University of Ghana http://ugspace.ug.edu.gh UNIVERSITY OF GHANA COLLEGE OF HUMANITIES BRIDEWEALTH PAYMENT AND MALE-PERPETRATED INTIMATE PARTNER VIOLENCE IN GHANA BY CHARLOTTE ABRA ESIME OFORI (10443245) THIS THESIS IS SUBMITTED TO THE UNIVERSITY OF GHANA, LEGON IN PARTIAL FULFILLMENT OF THE REQUIREMENT FOR THE AWARD OF DOCTOR OF PHILOSOPHY POPULATION STUDIES DEGREE REGIONAL INSTITUTE FOR POPULATION STUDIES JULY, 2019 University of Ghana http://ugspace.ug.edu.gh ACCEPTANCE Accepted by the College of Humanities, University of Ghana, Legon, in fulfillment of the requirement for the award of PhD in Population Studies degree. Thesis supervisors: 10/08/2020 Prof. Francis Nii Amoo Dodoo Date 10/08/2020 Dr. Naa Dodua Dodoo Date 01/08/2020 Dr. Adriana Andrea Ewurabena Biney Date i University of Ghana http://ugspace.ug.edu.gh DECLARATION I hereby declare that this work is the result of my own research undertaken under supervision except for references made to other people’s work and which have been duly acknowledged. Also, this work has neither in part nor whole been presented for another degree elsewhere. 10/08/2020 Charlotte Abra Esime Ofori Date (Student) ii University of Ghana http://ugspace.ug.edu.gh ACKNOWLEDGEMENT To the One who gives undeserved grace, I thank you. I am grateful to my advisors, Prof. Francis Dodoo, Dr. Naa Dodua Dodoo and Dr. Adriana Biney for all your support, constructive instruction and love throughout my graduate studies and for supervising this thesis. I thank you all for investing your time and resources in me. I am grateful to the William and Flora-Hewlett Foundation for funding my post-graduate studies. Also, my sincere gratitude goes to the entire faculty of the Regional Institute for Population Studies, who in various ways contributed to the success of this work. I am also grateful to Ghana Statistical Service and the DHS Program for giving me unrestricted access to the 2014 GDHS survey data for this academic work. I am indebted to my Mother, Peace Deh, my Father, Cephas Ofori (forever in our hearts), Dr. Delali Badasu, Dr. Ayaga Bawah, Dr. Abu Mumuni, my siblings, Cephas, Godfred, Eric and Belinda, and Mawusi for all their love and commitment to helping me get this far. To Akua Obeng- Dwamena, Martin Wiredu Agyekum, Isaac Yeboah, Crystal Bubune Letsa, Dr. Maame B. Peterson and Dr. Micheal Odame, God bless you all for holding my hands through it all. Thank you all! Charlotte Abra Esime Ofori iii University of Ghana http://ugspace.ug.edu.gh ABSTRACT Physical aggression, psychological mistreatment and sexual abuse against women in intimate unions are public health concerns globally. They constitute infringements on human rights, and are a huge economic burden to nations. Research suggests that individual level factors and women’s subordinate position to men predispose them to male-perpetrated intimate partner violence (IPV). In sub-Saharan Africa where marriages are near universal, little is known about the cultural context within which marriages are arranged and the implications for IPV. Of particular interest, and central to marriage in the sub-region, is bridewealth payment which legitimizes unions and maintains kinship ties. Over time, bridewealth payment has been misconstrued by some to mean that the woman has been bought and is the property of the man. Although this cultural practice is widespread, and has persisted over time, there is little empirical data on the relationship between various aspects of bridewealth payment and IPV. This study examined how bridewealth payment, specifically, whether or not bridewealth has been negotiated and completeness of bridewealth payment, explain men’s attitudes toward wife- beating and actual perpetration of physical violence against women. Using a mixed-methods approach, the study relies on a nationally representative dataset, 2014 Ghana Demographic and Health Survey (survey sample size = 1,893 men), and a population-based study conducted in selected communities in the Eastern Region of Ghana, Bridewealth Payment and Normative Constraints on Women’s Lives in Ghana (survey sample size = 579 men and 16 key informant in-depth interviews). The 2014 Ghana Demographic and Health Survey examined men’s attitudes toward wife-beating while the Bridewealth Payment and Normative Constraints on Women’s Lives in Ghana Study examined men’s self-reported actual perpetration of abuse. The results showed that both men’s approval of wife-beating and self-reported actual perpetration of iv University of Ghana http://ugspace.ug.edu.gh abuse against female partners were prevalent. While about one in ten men approved of wife- beating, nearly one in five men indicated ever physically abusing their current partners. Bridewealth payment was widespread. A large proportion of men indicated that bridewealth had been negotiated in their current union. With regard to completeness of payment, more than half of the men in both samples reported that bridewealth has been fully paid in their current union. The study showed that the implications of bridewealth payment on male-perpetrated intimate partner violence are complex. Completeness of bridewealth payment did not significantly predict men’s approval of wife-beating, but completeness of bridewealth payment was significantly associated with actual perpetration of abuse. Contrary to previous assertion, men were more likely to have ever beaten their wives when bridewealth has not been paid compared to when bridewealth has been fully paid. Religious affiliation, level of education, type of earning, and ethnicity were significant predictors of men’s attitudes toward wife-beating. The predictors of actual perpetration of abuse by men were religious affiliation, age, duration of marriage, number of children ever born, observing violence as a child, and ethnicity. Key informant interviews also showed that the implications of bridewealth payment on men’s use of violence in intimate unions are multifaceted. All key informants suggested that bridewealth payment does not give the husband authority to use violence against his wife in the union. Bridewealth payment places the man in a position of responsibility, ensuring that his wife and family are safe from harm. Bridewealth is exchanged to honour all stakeholders involved. The social support and control developed when bridewealth is exchanged potentially prevent men from being violent. Rather, the narratives showed that non-payment and partial bridewealth payment are related to conflict and abuse. Couples may argue