [h=Female sterilization] [i=Male sterilization] [j=Natural methods] [k=Withdrawal] [l=Lactational Amenorrhea (LAM)] [m=Other]. This proxy for measuring contraceptive use was used as a measure for all the known methods of contraception that a woman has ever used and a variable was created called ‘use of contraception’. Within this variable, Non use was coded 1, whiles Ever use was coded 2.  There were many questions asked on migration in the data, but for the purpose of this study all respondents were considered migrants if they were born outside the Greater Accra region. All others who were born in the Greater Accra region were considered non- University of Ghana http://ugspace.ug.edu.gh 35 migrants. The question asked was ‘Where were you born?’ Migration status was coded as migrants=1 and non migrants =2.  In this study, pregnancy loss was made up of miscarriages, abortions and still births, therefore it was measured using the question ‘Have you / your partner had a pregnancy that miscarried, was aborted, or ended in a still birth?’ and the responses were [1=Yes] [2=No].  For the variable marital status, it was coded into [1= Ever married] which consisted of divorced, widowed and separated, [2=Single] and [3= Currently married]. Two questions were combined in the data to derive the variable marital status. They were: ‘Are you currently married or living with a partner?’ and the responses were [1=Yes currently living married] [2=Yes living with a partner] [3=Not in union] all those who answered Not in union were coded single and ‘What is your marital status now?’ the options were [1=widowed] [2=Divorced] [3=Separated]. 3.7 Data limitation Age misreporting could be a major limitation to this study. This could be due to the fact that, some respondents may have misreported their ages. This is because they may have felt embarrassed by the fact that they may have initiated sexual intercourse earlier or they may have forgotten the age that they had their first sexual intercourse. Older women who may have memory lapses may have forgotten the time they had their first intercourse or the number of children they may have ever had. Other women may have misreported their ages due to religious doctrines that prohibit having sexual intercourse out of marriage. Age at menarche and parental supervision which could also determine a woman’s age at first sexual intercourse and her subsequent parities are not measured in the data. Therefore statistical University of Ghana http://ugspace.ug.edu.gh 36 tests of association cannot be carried out to determine their effects on the parities of women in these three selected urban poor communities in Accra. Media exposure which could also have an influence on a woman’s age at first sexual intercourse was not measured in the data. Therefore they were not included in the models to run statistical tests to determine its influence on age at first sex and parity on the women in the three selected urban poor communities. University of Ghana http://ugspace.ug.edu.gh 37 CHAPTER FOUR PROFILE OF STUDY AREA, SOCIO-ECONOMIC AND DEMOGRAPHIC CHARACTERISTICS OF RESPONDENTS 4.1Introduction Historically at the heart of Accra and Ghana, 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, constituting Ussher Town and James Town, 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 (GAMADA, 2008). Increasing urbanization and rapid population growth in urban Accra have created considerable pressure on the community, which in the past was a dynamic, well-planned and very lively area, due in large measure to the colonial presence in the community. Ironically, this part of Accra has not seen any form of modern development since independence, although it borders the main Central Business District. It is now one of the most deprived and neglected urban areas in Accra with housing and other infrastructure in deteriorating state. Ga Mashie is described as an urban poor community (Quartey-Papafio, 2006). Agbogbloshie is considered a slum, this is because it is a heavily populated area with informal structures with low standard housing. Slums lack reliable sanitation services, supply of clean water and unreliable supply of electricity and timely law enforcement .The dwellings in Agbogbloshie are wooden shacks that lack water and sanitation. University of Ghana http://ugspace.ug.edu.gh 38 4.2 Background characteristics of the study population Background characteristics of populations are important factors in determining the socio-economic behaviour of that population. This chapter looks at the background characteristics of the respondents that is educational level, religious affiliation, ethnicity, locality of residence, wealth status, contraceptive use, pregnancy loss, migration status and marital status and age composition. This chapter also briefly describes the mean age at first sexual intercourse and the mean parity in the study sample. Understanding the background characteristics of the study population is necessary for the analysis and interpretation of data on age at first sexual intercourse and parity. University of Ghana http://ugspace.ug.edu.gh 39 Table 4.0.1 Descriptive statistics of the dependent, independent and control variables (N=478) Variables Frequency Percent Dependent variable Mean (SD) Minimum, Maximum Parity 2.17 (2.13) 0, 15 Independent variable Age at first sexual intercourse 17.96, (3.51) Mean age at first sexual intercourse 9, 43 Control variables Current age of respondents 15-19 29 15.45 6.1 20-24 96 17.23 20.1 25-29 88 18.14 18.4 30-34 84 18.30 17.6 35-39 66 18.89 13.8 40-44 56 18.13 11.7 45-49 59 18.42 12.3 Education No education 40 8.3 Primary 129 27.0 Middle 215 45.0 Higher 94 19.7 Marital status Ever married 106 22.2 Single 124 25.9 Currently married 248 51.9 Religious Affiliation Catholic 26 5.4 Protestant/Pentecostal/Other Christian 371 77.6 Islam 45 9.4 Other religion 36 7.6 Ethnicity Akan 135 28.2 Ga-Dangme 259 54.2 Ewe 28 5.9 Northern tribes 24 5.0 Other 32 6.7 Locality Agbogbloshie 90 18.8 University of Ghana http://ugspace.ug.edu.gh 40 Jamestown 140 29.3 Ussher Town 248 51.9 Migration status Migrants 285 59.6 Non-migrants 193 40.4 Use of contraception Non-use 208 43.5 Ever use 270 56.5 Pregnancy loss Yes 177 37.0 No 301 63.0 Wealth status Poorest 109 22.8 Poorer 96 20.1 Middle 95 19.9 Richer 82 17.1 Richest 96 20.1 Source: Computed from the EDULINK 2011, Data file 4.3 Parity (Number of Children Ever Born) Table 4.1 above shows the number of children ever born by women aged 15-49.The table also indicates that the mean parity per woman is 2.17, the minimum parity is 0, while the maximum parity is 15. The standard deviation for the distribution of parity is 2.13. 4.4 Age at first sexual intercourse Age at first sexual intercourse is an important determinant of risk of exposure to childbearing. Table 4.1 indicates that the average age at first sexual intercourse of women from ages 15-49 is 17.96 years. The minimum age at which women in the study sample had first sex is 9 years, whiles the maximum age at first sexual intercourse is 43 years, and the standard deviation is 3.51. University of Ghana http://ugspace.ug.edu.gh 41 4.5 The average age at first sexual intercourse among women within the various age groups Table 4.1 indicates that the average age at first sexual intercourse among women from ages 15-19 is 15.45 years. Whereas the average age at first sexual intercourse among women aged 20-24 years is 17.23 years. Women between the ages 25-29 years recorded an average age at first sexual intercourse at 18.14 years, whiles those within the ages of 30-34 years recorded an average at first sexual intercourse at 18.30 years. Women between ages 35-39 years and those ages 40-44 years recorded an average age at first sexual intercourse of 18.89 and 18.13 respectively. The last category of women who are between the ages of 45-49 recorded an average age at first sexual intercourse at 18.42 years. 