on issues regarding legitimizing their unions and these are pointers of conflicts. The relative authority of the man may also be challenged with incomplete v University of Ghana http://ugspace.ug.edu.gh bridewealth payment, and this has implications on “wounded masculinities”, and hence the use and approval of violence. Further, the narratives showed that when bridewealth is paid the man may reprimand his partner or use violence if the woman does not perform certain duties. Therefore, the study concludes that the practice of bridewealth has vital implications for male- perpetrated IPV. The very practical consequences this important cultural custom has for women’s health and wellbeing cannot be overemphasized. It is necessary to focus on context, bearing in mind cultural norms, in the discourse of IPV and in developing interventions in sub- Saharan Africa if research seeks to reduce violence against women and reduce the health and economic burden of this menace. vi University of Ghana http://ugspace.ug.edu.gh TABLE OF CONTENTS ACCEPTANCE ............................................................................................................................... i DECLARATION ............................................................................................................................ ii ACKNOWLEDGEMENT ............................................................................................................. iii ABSTRACT ................................................................................................................................... iv TABLE OF CONTENTS .............................................................................................................. vii LIST OF TABLES ....................................................................................................................... xiii LIST OF FIGURES ..................................................................................................................... xiv LIST OF APPENDICES ............................................................................................................... xv LIST OF ABBREVIATIONS ...................................................................................................... xvi CHAPTER ONE ............................................................................................................................. 1 INTRODUCTION .......................................................................................................................... 1 1.1 Background to the study ........................................................................................................ 1 1.2 Statement of the problem ...................................................................................................... 7 1.3 Research questions of the study .......................................................................................... 12 1.4 Rationale for the study ........................................................................................................ 12 1.5 Objectives of the study ........................................................................................................ 16 1.6 Organization of the study .................................................................................................... 16 CHAPTER TWO .......................................................................................................................... 18 LITERATURE REVIEW ............................................................................................................. 18 2.1 Introduction ......................................................................................................................... 18 2.2 Extant measures of intimate partner violence ..................................................................... 19 2.2.1 Physical intimate partner violence ................................................................................ 20 2.2.2 Sexual intimate partner violence .................................................................................. 21 2.2.3 Psychological/emotional intimate partner violence ...................................................... 21 2.2.4 Economic intimate partner violence ............................................................................. 22 2.3 Prevalence of intimate partner violence .............................................................................. 24 2.4 Prevalence of IPV in sub-Saharan Africa ........................................................................... 26 2.5 Some predictors of male-perpetrated intimate partner violence ......................................... 28 vii University of Ghana http://ugspace.ug.edu.gh 2.5.1 Attitudes toward intimate partner violence .................................................................. 29 2.5.2 Male dominance and intimate partner violence ............................................................ 32 2.5.3 Gender differences and the risk of intimate partner violence ....................................... 34 2.5.4 Poverty/economic dependence on men and intimate partner violence ......................... 37 2.5.5 Social support and intimate partner violence ............................................................... 39 2.5.6 Education and intimate partner violence ...................................................................... 40 2.5.7 Observing violence as a child and intimate partner violence ....................................... 42 2.5.8 Alcohol use and intimate partner violence ................................................................... 44 2.5.9 Religion and intimate partner violence ......................................................................... 46 2.6 Bridewealth payment in sub-Saharan Africa ...................................................................... 48 2.6.1 Bridewealth payment and intimate partner violence (IPV) .......................................... 52 2.7 Legislation on domestic violence in Ghana ........................................................................ 55 2.8 Gaps in the Literature .......................................................................................................... 59 2.9 Theories and conceptualization ........................................................................................... 60 2.9.1 Integrated ecological framework .................................................................................. 61 2.9.2 Resource theory ............................................................................................................ 65 2.9.3 Exchange/Social control theory .................................................................................... 66 2.10 Conceptual framework ...................................................................................................... 69 CHAPTER THREE ...................................................................................................................... 73 DATA SOURCES, SAMPLING PROCEDURE AND METHODOLOGY ............................... 73 3.1 Introduction ......................................................................................................................... 73 3.2 The 2014 Ghana Demographic and Health Survey ............................................................. 75 3.3 Bridewealth Payment and Normative Constraints on Women’s Lives in Ghana (Bridewealth Study) .................................................................................................................. 77 3.4 Qualitative data collection ................................................................................................... 80 3.5 Measurement and description of variables .......................................................................... 81 3.5.1 Objectives 1 and 2: bridewealth negotiation status, completeness of bridewealth payment and attitudes toward wife-beating among men ....................................................... 81 3.5.1.1 Dependent variable 1: attitudes toward wife-beating ............................................ 81 3.5.1.2 Main independent variable 1: bridewealth negotiation status................................ 84 3.5.1.3 Main independent variable 2: completeness of bridewealth payment ................... 84 viii University of Ghana http://ugspace.ug.edu.gh 3.5.2 Objectives 3 and 4: whether or not bridewealth has been negotiated, completeness of bridewealth payment and physical violence perpetration by men ......................................... 85 3.5.2.1 Dependent variable 2: physical violence perpetration ........................................... 85 3.5.2.2 Main independent variable1: bridewealth negotiation status................................. 85 3.5.2.3 Main independent variable 2: completeness of bridewealth payment ................... 86 3.5.3 Control variables........................................................................................................... 87 3.6 Methods of analyses ............................................................................................................ 90 3.6.1 Quantitative data analyses ............................................................................................ 90 3.6.2 Qualitative data analysis ............................................................................................... 92 3.7 Limitations of the study....................................................................................................... 93 CHAPTER FOUR ......................................................................................................................... 95 BACKGROUND CHARACTERISTICS OF STUDY RESPONDENTS ................................... 95 4.1 Introduction ......................................................................................................................... 95 4.2 Ghana Demographic and Health Survey, 2014 ................................................................... 95 4.2.1 Men’s attitudes toward wife-beating, GDHS ............................................................... 95 4.2.2 Bridewealth negotiation status, GDHS ......................................................................... 97 4.2.3 Completeness of bridewealth payment, GDHS ............................................................ 97 4.2.4 Socio-demographic and economic characteristics of respondents, GDHS .................. 98 4.3 2016 Bridewealth Study .................................................................................................... 101 4.3.1 Physical violence perpetration, Bridewealth Study .................................................... 101 4.3.2 Bridewealth negotiation status, Bridewealth Study .................................................... 101 4.3.3 Completeness of bridewealth payment, Bridewealth Study ....................................... 102 4.3.4 Socio-demographic and economic characteristics of respondents, Bridewealth Study ............................................................................................................................................. 103 4.4 Socio-demographic characteristics of qualitative study participants, Bridewealth Study 105 4.5 Discussion ......................................................................................................................... 106 CHAPTER FIVE ........................................................................................................................ 109 BRIDEWEALTH PAYMENT, SOCIO-DEMOGRAPHIC AND ECONOMIC CORRELATES OF MEN’S ATTITUDES TOWARD WIFE-BEATING AND ACTUAL PERPETRATION OF PHYSICAL ABUSE ................................................................................................................... 109 5.1 Introduction ....................................................................................................................... 109 5.2 Bivariate results on men’s attitudes toward wife-beating, GDHS .................................... 110 ix University of Ghana http://ugspace.ug.edu.gh 5.2.1 Bridewealth negotiation status and men’s attitudes toward wife-beating .................. 110 5.2.2 Completeness of bridewealth payment and men’s attitudes toward wife-beating, GDHS .................................................................................................................................. 111 5.2.2.2 Association between socio-demographic characteristics and men’s attitudes toward wife-beating, GDHS ........................................................................................................ 113 5.3 Physical violence perpetration, Bridewealth Study........................................................... 118 5.3.1 Bridewealth negotiation status and physical violence perpetration, Bridewealth Study ............................................................................................................................................. 