4.6 Current age of respondents Age is a very important demographic variable. Age data may be tabulated in single years of age, five year age group or broader age groups. Age is the estimated interval of time between the date of birth and the date of the survey or census expressed in completed years. With reference to Table 4.1, the highest percentage of women in their reproductive age is between the age group 20-24. They recorded the highest percentage of 20.1 percent, while the lowest percentage is 6.1 representing 29 women within the age group 15-19. University of Ghana http://ugspace.ug.edu.gh 42 4.7 Education In recent years, fertility decline has been observed to follow expansion of education in society. Increasingly, education, particularly of women has been singled out as a major determinant of fertility decline. Rather than accepting traditional norms regarding fertility, education has assisted women in making informed choices. Table 4.1 above shows the percentage distribution of the educational level of the respondents. The table indicates that most women have attained middle school education with a percentage of 45.0. A relatively low percentage of women have attained higher education, with a percentage of about 19.7. The percentage of women who do not have any education is 8.3, a relatively small number, while 27.0 percent of the women have primary education. 4.8 Marital status Marriage affects fertility, in some societies, marriage marks the onset of sexual intercourse and hence the beginning of childbearing, in such societies therefore, the earlier a woman’s exposure to marriage, the longer her exposure to childbearing and the longer she remains in a marital union the longer her exposure to childbearing. Table 4.1 above shows the percentage distribution of women in the study sample who are currently married, ever married and single. Those who are ever married is made up of those who are widowed, divorced and separated. The percentage of women who are currently married is 51.9, whiles the percentage of women who are single is 25.9. The smallest percentage is for the women who have ever married, they recorded 22.2 percent of the study sample. University of Ghana http://ugspace.ug.edu.gh 43 4.9 Religious affiliation One of the variables that indirectly affect the fertility of women in their reproductive age is religion. A woman’s religious beliefs and norms may influence her behaviour towards fertility. From Table 4.1, 5.4 percent of the women in the study sample are Catholics whiles 77.6 percent belong to the Protestant, Pentecostal and other Christian faiths. The percentage of Moslims in the study sample is 9.4. About 7.6 percent of the women in the study sample belong to other religions. 4.10 Ethnicity Ethnicity is one of the demographic variables that influence first sexual intercourse and fertility. Early marriage, child betrothal and rites of passage such as Dipo, which is practiced among the Krobos of Ghana, are some of the cultural practices that have an influence on fertility. For this study, the major ethnic groups in Ghana are used, they are Akan, Ga-Dangme, Ewe, Northern ethnic groups, and other ethnic groups. Table 4.1 above shows that 54.2 percent which is highest percentage of women belong to the Ga-Dangme ethnic group, whiles 28.2 percent are Akan. The lowest percentage of women in the study sample belongs to the Northern ethnic groups. Whiles the women who belong to other ethnic groups are 6.7 percent. 4.11 Locality of residence The society within which young women grow up may have an influence on their development, relationships and adjustments. The expectations of society may mould their personalities, influence their roles, and guide their future. Table 4.1 shows the percentage distribution of women by their locality or place of residence. The table indicates that 51.9 percent of the total women in their reproductive age reside in Ussher Town, followed by James Town with a percentage of 29.3, University of Ghana http://ugspace.ug.edu.gh 44 representing 140 women of the total sample size of 478. The locality with the least women is Agbogbloshie representing 18.8 percent of the total sample of women. 4.12 Migration Migration as a component of fertility change can affect fertility. Migrants’ fertility may be disrupted once they move to a new place. They may also adopt the fertility norms of their destination communities when they stay longer, this may affect their fertility behavior. Table 4.1 indicates that 59.6 percent of women in the study sample are migrants whiles 40.4 percent are non -migrants. 4.13 Use of contraception Contraception can be defined as measures that are taken to prevent sexual intercourse from resulting in pregnancy. There are two categories of contraception that is modern and traditional methods. All these methods are used to deliberately postpone childbirth till such a time that one is ready for childbearing. Contraceptive use and frequency of sexual intercourse are the primary factors that may determine whether the sexually active become pregnant. The use of known methods, the availability of contraceptive device as well as norms and beliefs regarding contraceptive use affect fertility. Table 4.1 above shows that 56.5 percent of the women in the study sample have ever used a known method of contraception whiles 43.5 percent of the study sample responded that they have never used any form of contraception. University of Ghana http://ugspace.ug.edu.gh 45 4.14 Pregnancy loss From Table 4.1, 37.0 percent of the women in the study sample responded that they have ever lost a child through miscarriages, abortions or stillbirths, whereas 63.0 percent of the women responded that they have never had a pregnancy loss. 4.15 Wealth status From Table 4.1 above, 22.8 percent of the women in the study sample are very poor, whiles 20.1 percent of them are poorer while 19.9 percent of the women are middle class people, 20.1 percent of the women in the study sample belong to the richest quintile whiles 17.1 percent of the women are richer. Table 4.0.2 Descriptive Statistics of Parity of Women Parity(children everborn) Number of women Percentage 0 120 25.1 1 101 21.1 2 81 16.9 3 74 15.5 4 37 7.7 5 32 6.7 6 13 2.7 7 7 1.5 8 10 2.1 10+ 3 0.6 Total 478 100.0 Source: Computed from the EDULINK 2011, Data file University of Ghana http://ugspace.ug.edu.gh 46 Table 4.2 indicates that about 25.1 percent of women in the study sample representing 120 women had 0 parity,whiles about 0.6 percent of the women in the sample had 10 plus parities. In a summary this chapter presented the means, standard deviations, the minimum and maximum values of the dependent variable which is parity and the independent variable which is age at first sexual intercourse. The chapter also presented the percentage distribution of the respondents by current age, marital status, religious affiliation, ethnicity, locality of residence, migration, use of contraception, pregnancy loss and wealth status as well as the parity levels of women in the study sample. University of Ghana http://ugspace.ug.edu.gh 47 CHAPTER FIVE AGE AT FIRST