118 5.3.2 Completeness of bridewealth payment and physical abuse perpetration, Bridewealth Study .................................................................................................................................... 119 5.3.3 Association between socio-demographic characteristics and physical abuse, Bridewealth Study ............................................................................................................... 120 5.4 Discussion ......................................................................................................................... 122 CHAPTER SIX ........................................................................................................................... 125 ASSOCIATION BETWEEN COMPLETENESS OF BRIDEWEALTH PAYMENT AND MEN’S ATTITUDES TOWARD WIFE-BEATING CONTROLLING FOR OTHER SELECTED CHARACTERISTICS ........................................................................................... 125 6.1 Introduction ....................................................................................................................... 125 6.2 Bridewealth negotiation status and men’s attitudes toward wife-beating ......................... 126 6.3 Completeness of bridewealth payment and other correlates of men’s attitudes toward wife- beating ..................................................................................................................................... 129 6.4 Discussion ......................................................................................................................... 133 CHAPTER SEVEN .................................................................................................................... 139 BRIDEWEALTH PAYMENT AND PERPETRATION OF INTIMATE PARTNER VIOLENCE BY MEN ................................................................................................................ 139 7.1 Introduction ....................................................................................................................... 139 7.2 Bridewealth negotiation status and having ever perpetrated intimate partner physical abuse ................................................................................................................................................. 139 7.4 Discussion ......................................................................................................................... 144 CHAPTER EIGHT ..................................................................................................................... 151 PERCEPTION OF BRIDEWEALTH PAYMENT AND INTIMATE PARTNER VIOLENCE AMONG THE PARTICIPANTS ............................................................................................... 151 8.1 Introduction ....................................................................................................................... 151 8.2 Reasons for or benefits of bridewealth payment ............................................................... 152 x University of Ghana http://ugspace.ug.edu.gh 8.2.1 Honour to parents, wife and husband ......................................................................... 152 8.2.2 Bridewealth payment legitimizes the union ............................................................... 154 8.2.3 Women are valued ...................................................................................................... 155 8.2.4 Protect women ............................................................................................................ 155 8.2.5 Right to inheritance .................................................................................................... 156 8.2.6 Economic support/security ......................................................................................... 157 8.2.7 Discussion ................................................................................................................... 157 8.3 Implications of completeness of bridewealth payment ..................................................... 158 8.3.1 Implications of non-payment of bridewealth .............................................................. 159 8.3.1.1 Breeds conflicts .................................................................................................... 159 8.3.1.2 Woman has been “stolen” .................................................................................... 161 8.3.1.3 Physical abuse/beating ......................................................................................... 161 8.3.1.4 Children do not belong to their biological father ................................................. 162 8.3.1.5 Authority .............................................................................................................. 163 8.3.1.6 No sanctions against the woman .......................................................................... 164 8.3.1.7 Poor parental support ........................................................................................... 165 8.3.1.8 Non-payment does not guarantee abuse .............................................................. 166 8.3.2 Implications of partial bridewealth payment .............................................................. 166 8.3.2.1 Breeds conflicts .................................................................................................... 167 8.3.2.2 Man has some authority ....................................................................................... 168 8.3.2.3 Woman is not valued ........................................................................................... 169 8.3.2.4 Children do not belong to the man ....................................................................... 169 8.3.2.5 Type of union ....................................................................................................... 170 8.3.2.6 No sanctions if the woman cheats ........................................................................ 171 8.3.2.7 Poor parental support ........................................................................................... 172 8.3.3 Implications of full bridewealth payment ................................................................... 173 8.3.3.1 Position of responsibility ..................................................................................... 174 8.3.3.2 Man has authority over the woman ...................................................................... 175 8.3.3.3 Does not guarantee physical abuse ...................................................................... 176 8.3.3.4 Woman has not been bought ................................................................................ 177 8.3.3.5 Physical abuse/beating ......................................................................................... 178 xi University of Ghana http://ugspace.ug.edu.gh 8.3.3.6 Authority to reprimand the woman ...................................................................... 179 8.3.3.7 Social support....................................................................................................... 179 8.3.3.8 Discussion ............................................................................................................ 181 8.4 Sanctions/strategies for dealing with abuse ...................................................................... 186 8.4.1 Divorce ....................................................................................................................... 187 8.4.2 Legal or police case .................................................................................................... 188 8.4.3 Separation ................................................................................................................... 188 8.4.4 Conflict resolution ...................................................................................................... 189 8.4.5 Warning ...................................................................................................................... 190 8.4.6 Deny the husband sex ................................................................................................. 190 8.4.7 Discussion ................................................................................................................... 191 CHAPTER NINE ........................................................................................................................ 193 SUMMARY, RECOMMENDATIONS AND CONCLUSIONS............................................... 193 9.1 Introduction ....................................................................................................................... 193 9.2 Summary ........................................................................................................................... 194 9.2.1 Bridewealth payment and men’s attitudes toward wife-beating ................................ 194 9.2.2 Bridewealth payment and men’s self-reported actual perpetration of violence ......... 197 9.3 Recommendations ............................................................................................................. 201 9.4 Conclusion ......................................................................................................................... 204 References ................................................................................................................................... 208 xii University of Ghana http://ugspace.ug.edu.gh LIST OF TABLES Table 3. 1: Description and measurement of variables ................................................................. 83 Table 3. 2: Description and measurement of variables ................................................................. 86 Table 3. 3: Description and measurement of control variables .................................................... 88 Table 4. 1: Percentage distribution of men by socio-demographic and economic characteristics ..................................................................................................................................................... 100 Table 4. 2: Percentage distribution of men by socio-demographic and economic characteristics ..................................................................................................................................................... 104 Table 4. 3: Characteristics of participants in the qualitative study ............................................. 105 Table 5 1: Percentage distribution of men by bridewealth negotiation status and approval of wife- beating ......................................................................................................................................... 110 Table 5 2: Percentage distribution of the association between socio-demographic characteristics and men’s attitudes toward wife-beating .................................................................................... 116 Table 6. 1: Odds ratios (and Confidence Intervals) from binary logistic regression models of approval of wife-beating by bridewealth negotiation status and selected characteristics, among men in Ghana .............................................................................................................................. 127 Table 6. 2: Odds ratios (and Confidence Intervals) from binary logistic regression models of approval of wife-beating by completeness of bridewealth payment and selected characteristics, among men in Ghana .................................................................................................................. 131 Table 7. 1: Odds ratios (and Confidence Intervals) from binary logistic regression models of perpetrating physical abuse by bridewealth negotiation status and selected characteristics, among men .............................................................................................................................................. 141 Table 7. 2: Odds ratios (and Confidence Intervals) from binary logistic regression models of perpetrating physical abuse by completeness of bridewealth payment and selected characteristics, among men in Ghana .................................................................................................................. 143 xiii University of Ghana http://ugspace.ug.edu.gh LIST OF FIGURES Figure 2.1: Integrated ecological model showing the association between the different levels of the ecology and their influence of IPV ......................................................................................... 64 Figure 2.2: Conceptual framework indicating the relationship between men’s individual, situational, and cultural factors and IPV ....................................................................................... 72 Figure 5. 1: Percentage distribution of men by completeness of bridewealth payment and attitudes toward wife-beating ...................................................................................................... 112 Figure 5. 