SEXUAL INTERCOURSE AND SOCIO-DEMOGRAPHIC DETERMINANTS OF PARITY 5.0 Introduction Age at first sexual intercourse marks the beginning of the reproductive life of a woman. The age at first intercourse represents a critical indicator for the risk of pregnancy which has the implication of increasing her parity. Women who indulge in relatively early sexual intercourse are more likely to have larger families. This study examined the influence of age at first sexual intercourse on fertility so as to provide a better understanding of the relationship between the two variables. There are other factors that may affect parity apart from the age at first sexual intercourse, these are current age, marital status, education, religion, migration, locality of residence, contraceptive use, pregnancy loss, wealth status and ethnicity. This chapter examines the effect that the various socio- economic and demographic factors had on parity that is the extent to which these variables affected the mean number of children ever born by a woman. A correlation was used to determine the influence of age at first sexual intercourse on parity whiles a one way Analysis of Variance (ANOVA) was used to determine the influence of current age, marital status, education, religion, locality of residence, wealth status and ethnicity on parity. An independent t-test was used to determine the effect of migration, pregnancy loss and use of contraception on parity. Tests of significance to determine associations were conducted at 95 percent significance level. University of Ghana http://ugspace.ug.edu.gh 48 5.1 Parity Levels 5.1.1 The relationship between Age at first sexual intercourse and Parity A correlation was used to determine the relationship between age at first sexual intercourse and parity and indicated that there was a significant relationship between age at first sexual intercourse and parity (p=0.006). The Pearson correlation value of -0.126 indicated that there was a negative correlation between age at first sexual intercourse and parity, in that the lower a woman’s age at first sexual intercourse the higher her parity, the magnitude of this relationship was about 12.6 percent. Table 5.1 One-way Anova table showing the relationship between background characteristics and parity Characteristic Mean Parity No of Women F value Significance value Current age of respondent 15-19 0.37 29 35.51 0.000 20-24 0.68 96 25-29 1.55 88 30-34 2.32 84 35-39 2.93 66 40-44 3.78 56 45-49 3.77 59 Marital Status Ever married 2.64 106 57.70 0.000 Single 0.58 124 Currently married 2.76 248 Locality of residence Agbogbloshie 2.35 90 1.52 0.219 James Town 1.91 140 Ussher Town 2.25 248 Wealth Status Poorest 2.52 109 1.44 0.217 Poorer 2.18 96 Middle 1.94 95 Richer 2.26 82 Richest 1.80 96 Ethnicity Akan 2.14 135 University of Ghana http://ugspace.ug.edu.gh 49 Ga-Dangme 2.13 259 0.69 0.599 Ewe 2.10 28 Northern ethnic groups 2.87 24 Other 2.09 32 Education No education 3.02 40 17.66 0.000 Primary 2.43 129 Middle 2.15 215 Higher 0.96 94 Religion No religion 2.37 29 2.60 0.036 Catholic 2.34 26 Protestant/Pentecostal/Other Christian 2.07 371 Islam 2.3 45 Traditional/Spiritualist 4.5 7 SOURCE: Computed from EDULINK 2011, Data file. 5.1.2 Relationship between current age and parity Table 5.1 above shows the mean parity of women 15-49 years. The mean parity of women between ages 15-19 is 0.4 births per woman while that of women aged 45-49 is 3.8 births per woman. This indicates that parity increases with age. The women in this age group may have such high number of children ever born because they may have achieved their desired fertility. The variations in fertility are due to the age at which the woman is exposed to childbearing. This indicates that when a woman is exposed to childbearing relatively early in life, there is the higher probability of her having a large family size. 5.1.3 Relationship between marital status and parity Marriage marks the beginning of a woman’s exposure to regular sexual intercourse which may lead to childbearing. Most births occur to women who are currently married as evident in the table. From Table 5.1 above, the average parity of women who are currently married in the study sample is 2.8 births per woman, which is the highest. This is followed by women who have ever married, University of Ghana http://ugspace.ug.edu.gh 50 this include, the separated, widowed and the divorced, who have a relatively higher fertility of 2.6. This could mean that the women who are widowed, divorced or separated may be older women who may have completed their desired family size. The single women accounted for the lowest mean parity of 0.58. This could be due to the fact that because they are not in a stable union, they may not want to have children. 5.1.4 Relationship between locality of residence and parity The environment can influence a woman’s parity, some conditions that may be present at a woman’s locality of residence may affect a woman’s fertility levels. A study conducted by Grief et.al (2011) indicated that early sexual initiation among women in slum dwellings differ from non- slum dwellers due to poverty. According to Zulu et.al (2002), early age at first sexual intercourse is common among slum dwellers due to economic hardships that they face. In slums where employment opportunities are low leading to low wages, women engage in a lot of transactional sex which may result in unwanted pregnancies. Also within slum dwellings, certain conditions apart from poverty may allow for early sexual initiation that may lead to increased parities, these are residential arrangements that do not afford privacy for sexual intercourse within households and a social context that is acceptable of prostitution. All these conditions may have implications for fertility. From Table 5.1, even though place of residence is not significantly associated with parity, there are variations in the mean parities of the women who reside in the various localities. The table indicates that women in Agbogbloshie, which is a slum area have the highest mean parity of 2.3 whiles the two urban poor communities, James Town and Ussher Town recorded lower parities of 1.9 and 2.2 births per woman respectively. University of Ghana http://ugspace.ug.edu.gh 51 5.1.5 Relationship between wealth status and parity From Table 5.1, the wealth status of a woman is not significantly associated with her parity, but there are variations within the various categories of wealth status. The poorest women recorded the highest mean parity of 2.5 births per woman, whiles the richest women recorded the least mean parity of 1.8 births per woman. There is a difference in the mean parities of middle and richer women. Women who belong to the middle wealth status category had a mean parity of 1.9 while those who belong to the richer wealth status have a mean parity of 2.3. 5.1.6 Relationship between ethnicity and parity Table 5.1 indicates that there are slight variations in the mean parities o women in the various ethnic groups even though there is no significant relationship between ethnicity and parity. The women in the northern ethnic group recorded the highest mean parity of 2.9, whiles the women from Akan, Ga-Dangme, Ewe and other ethnic groups have relatively same mean parities of 2.1 respectively. 