2: Percentage distribution of men by bridewealth negotiation status and having ever perpetrated physical abuse .......................................................................................................... 118 Figure 5. 3: Percentage distribution of men by completeness of bridewealth payment and having ever perpetrated physical abuse .................................................................................................. 119 Figure 8. 1: Thematic network of the reasons for/ benefits of bridewealth payment ................. 153 Figure 8. 2: Thematic network of the implications of non-payment of bridewealth .................. 160 Figure 8. 3: Thematic network of the implications of partial bridewealth payment ................... 168 Figure 8. 4: Thematic network of the implications of full bridewealth payment ....................... 173 Figure 8. 5: Thematic network representing sanctions or strategies for dealing with abuse ...... 187 xiv University of Ghana http://ugspace.ug.edu.gh LIST OF APPENDICES Appendix A: Bridewealth Payment and Normative Constraints on Women’s Lives in Ghana Study: Sample Vignette (Questionnaire)………………………………………………………231 Appendix B: List of communities and the respective sample sizes…………………………...236 Appendix C: Selected characteristics of participants for the key informant interviews………237 Appendix D: Interview guide for qualitative data collection………………………………….238 Appendix E: Setting survey weight (STATA command) for 2014 Ghana Demographic an Health Survey, Men’s Sample…………………………………………………………………………239 Appendix F: Distribution of men by marital status and completeness of bridewealth payment…………………………………………………………………………………………239 Appendix G: Percentage distribution of men by number of items they approve of wife-beating on …………………………………………………………………………………………………..240 Appendix H: Organizing themes and basic themes for reasons for/benefits of bridewealth payment…………………………………………………………………………………………241 Appendix I: Organizing themes and basic themes for implications of non-payment of bridewealth……………………………………………………………………………………...245 Appendix J: Organizing themes and basic themes for implications of partial bridewealth payment…………………………………………………………………………………………250 Appendix K: Organizing themes and basic themes for implications of full bridewealth payment…………………………………………………………………………………………253 Appendix L: Organizing themes and basic themes for sanctions/strategies for dealing with abuse……………………………………………………………………………………………259 xv University of Ghana http://ugspace.ug.edu.gh LIST OF ABBREVIATIONS AIDS Acquired Immunodeficiency Syndrome DHS Demographic and Health Survey DOVSSU Domestic Violence and Victim Support Unit GDHS Ghana Demographic and Health Survey GHS Ghana Health Service GSS Ghana Statistical Service HIV Human Immunodeficiency Virus IDS Institute of Development Studies IPV Intimate Partner Violence IRD Institute for Resource Development NMIMR Noguchi Memorial Institute for Medical Research NGO Non-Governemental Organization PTSD Post-Traumatic Stress Disorder RIPS Regional Institute for Population Studies TFR Total Fertility Rate UN United Nations WHO World Health Organization WAJU Women and Juvenile Unit xvi University of Ghana http://ugspace.ug.edu.gh CHAPTER ONE INTRODUCTION 1.1 Background to the study Physical aggression, psychological mistreatment and sexual abuse in intimate unions are increasingly acknowledged to be major public health concerns globally. Studies suggest that a male intimate partner is often the common perpetrator of abuse suffered by females in intimate unions (Carney & Barner, 2012; Walby & Allen, 2004). Male-perpetrated intimate partner violence (IPV) predisposes women to sexual, reproductive, physical and psychological harm, and constitute an infringement on their human rights (Eriksson & Mazerolle, 2015; Krug, Dahlberg, Mercy, Zwi, & Lozano, 2002; Rigby & Johnson, 2017). Specifically so for women, the number of IPV related injuries and deaths is significant (Adinkrah, 2008; Kishor & Bradley, 2012; Tjaden & Thoennes, 2000b). The effects of IPV extend beyond physical injury to very severe psychological and emotional conditions (Ellsberg, Jansen, Heise, Watts, & García-Moreno, 2008; Ozaki & Otis, 2017; Rees et al., 2017). Violence against women is generally entrenched in gendered power relationships where the differential resources of men and women often determine the extent of IPV perpetrated (Frost & Dodoo, 2009; Michau, Horn, Bank, Dutt, & Zimmerman, 2015). Some studies have shown that IPV cuts across socio-economic, demographic, and cultural groups (Krug et al., 2002). Globally, nearly 30 percent of women have suffered physical or sexual intimate partner violence (World Health Organization, 2013). However, in sub-Saharan Africa, a higher percentage, (about 37 percent) of women have suffered physical or sexual intimate partner abuse (World Health Organization, 2013). Like many sub-Saharan African countries, male-perpetrated intimate 1 University of Ghana http://ugspace.ug.edu.gh partner abuse against females is predominant in Ghana (Asante et al., 2019; Ghana Statistical Service, Ghana Health Service, & ICF-Macro, 2009; Kishor & Bradley, 2012). In 2016, about 21, 9, 36 and 27 percent of male-perpetrated intimate partner physical, sexual, emotional economic violence, respectively, were reported in the twelve months preceding the survey (Asante et al., 2019). Until recently, the issue of IPV has been considered a private affair, thus, receiving very little attention from the public, researchers and policy makers. This situation has persisted mainly due to the dependence of women on men and the cultural norms regarding marriages (Amoakohene, 2004; Bowman, 2003; Ofei-Aboagye, 1994b). Power dynamics between intimate partners have been deepened not only by economic or social advantages of men relative to women (Dodoo & Frost, 2008; Kaukinen, 2004; Panda & Agarwal, 2005; Weitzman, 2014), but also by the cultural context within which unions are arranged. According to some commentators, this can fundamentally encourage approval and actual perpetration of violence against women (Bowman, 2003; Frost & Dodoo, 2009, 2010; Horne, Dodoo, & Dodoo, 2013). Gender socialization and norms that give men more authority in marriage, and the use of violence to maintain such authority have implications for women’s well-being (Frost & Dodoo, 2009, 2010). A number of studies have shown that men’s attitude toward wife-beating is a major predictor of male-perpetrated IPV against women (Speizer, 2010; Spencer, Morgan, Bridges, Washburn-Busk, & Stith, 2017). Attitudes toward wife-beating are important because they are related to traditional norms that encourage male subordination of women by normalizing different types of violence used against women (Namy et al., 2017; Rani, Bonu, & Diop-Sidibe, 2004; Speizer, 2010), thereby, putting women at risk of male-perpetrated IPV (Amoakohene, 2004; Namy et al., 2017). When normative systems endorse male-perpetrated violence in 2 University of Ghana http://ugspace.ug.edu.gh marriage, men may be more likely to be violent towards women (Namy et al., 2017; Speizer, 2010). Over the past few decades, although age at first marriage has been increasing in sub- Saharan Africa, marriage still occurs at relatively early ages, and has been widespread (Hertrich, 2017). Therefore, the implication for IPV should be a priority. Of particular interest, and central to marriage in sub-Saharan Africa, is bridewealth payment, which transfers rights to women’s sexual, reproductive and domestic services to men (Bawah, Akweongo, Simmons, & Phillips, 1999; Dekker & Hoogeveen, 2002; Dodoo, 1998a; Dodoo & Frost, 2008; Dodoo, Horne, & Biney, 2014; Frost & Dodoo, 2010; Goody, 1973; Horne et al., 2013). Traditionally, bridewealth is exchanged to legitimize unions, maintain kinship ties and ensure marital stability (Aborampah, 1999). The practice of bridewealth payment is widespread (Anderson, 2007; Oheneba-Sakyi & Takyi, 2006), and has persisted over time (across pre-colonial and post-colonial times) (Horne et al., 2013). It is an essential aspect of the marriage contract in Africa (Frost & Dodoo, 2010; Horne et al., 2013; Rudwick & Posel, 2015; Wojcicki, Straten, & Padian, 2010). Bridewealth payment involves the exchange of cash, livestock (usually cattle, goats and other small animals etc.) or other tangibles such as land from the family of the groom to the family of the bride (Aborampah, 1999; Ansell, 2001). As part of the marriage process, the amount of bridewealth to be transferred is negotiated between the two families (Aborampah, 1999; Horne et al., 2013; Ngubane, 1987; Wojcicki et al., 2010). In a recent study, the results showed significant differences between women who stated that they were married but bridewealth had not been negotiated and those who reported that bridewealth had been negotiated (Wojcicki et al., 2010). This suggests that negotiating bridewealth between the families of the couple has significant implications for social outcomes. 3 University of Ghana http://ugspace.ug.edu.gh After bridewealth has been negotiated between the two families, the commonest situation is for the negotiated amount to be paid in full. In some societies, the negotiated amount is paid instalment over time (Hakansson, 1990; Kaufman, Wet, & Stadler, 2001). Allman & Tashjian (2000) noted that in some poor communities, it may take several years for the negotiated amount to be paid in full because bridewealth payments are substantial. A third scenario is where bridewealth has been negotiated and no amount has been paid yet, but the couple may be allowed to live together as married partners (Dodoo & Dodoo, 2017; Horne et al., 2013; Wojcicki et al., 2010). These three scenarios suggest different levels of completeness of bridewealth payment. However, it is only after the negotiated amount agreed upon by the two families has been paid in full that all rights (reproductive rights and right to domestic services) will be transferred to the man/his family (Horne et al., 2013). The authors argue that once bridewealth has been paid in full, and rights transferred, any reproductive decisions taken by the woman that defy the man’s preferences will provoke disapproval. According to Horne et al. (2013; p. 5), “bridewealth payment activates norms constraining a woman’s reproductive autonomy”. Therefore, it is possible that bridewealth payment may predict men’s attitudes about wife-beating in both the reproductive and domestic domains of women’s lives if women do not meet these obligations. Differences in social organization and marriage practices (for example, the amount of bridewealth paid) have been found to exist among different lineages and ethnic groups (Nukunya, 2003; Takyi & Dodoo, 2005), and these have implications for right to children, property, and rights of partners across the lineages and ethnic groups. When bridewealth is paid, the man has access to the woman’s household (cooking, cleaning, child-care responsibilities and housekeeping), and sexual and reproductive services, in both matrilineal and patrilineal kinship systems (Dodoo, Horne, & Biney, 2014; Frost & Dodoo, 2010; Horne et al., 2013). In matrilineal 4 University of Ghana http://ugspace.ug.edu.gh societies where bridewealth payments are less substantial (Abu, 1983; Horne et al., 2013), a woman, whether alive or dead, belongs to her lineage (Fortes, 1950; Fortes, Steel, & Ady, 1947), and the children she bears belong to her kinsmen rather than to those of their biological father (Chukwukere, 1978; Dodoo, Horne, & Dodoo, 2014; Fortes et al., 1947). In this context where bridewealth is token, the husband has fewer entitlements (Horne et al., 2013; Schneider, 1964). Conversely, in patrilineal societies bridewealth is substantial, and the children the woman bears belong to her husband’s lineage (Bawah et al., 1999; Dodoo & Frost, 2008; Dodoo, Horne, & Dodoo, 2014; Gluckman, 1950; Horne et al., 2013). Both matrilineal and patrilineal kinship systems distinguished by various ethnic affiliations are of particular interest (Fortes, 1972; Frost & Dodoo, 2010; Horne et al., 2013; Nukunya, 1992; Takyi & Mann, 2006) because the nature of marriage arrangements, completeness of bridewealth payment, and rights obtained thereof, are significantly different. These have implications for power relations and in this case male- perpetrated violence against their female intimate partners. Indeed, anthropological research has consistently indicated that bridewealth payment specifically transfers control over women’s fertility to men in patrilineal societies (Goody, 1973); recent work suggests the same is not true in matrilineal communities (Frost & Dodoo, 2010). Bridewealth payment contributes to the wider control of wives and could necessitate the use of violence to gain such control (Frost & Dodoo, 2010). In settings where the amount of bridewealth payment is very high, women could be trapped in abusive relationships because their families are unable or reluctant to return the bridewealth they received from the groom and his family (Isiugo-Abanihe, 1994a; Takyi, 2001; Takyi & Gyimah, 2007). Despite