5.1.7 Relationship between education and parity Education is an important demographic factor in the analysis of fertility levels. It widens a woman’s horizon by giving her adequate exposure to ideals that are likely to affect her efficient use of contraception and for her to make informed choices about her fertility. From Table 5.1 above, there is a significant association between a woman’s education and her parity. There is a clear trend from the table which shows that women with no education have the highest mean parity of 3.0 births per woman. The next category of women who have the highest fertility are women with primary education with a mean parity of 2.4 births per woman. Women with middle education University of Ghana http://ugspace.ug.edu.gh 52 recorded mean parity of 2.1 births per woman, whiles women with higher education recorded the least mean parity of 0.9 births per woman. This means that fertility decreases with increasing level of education. 5.1.8 Relationship between religious affiliation and parity The survey collected data on the religious affiliation of the women aged 15-49. While some of the women were not affiliated to any religion, others belonged to the Catholic, Protestant/Pentecostal/Other Christian, Islam and Traditional or Spiritual religions. From Table 5.1, there is no difference in the average parity of women with no religion, Catholic women and Moslems. The mean parity of these women is 2.3 births per woman. Women who are traditionalists recorded the highest parity of 4.5 births per woman, this may be due to the fact that traditional religion is a pronatalist religion (Adongo et.al, 1998). Table 5.2 An independent t-test result showing the relationship between migration status, pregnancy loss and use of contraception and parity Characteristic Mean parity Number of women T-value Significance level Migration status Migrant 2.33 285 2.02 0.044 Non- Migrant 1.93 193 Pregnancy loss(abortion,miscarriage,stillbirth) Yes 2.27 177 0.82 0.409 No 2.10 301 Use of contraception Ever use 2.18 208 0.100 0.920 Non-use 2.16 270 SOURCE: Computed from EDULINK 2011, Data file University of Ghana http://ugspace.ug.edu.gh 53 5.1.9 Relationship between migration status and parity Migration may disrupt fertility, this disruption may either increase fertility or decrease fertility. When migrants live long in their destination areas they may tend to adopt the fertility norms of those places. Moreover if younger migrants stay longer in their destination areas, they may settle and start having children there. For the purpose of this study, place of birth was used as the measure for migration, it was found out that most women in the three communities were not born at their current place of residence, which means that they migrated from other places. From Table 5.2 there is a significant difference in the mean number of children ever born by migrants and non-migrants. Migrants recorded the highest mean parity of 2.3 births per woman, while non-migrants recorded the least mean parity which is 1.9 births per woman. University of Ghana http://ugspace.ug.edu.gh 54 5.1.10 Relationship between pregnancy loss and parity The number of pregnancy losses a woman may have suffered throughout her reproductive years may affect her total number of children ever born. This is because all the children that she lost could have added up to her total fertility if they were alive. Table 5.2 indicates that pregnancy loss is not significantly related to the parity of a woman. The average parity of women who have ever suffered a pregnancy loss is 2.2 whiles that of women who have never suffered any pregnancy loss recorded a mean parity of 2.2 births per woman. 5.1.11 Relationship between contraceptive use and parity Table 5.2 indicates that there is no significant relationship between contraceptive use and parity. The mean parity of women who have ever used any form of contraception is 2.1 whiles that of women who have never used any form of contraception recorded a mean parity of 2.1 births per woman. In conclusion, the bivariate analyses that were used to determine the influence of the independent and control variables on the dependent variable indicated that a woman’s age at first sexual intercourse, current age, marital status, educational background, religion and migration status have a significant relationship with parity or the number of children a woman may have. Other variables such as ethnicity, pregnancy loss, use of contraception, locality of residence and wealth status of women in the study sample are not significantly related to parity. University of Ghana http://ugspace.ug.edu.gh 55 CHAPTER SIX RELATIONSHIP BETWEEN AGE AT FIRST SEXUAL INTERCOURSE AND DEMOGRAPHIC AND SOCIO-ECONOMIC FACTORS AND PARITY 6.1 Introduction This chapter presents the regression analysis of the dependent variable, the independent variable, and the control variables. Various tests of significance to determine associations were done at 95 percent significant levels. Two models were run. A simple linear regression was used to determine the relationship between age at first sexual intercourse and parity in the first model. The combined effects of age at first sexual intercourse of a woman, current age of respondents, marital status, educational background, religious affiliation, contraceptive use, locality of residence, migration status, ethnicity, pregnancy loss and wealth status of respondents on parity were determined using multiple linear regression in the second model. The independent and dependent variables were treated as continuous variables whiles the control variables were treated as categorical variables. In each of these categorical variables, one category was taken to be a reference category (R.C) while the remaining categories were treated as dummy variables. The reference category was represented by the value zero (0). Table 6.0.1 A simple linear regression analysis of the relationship between age at first sexual intercourse and parity Independent variable B Coefficient Standard Error Beta- Coefficient Significant T Significance level Age at first sexual intercourse -0.076 0.028 -0.126 -2.766 0.006 Adjusted R square = 0.014 F statistic =7.648 University of Ghana http://ugspace.ug.edu.gh 56 Table 6.1 shows that about 1.4 percent of the variation in parity is explained by age at first sexual intercourse. The results in the table above shows that age at first sexual intercourse is inversely related to parity, in that a unit change in age at first sexual intercourse will lead to a 0.076 decrease in parity, which means that as age at first sexual intercourse decreases, parity increases. Table 6.2 A multiple regression analysis table showing the influence of the independent variable and control variables on parity Variable Categories B- Coefficie nt Standard Error Beta Co- efficient Significant T Significance value Age at first sexual intercourse -0.099 0.023 -0.163 -4.296 0.000 Marital status Currently married (R.C) Single -1.154 0.206 -0.237 -5.616 0.000 Ever married -0.730 0.200 -0.142 -3.657 0.000 Contraceptive use Non use (R.C) Ever use 0.265 0.156 0.062 1.705 0.089 Pregnancy loss No (R.C) Yes -0.122 0.161 -0.028 0.758 0.449 Migration status Non migrants (R.C) Migrants 0.027 0.168 0.006 0.160 0.873 Ethnicity Akan (R.C) Ga-Dangme -0.055 0.194 -0.013 -0.285 0.776 Ewe -0.154 0.349 -0.017 -0.442 0.658 Northern Ethnic groups -0.111 o.399 -0.011 -0.279 0.781 Other -0.539 0.374 -0.063 -1.443 0.150 Current age 15-19 -3.255 -0.414 -0.364 -7.862 0.000 20-24 -2.906 0.302 -0.546 -9.632 0.000 25-29 -2.139 0.285 -0.389 -7.519 0.000 30-34 -1.467 0.284 -0.262 -5.166 0.000 35-39 -0.945 0.299 -0.153 -3.158 0.002 40-44 -0.209 0.308 -0.032 -0.678 0.498 University of Ghana http://ugspace.ug.edu.gh 57 Variable Categories B- Coefficie nt Standard Error Beta Co- efficient Significant T Significance value 45-49 (R.C) Locality Agbogbloshie (R.C) James Town -0.213 0.252 -0.045 -0.846 0.398 Ussher Town 0.012 0.246 0.003 0.049 0.961 Education No education (R.C) Primary -0.012 0.311 -0.003 -0.040 0.968 Middle -0.414 0.308 -0.097 -1.345 0.179 Higher -0.911 0.351 -0.170 -2.597 0.010 Wealth status Poorest (R.C) Poorer -0.136 0.238 -0.026 -0.570 0.569 Middle -0.425 0.240 -0.080 -1.773 0.077 Richer -0.415 0.251 -0.073 -1.654 0.099 Richest -0.320 0.252 -0.060 -1.267 0.206 Religion Catholic (R.C) No religion 0.169 0.459 0.019 0.368 0.713 Protestant/Pentecostal -0.149 0.346 -0.029 -0.429 0.668 Islam 0.274 0.416 0.037 0.659 0.510 Traditional/Spiritual 1.310 0.728 0.074 1.800 0.072 Adjusted R square = 0.427 F statistic =13.257 SOURCE: Computed from EDULINK 2011, Data file Table 6.2 indicates that 42.7 percent of the variation in parity is explained by age at first sexual intercourse, which is the main independent variable, and control variables which are marital status, contraceptive use, pregnancy loss, migration status, ethnicity, current age of respondents locality of residence, education, wealth status and religion. Table 6.2 indicates that there is a significant relationship between age at first sexual intercourse, marital status, current age, education and parity. The regression results indicate that there is an inverse relationship between age at first sexual intercourse and parity in that the lower a woman’s age at first sexual intercourse, the higher her parity. University of Ghana http://ugspace.ug.edu.gh 58 The results show a highly significant relationship between marital status and parity. Single women show a high significant relationship with parity. There is a 1.15 decrease in parity among women who are single as compared to women who are currently married. The relationship between women who have ever married and parity is also significant, from the results there is a 0.73 decrease in parity among women who have ever married as compared to their currently married counterparts. The regression results indicate that there is a significant relationship between the current age of a woman and parity. The parity of women aged 15-19 is 3.25 points lower than that of women aged 45-49. The results also indicate that for women aged 20-24 their parity is 2.90 points lower than that of women aged 45-49. For the women aged 25-29, their parity is 2.13 points lower than those aged 45-49. The parity of women aged 30-34 is 1.46 points lower than that of women aged 45-49. The parity of women aged 35-39 is 0.94 points lower than that of women who are 45-49. For women who are between ages 40-44, they have a parity of 0.20 points lower than those aged 45- 49. The regression results also show that there is an inverse relationship between educational level of the women aged 15-49 years and parity. This indicates that as educational level increases the mean parity decreases. From the table there is a 0.9 decrease in mean parity as educational level of women increase from no education to higher. The regression results indicate that there is no significant relationship between ethnicity, contraceptive use, pregnancy loss, migration status, locality of residence, wealth status, religion and parity. In summary, the regression results indicate that there is an inverse relationship between age at first sexual intercourse and parity, in that a lower age at first sexual intercourse increases the parity of University of Ghana http://ugspace.ug.edu.gh 59 a woman, hence confirming the hypothesis. Other variables, which are current age of respondents, marital status and educational background of respondents also have a significant relationship with parity. 6.2 Discussion The legal age of consent for sexual intercourse in Ghana is 16 years old ( UNHCR, 2011), while the average age at first sexual intercourse among women in the three selected communities was 17.96. Even though this age is relatively higher than that of the national average, there are still implications for fertility. The average parity of 2.17 births per woman in the three selected communities is still above replacement level and if the average age at first sexual intercourse among women is not checked by raising and legalising a higher age, things may get out of hand. A lower age at first sexual intercourse increases the parity of a woman (Zulu et. al, 2002; Zaba et. al, 2004; Ankomah et. al, 2011). Age at first sexual intercourse is an important indicator of exposure to the risk of pregnancy and a lower age at first sexual intercourse by a woman leads to a prolonged exposure to childbearing hence increasing her parity as confirmed in the multivariate analysis. At the bivariate level, the p value of 0.006 indicated that there was a strong relationship between age at first sexual intercourse and parity. The linear regression analysis indicated that there was an inverse relationship between age at first sexual intercourse and parity. This indicated that a lower a woman’s age at first sexual intercourse the higher her parity. The earlier a woman is exposed to sexual intercourse, the larger her family size. Other predictor variables such as marital status, contraceptive use, current age of respondents, locality of residence, ethnicity, wealth status, pregnancy loss, migration status, religion and education were controlled for to eliminate their effects on parity. The results showed that there was a strong relationship between marriage, current age, education and parity. The combined effects of all these variables still showed that that University of Ghana http://ugspace.ug.edu.gh 60 was a significant relationship between age at first sexual intercourse and parity among women in these three selected urban poor communities of the study. In urban poor communities, where there is a lot of poverty, young women may engage in prostitution that is having sex in exchange of money to cater for themselves (Zulu et. al, 2002). Also the poor housing conditions in urban poor communities may predispose adolescents into having early sexual intercourse. According to Zulu et.al (2002), in urban poor communities where children may share same rooms with their parents, there may not be privacy for sexual intercourse, hence these children may start having sexual intercourse early. Literature supports the fact that as age increases parity also increases. Bongaarts (1983) said that age affects the number of children ever born or parity. The current age of a woman may affect her fertility, the older a woman is the more children she may have had. The reverse is also true, in that the younger a woman, the lower her fertility. Marital status was significantly associated with parity, as Table 6.2 indicated. Even though in Africa, childbearing also occurs outside marriage (Zaba et. al, 2004), many births occur within marriage. Marriage may mark the frequency of sexual intercourse that may put a woman at risk of frequent childbearing and subsequent high parity ( Bongaarts, 2003). Education, from the regression results in Table 6.2 shows an inverse relationship with parity. As educational level increases, parity decreases. As women become educated, they are able to make informed decisions about their fertility and reject cultural practices that may be pronatalist. Also the longer years that a woman may spend in school, may lead to a postponement in her fertility, and subsequently shortening her reproductive years (Basu, 2002: Osili &Long, 2008). University of Ghana http://ugspace.ug.edu.gh 61 Contraceptive use, pregnancy loss, migration status, ethnicity, locality, religion and wealth status were not significantly related to parity. There was no significant relationship between contraceptive use and parity, even though it was expected that contraceptive use should affect fertility. This could be due to the fact that the measure that was used to operationalise contraceptive use that is ‘ever use’ did not actually reflect the relationship between contraceptive use and parity. This is because the women may have ever used a form of modern contraception but may have stopped, even though they may still be in their reproductive years. Table 6.2 indicated that there was no difference in the mean parities of migrants and non-migrants. This could be due to the fact that the migrant women may have stayed longer and hence adopted the fertility norms of their destination areas. Akan women were expected to have higher parity due their pronatalist nature, but the regression results indicated that there was no significant relationship between the mean parities of Akans and the other ethnic groups. There was no significant relationship between the wealth status of women and parity, even though the expected result was that women within the poorest quintile would have higher parities since poor people are thought to have higher parities. Religious affiliation was not significantly related to parity, even though Catholics, Traditionalists and Muslims were expected to have higher parities due to the pronatalist nature of their doctrines and the non-use of contraception by their members. University of Ghana http://ugspace.ug.edu.gh 62 CHAPTER SEVEN SUMMARY, CONCLUSION AND RECOMMENDATIONS 7.1 Summary This study of age at first sexual intercourse and fertility had the main objective of establishing the relationship between age at first sexual activity and parity among women in their reproductive years in three selected urban poor communities in Accra. The specific objectives of the study were to; describe the age at first sexual intercourse and the background characteristics of women in urban poor communities, determine the average age at first sexual intercourse, and to determine the influence of age at first sexual intercourse and parity among women in urban poor communities in Accra as well as to suggest evidence-based interventions for effective policy planning. The data for this study were obtained from the EDULINK 2011 wave two, with a sample size of 478 women between the age groups 15-49 years. Percentages and means were used to describe the background characteristics of the respondents and to determine the relationship between the background characteristics and the mean parity of women in the study sample. A simple linear regression was used to determine the influence of age at first sexual intercourse on parity whilst a multiple linear regression was used to further examine the influence of age at first sexual intercourse on parity among women in the study sample. The mean age at first sexual intercourse of women in the age group 15-49 in the three urban poor communities was found to be 17.96 years. The age structure of the women was found to be relatively young, this is because about 20.1 percent of the women were between the ages of 20-24. About 45 percent of the women in the study sample had Middle education, whiles 51.9 percent were currently married. Protestants, Pentecostals and Other Christians formed 77.6 percent of the total sample of women. University of Ghana http://ugspace.ug.edu.gh 63 The ethnic groups that were predominant in the three communities were Ga-Dangme representing 54.2 percent, while 28.2 percent of the women were Akan. About 6 percent of the sample of women were Ewes while 5.0 percent and 6.7 percent belonged to Northern and Other ethnic groups respectively. Nearly 52 percent of the women lived in Ga Mashie, 29.3 percent lived in James Town whiles 18.8 percent of the women lived in Agbogbloshie. Migrants constituted 59.6 percent of the women in the study sample, while 40.4 percent were non migrants, 56.5 percent of women in the study sample recorded that they have ever used a form of contraception whiles 43.5 percent of the women have never used any form of contraception. 37 percent of the women have ever lost a pregnancy while 63 percent have never lost a pregnancy. With wealth status, 22.8 percent of the women belonged to the poorest quintile. The mean parity of the women was 2.17 births per woman, but the mean parities varied by the various socio-demographic characteristics. For instance, the mean parity of women within the various age groups differed. As age increased, the mean parities of the women increased, as women moved from age 15 to 49 years, the mean parity increased from 0.37 to 3.77. Linear regression analysis was used to determine the effects of age at first sexual intercourse on parity separately and jointly with the background characteristics. The results of both regression analyses showed an inverse relationship between age at first sexual intercourse and parity, meaning that the lower a woman’s age at first sexual intercourse the higher her parity. The results in Table 6.2 further showed that parity increased with advancement in the current age of the respondents. University of Ghana http://ugspace.ug.edu.gh 64 7.2 Conclusion The study has shown that low age at first sexual intercourse contributes to a high parity among women aged 15 to 49 years in the selected urban poor communities in Accra, which confirms the hypothesis. Economic stresses such as unemployment that exist within urban poor communities may have contributed to this low age at first sexual intercourse among the woman. Young women may have sexual intercourse with men so as to get some money to cater for themselves. According to Zulu et. al (2002), residential arrangements in urban poor communities that do not allow for privacy of sexual intercourse within households predispose young women to very early sexual initiation. Lower age at first sexual intercourse that predisposes a woman to a prolonged period of childbirth and subsequently increasing her parity has socio-demographic implications. A high parity among women may lead to a young population structure that must be educated and absolved into the labour force. Considering the economic situation in Ghana, such rise in the population structure will put pressure on the limited education and employment facilities that exist. Further, an early age at first sexual intercourse predisposes a woman to many viral infections such as HIV and cervical cancer (Kaestle et. al, 2004). The projected rapid population growth in African cities and the widening disparity between the rich and the poor in them have obvious implications for measures to control fertility. If age at first sexual intercourse can be increased so as to reduce parity in urban poor communities, educational campaigns should be organized by social workers and NGOs to educate young women on the negative implications of early age at sexual initiation. University of Ghana http://ugspace.ug.edu.gh 65 Considerable work remains to be done in the research area of the study presented in this study. The results suggest that a lower age at first sexual intercourse increases the parity of a woman, but these results need more support in the form of direct evidence from surveys in other urban poor communities. It is also crucial to establish whether lower age at first sexual intercourse, is truly a consequence of living in an urban poor community or whether it is in fact a consequence of poverty in general. 7.3 Recommendation Laws governing the age at sexual intercourse in Ghana should be made stricter by raising the age, this would help manage young women’s sexual debut and check their risks to mistimed and unwanted pregnancies that could increase their parities. Extensive campaign on the knowledge and use of contraception and family planning methods should be embarked on by government, NGOs and other stakeholders to educate women in these three urban poor communities on the use and benefits of modern family planning methods. Since education is significantly related with parity, women should be encouraged to attend school, especially secondary and higher education. Due to the long duration of secondary and tertiary education, girls may postpone marriage and thus childbearing, subsequently reducing their fertility. Education of women in urban poor communities may also enable women to use contraception which may also reduce their parity. Religious bodies can also embark on campaigns to teach the women especially, the young women in their congregation about delaying sexual debut for a longer period until such a time that they are ready, or married. University of Ghana http://ugspace.ug.edu.gh 66 REFERENCES Addai, I. (2014). Does religion matter in contraceptive use among Ghanaian women ?, 40(3), 259–277. Agyei-Mensah, S. (2006). Fertility transition in Ghana : looking back and looking forward, 477, 461–477. Akintola, O. N. L. (2012). “I did it for him, not for me”: An explanatory study of factors influencing sexual debut among female university students in Durban, South Africa: A cohort study. BMJ. Akuffo, F. O. (1987). Teenage pregnancies and school drop outs, the relevance of family life education and vocational training to girls employment opportunities. Amos, P. M. (2006). Parenting and culture – evidence from some African communities. Ankomah A.,Mamman-Duara F,Omoregie G, A. J. (2011). Reasons for delaying or engaging in early sexual initiation among adolescents in Nigeria, Adolescent health medicine and therapeutics. Aryeetey, E. (2011). The global food and financial crises and the poor in Africa. Asampong, E., Osafo, J., Bingenheimer, J. B., & Ahiadeke, C. (2013). Adolescents and parents’ perceptions of best time for sex and sexual communications from two communities in the Eastern and Volta Regions of Ghana: implications for HIV and AIDS education. BMC International Health and Human Rights, 13(1), 40. doi:10.1186/1472-698X-13-40. Banchani, E., & Tenkorang, E. Y. (2014). Implementation challenges of maternal health care in Ghana: the case of health care providers in the Tamale Metropolis. BMC Health Services Research, 14, 7. doi:10.1186/1472-6963-14-7 Bankole, A. & Singh S.(1998).Couple's fertility and contraceptive decision making in developing countries: hearing the man's voice. International family planning perspectives. vol.24(1): pp 15-24 Basu, A. M. (2002). Why does education lead to lower fertility? A critical review of some of the possibilities. World Development, 30(10), 1779–1790. doi:10.1016/S0305-750X(02)00072- 4 Blanc, A. K., & Way, A. A. (2014). Sexual behaviour and contraceptive knowledge and use among developing countries, 29(2), 106–116. Bledsoe C., H. J. (1999). Critical perspectives on schooling and fertility in developing world. National Academy Press, Washington D.C. University of Ghana http://ugspace.ug.edu.gh 67 Blum, R. W., Beuhring, T., Shew, M. L., Bearinger, L. H., Sieving, R. E., & Resnick, M. D. (2000). The effects of race/ethnicity, income, and family structure on adolescent risk behaviors. American Journal of Public Health, 90(12), 1879–84. Retrieved from http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=1446419&tool=pmcentrez&ren dertype=abstract Bongaarts, J. (1983). Fertility, biology and behaviour: An analysis of the proximate determinants. Academic Press Inc. New York. Bongaarts, J. (2003). Completing the fertility transition in the developing world: the role of educational differences and fertility preferences. Population Studies, 57(3), 321–35. doi:10.1080/0032472032000137835 Bongaarts, J. (2011). Can family planning programmes reduce high desired family size in Sub- Saharan Africa,international perspectives on sexual and reproductive health: Guttmacher Institute. Bongaarts, J., & Casterline, J. (2013). Fertility Transition: Is Sub-Saharan Africa different? Population and Development Review, 38(Suppl 1), 153–168. doi:10.1111/j.1728- 4457.2013.00557.x Brockerhoff, M. (1995). Fertility and family planning in African cities , the impact of female migration. Journal of Biosocial Science, Vol. 27(issue 03). Caldwell, J. C. (1982). Theory of fertility decline. Academic Press, London. Collins, R. L., Elliott, M. N., Berry, S. H., Kanouse, D. E., Kunkel, D., Hunter, S. B., & Miu, A. (2004). Watching sex on television predicts adolescent initiation of sexual behavior. Pediatrics, 114(3), e280–9. doi:10.1542/peds.2003-1065-L Denzer, L., & Ebeling, M. F. E. (2002). Understanding fertility in Africa, (11). Dodoo, F.N.A., Zulu, M. E. & Ezeh, A. C. (2006). Urban-Rural differences in the socio- economic deprivation-sexual behaviour link in Kenya. Erkut, S., Grossman, J. M., Frye, A. A., Ceder, I., Charmaraman, L. &, & Tracy, A. J. (2012). Can sex education delay early sexual debut? The Journal of Early Adolescence, 33(4), 482– 497. doi:10.1177/0272431612449386 Ezeh, A. C., Kodzi, I., & Emina, J. (2010). Reaching the urban poor with family planning services. Studies in Family Planning, 41(2), 109–16. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/21466110 Farrar, K. M. (2012). Sex on television: A review of socialization effects and the role of context and individual differences. University of Ghana http://ugspace.ug.edu.gh 68 Fraser, A. ., Brockert, J. . ., & Ward, R. H. (1995). Association of young maternal age with adverse reproductive outcomes. The New England Journal of Medicine, 332(17), 1113–7. doi:10.1056/NEJM199504273321701 Frøen, J. Cacciatore, F. & McClure, E. (2011). Stillbirths: why they matter. The Lancet’s Stillbirths Series Steering Committee. doi:10.1016/S0140- 6736(10)62232-5. GAMADA, Factsheet. (2008). Background Information of Ga Mashie (Old Accra) Development Project. Accra,Ghana. Ghana Statistical Service Report. (2005). Ghana Statistical Service Report. (2010). Ghana Statistical Service, Ghana Health Service, ICF, Macro, Ghana Demographic and Health Survey, 2008. (2009). N. GSS, GHS,ICF Macro. Goldenberg, R. , McClure, L. & Bhutta, A. (2011). Stillbirths: the vision for 2020. Lancet, 377(9779), 1798–805. doi:10.1016/S0140-6736(10)62235-0 Goldstein S., Goldstein, A. (1983). Migration and Fertility in Peninsula Malaysia: An analysis using life history data. Grant, M. J., & Hallman, K. (2008). Prior School Performance in Africa, 39(4), 369–382. Green, T. (2001). The impact of religious organisations in promoting HIV /AIDS prevention. Paper Presented at the Challenges For The Church: AIDS, Malaria and Tuberculosis, Vol.3, 25–26. Greif, M. J., Dodoo, F. N. A., & Jayaraman, A. (2010). Urbanisation, Poverty and Sexual Behaviour: The Tale of five African cities. Urban Studies, 48(5), 947–957. doi:10.1177/0042098010368575 Guttmacher Institute Report. (2014). Hardoy, J. E., Mitlin, D. Scatterthwaite, D. (2014). Environmental problems in an urbanizing world. Henry, R. & Fayorsey, C. (2002). Coping with Pregnancy Experiences of Adolescents in Ga Mashi Accra. Imoagene, A. (1990). The Ibo of East-Central Nigeria. New Era Publishers. Ibadan, Nigeria. Kabiru, C. W., Beguy, D., & Undie, C. (2010). Transition into first sex among adolescents in slum and non-slum communities in Nairobi , Kenya, (June 2014), 37–41. doi:10.1080/13676261003801754 University of Ghana http://ugspace.ug.edu.gh 69 Kabiru, C. W., Beguy, D., Undie, C.-C., Zulu, E. M., & Ezeh, A. C. (2010). Transition into first sex among adolescents in slum and non-slum communities in Nairobi, Kenya. Journal of Youth Studies, 13(4), 453–471. doi:10.1080/13676261003801754 Kaestle, C. E., Halpern, C. T., Miller, W. C., & Ford, C. . . (2005). Young age at first sexual intercourse and sexually transmitted infections in adolescents and young adults. American Journal of Epidemiology, 161(8), 774–80. doi:10.1093/aje/kwi095 Kaplan, H. (1996). A theory of fertility and parental investment in traditional and modern human societies, 91–135. Khan, S., & Mishra, V. (2008). Youth Reproductive and Sexual Health,DHS Comparative Report 19, (August). Kollehlon, K. T. (1994). Religious Affiliation and Fertility in Liberia,Journal of Biosocial science, Vol,26, 493–507. Kollehlon, K. T. (2003). Ethnicity and fertility in Nigeria. Biodemography and Social Biology, 50(3-4), 201–221. doi:10.1080/19485565.2003.9989072 Kulu, H. (2013). Why do fertility levels vary between urban and rural areas? Regional Studies, 47(6), 895–912. doi:10.1080/00343404.2011.581276 Levandowski, B., Kalilani-Phiri, L., Kachale, F., Awah, P., Kangaude, G., & Mhango, C. (2012). Investigating social consequences of unwanted pregnancy and unsafe abortion in Malawi: the role of stigma. International Journal of Gynaecology and Obstetrics: The Official Organ of the International Federation of Gynaecology and Obstetrics, 118 Suppl, S167–71. doi:10.1016/S0020-7292(12)60017-4 Lockwood, M. (2014). Structure and Behavior in the Social Demography of Africa, 21(1), 1–32. Magadi, M. (2013). The disproportionate high risk of HIV infection among the urban poor in sub-Saharan Africa. AIDS and Behavior, 17(5), 1645–54. doi:10.1007/s10461-012-0217-y Marino, J. L., Skinner, S. R., Doherty, D., Rosenthal, S. L., Cooper Robbins, S. C., Cannon, J., & Hickey, M. (2013). Age at menarche and age at first sexual intercourse: a prospective cohort study. Pediatrics, 132(6), 1028–36. doi:10.1542/peds.2012-3634 Mberu B, W. M. (2011). Internal Migration And Health,Premarital Sexual Initiation in Nigeria. Social Science & Medicine, 72. Mott, F., Fondell, M., Hu, P. N., Kowaleski-Jones, L., & Menaghan, E. (1996). Adolescent Population, 28(1), 13–18. Mullen, S. (2001). Parent- Adolescent Reproductive Health Health Communication in Lome Togo, Does it make a Difference? 129th Annual Meeting of APHA. University of Ghana http://ugspace.ug.edu.gh 70 Ngom, P., Magadi, M., & Owuor, T. (2003). Parental presence and adolescent reproductive health among the Nairobi urban poor. Journal of Adolescent Health, 33(5), 369–377. doi:10.1016/S1054-139X(03)00213-1 Nield, J., Magnusson, B., Chapman, D., Lapanekale, L. (2014). Age at Menarche and Risky Sexual partnerships in adulthood: Does a Biosocial model explain any associations? Journal of women’s Health. Odimegwu, C. (2005). Influence of religion on adolescent sexual attitudes and behaviour among Nigerian University students: Affiliation or committment. Okuma , M., & Chege, I. (1994). Females adolescent health and sexuality in Kenyan secondary schools. A survey report,Nairobi,Kenya. Medical Research Foundation. Osili, U. O., & Long, B. T. (2008). Does female schooling reduce fertility? Evidence from Nigeria. Journal of Development Economics, 87(1), 57–75. doi:10.1016/j.jdeveco.2007.10.003 Owusu, G., & Afutu-kotey, R. L. (2010). Poor Urban Communities and Municipal Interface in Ghana : A Case Study of Accra and Sekondi-Takoradi Metropolis, 12(1), 1–16. Parker, J. (2000). Making the town: Ga state and society in early colonial Accra. Oxford: James Currey. Parkes, A., Wight, D., Hunt, K., Henderson, M., & Sargent, J. (2013). Are sexual media exposure, parental restrictions on media use and co-viewing TV and DVDs with parents and friends associated with teenagers’ early sexual behaviour? Journal of Adolescence, 36(6), 1121–33. doi:10.1016/j.adolescence.2013.08.019 Pfeiffer, C., Kleeb, M., Mblewa, A., Ahorlu, C. (2014). The use of social media among adolescents in Dar es Salaam and Mtwara Tanzania. Population Reference Bureau, Factsheet. (2013). Prata, N., Weidert, K., & Sreenivas, A. (2013). Meeting the need: youth and family planning in sub-Saharan Africa. Contraception, 88(1), 83–90. doi:10.1016/j.contraception.2012.10.001 Quartey-Papafio, J. (2006). “Changing Livelihoods in Ga Mashie, A Coastal Settlement in Southern Ghana from 1957-2004.” Ph.D Thesis, Department of Geography and Resource Development, University of Ghana,2006,Legon, Accra. Romanuik, A. (2011). Persistence Of High Fertility in Tropical Africa: The case of the Democratic Republic Of Congo. Population and Development Review, 37, 1–29. Sander, W. (1992). Catholicism and the Economics of Fertility. Population Studies, 46(3), 477– 489. doi:10.1080/0032472031000146466 University of Ghana http://ugspace.ug.edu.gh 71 Schwandt, H., Creanga, A., Adanu, R., Danso, K., Agbenyega, T., & Hindin, M. (2013). Pathways to unsafe abortion in Ghana: the role of male partners, women and health care providers. Contraception, 88(4), 509–17. doi:10.1016/j.contraception.2013.03.010 Sear, R .Coall, D. (2011). How much does family matter? cooperative breeding and the demographic transition. Population Development Review, 37, 81–112. Shapiro, D. (2014). Enduring economic hardship, women’s education, marriage and fertility transition in Kinshasa. Journal of Biosocial Science, 1–17. doi:10.1017/S0021932014000091 Simons, J. (2003). Religious Population: doctrines of "Paul Demeny and Geoffrey McNicoll (eds), Encyclopedia of population,New York, Macmillan Reference ,USA, Vol 2. Smock, P. & Greenland, F. (2010). Diversity in pathways to parenthood in the U.S: patterns, implications and emerging research directions. Journal of Marriage and Family, Vol. 72. Speizer, I. S., Fotso, J. C., Davis, J. T., Saad, A., & Otai, J. (2013). Timing and circumstances of first sex among female and male youth from select urban areas of Nigeria, Kenya, and Senegal. The Journal of Adolescent Health : Official Publication of the Society for Adolescent Medicine, 53(5), 609–16. doi:10.1016/j.jadohealth.2013.06.004 Tawiah, E.O . (1984). Determinants of cumulative fertility in Ghana. Demography, 21, 1–8. UNFPA. (2012). United Nations Population Fund Report. United Nations Millennium Development Goal Report: New York. (2012). United States Census Bureau Factsheet. (2013). Weeks, J.(2008). Population. (B. T. Caldeira, Chris, Weeks Deanna, Ed.) (pp. 171–173). Wadworth Language Learning. Weeks, J., Allan, H. (2013). Introduction to the Accra school: An overview of Health, Poverty and Place in Accra. WHO/UNAIDS/UNICEF: Global HIV/AIDS response: Epidemic update and health sector progress towards universal access, progress report. (2011). World Health Organisation Report. (2014). Zaba, B, Pisani E, Boerma, J. (2004). Age at first sex: understanding recent trends in African demographic surveys. Zulu,E., Dodoo, F.N.A, Ezeh, A. (2002). Sexual risk-taking in the slums of Nairobi, Kenya, 1993-8. Population Studies, 56(3), 311–23. doi:10.1080/00324720215933 University of Ghana http://ugspace.ug.edu.gh 72 University of Ghana http://ugspace.ug.edu.gh