considerable reports of violence perpetrated by men in sub-Saharan Africa, the nuances with regard to men’s approval or perpetration of violence against women are not fully 5 University of Ghana http://ugspace.ug.edu.gh understood. In particular, how bridewealth payment may feed into such behaviour is not clear. For example, in sub-Saharan Africa, fertility decisions are very weighty, given the obligations and costs they impute on men and women for many years. It stands to reason, therefore, that disagreement about fertility could provoke more contention than, say, those about burning food cooked for meals, which frequently results in violence. With the substantial discordance in fertility preferences noted in sub-Saharan Africa (Bongaarts & Casterline, 2013; Dodoo, 1998b; Isiugo-Abanihe, 1994b), attribution of control over women’s fertility provided by bridewealth payment, potentially implicates bridewealth as a cause of violence. Indeed, a recent work by Dodoo & Dodoo (2017) evidenced an unexpected relationship between bridewealth payment, fertility preference and contraceptive usage. Levels of disagreement seemed to be elevated when bridewealth payments are partial compared to when no payments have been made or when it had been fully paid. Without the data to disentangle explanations, Dodoo & Dodoo (2017) speculated that this might be as a result of disagreement between couples on issues regarding the number of children they should have. It is feasible that bridewealth may be associated with marital violence, and this study examines IPV and the role the practice of bridewealth plays with regard to men’s attitudes and their actual perpetration of violence against women in intimate unions. This study was conducted in response to the scarcity of research on intimate partner violence as it examines the implications of both individual and cultural variables for intimate partner violence. 6 University of Ghana http://ugspace.ug.edu.gh 1.2 Statement of the problem Approval and justification of violence against women is a major risk factor for IPV (Adomako Ampofo, 1993; Bowman, 2003; Doku & Asante, 2015; Ozaki & Otis, 2017; Spencer et al., 2017; Takyi & Lamptey, 2016). In Adjei's (2015a) study, men indicated various reasons for perpetrating violence, stating that men had to control women, for example. This reflects an inherent perception that husbands have power over their wives. What these studies do not clarify is whether the underlying basis is simply gender (that men have power because they are men) or whether there are interceding causes like bridewealth payment that condition such IPV. If bridewealth payment increases public disapproval of women’s sole reproductive decisions (Horne et al., 2013), and it is associated with disagreement between spouses on fertility preferences (Dodoo & Dodoo, 2017), then it stands to reason that bridewealth payment may be associated with attitudes about wife-beating. Spencer et al. (2017) have shown that men who justify the use of IPV have the highest likelihood of perpetrating sexual and physical abuse. More alarming however, is the fact that both men and women support wife-beating (Alesina, Brioschi, & Ferrara, 2016; Rani et al., 2004; Uthman, Lawoko, & TaherehMorad, 2010). Where abuse against females is normalized and encouraged, male-perpetrated abuse is prevalent (Namy et al., 2017; Rimonte, 1991; Speizer, 2010; Spencer et al., 2017; Tenkorang, Owusu, Yeboah, & Bannerman, 2013). While the percentage of males who reported that wife-beating is tolerable with regard to various domestic reasons is decreasing in Ghana, possibly due to increasing levels of education, a fair number of them still believe that wife-beating is justified; 13 percent of men (GSS et al., 2015a; Ghana Statistical Service & Macro International Inc, 1999). There is a dearth of research on the practice of bridewealth payment: whether bridewealth has been negotiated or not 7 University of Ghana http://ugspace.ug.edu.gh (bridewealth negotiation status) and completeness of bridewealth payment, and approval of wife- beating (Horne et al., 2013). It must be pointed out that the rights and control that bridewealth payment bestows on men are all rooted in cultural socialization of individuals, and these are potentially related to attitudes toward violence in intimate unions. Recent studies have shown that for women, being in unions where bridewealth has been negotiated has implications for positive reproductive health outcomes (Wojcicki et al., 2010). Furthermore, research has shown that, globally, male-perpetrated intimate partner violence is prevalent. Using data from eight developing countries, Flemming et al. (2015) indicated that self-reported IPV perpetration was about 17 percent in Mexico, and 45 percent in the Democratic Republic of Congo (DRC). The estimates for IPV are higher in Africa; about 37 percent of women have been victims of male-perpetrated IPV (WHO, 2013). In Tanzania, about 13.2 percent (Mulawa et al., 2018), and 40 percent of men in Uganda (Speizer, 2010) reported that they had ever committed violence against their wives. Research has shown that men are more likely to commit such acts of abuse against intimate partners than women (Adinkrah, 2008; Caman, Kristiansson, Granath, & Sturup, 2017; Kishor & Bradley, 2012; Stöckl et al., 2013; Vatnar, Friestad, & Bjørkly, 2018). Additionally, the health and social consequences of individuals affected by IPV cannot be overstated. IPV is associated with serious depression disorders (Beydoun, Al-Sahab, Beydoun, & Tamim, 2010; Beydoun, et al., 2012; Reid et al., 2008; White & Satyen, 2015), and suicidal inclinations among both perpetrators and victims of IPV (Devries et al., 2011; Naved & Akhtar, 2008). Other mental and physical disorders have also been associated with IPV (Beydoun et al., 2017; Campbell, 2002; García-Moreno et al., 2015; Rhodes et al., 2009; Tjaden & Thoennes, 2000a; Verduin, Engelhard, Rutayisire, Stronks, & Scholte, 2013). Similarly, male-perpetrated 8 University of Ghana http://ugspace.ug.edu.gh IPV is associated with HIV infection (Durevall & Lindskog, 2015; Li et al., 2014; Rigby & Johnson, 2017). The impact of violence in intimate unions is pervasive, affecting perpetrators, victims (Durevall & Lindskog, 2015; Naved & Akhtar, 2008; Tjaden & Thoennes, 2000a), and their children (Eriksson & Mazerolle, 2015; Gage, 2016), and has economic ramifications (Max, Rice, Finkelstein, Bardwell, & Leadbetter, 2004; Peterson et al., 2018; Tjaden & Thoennes, 2000a). In a recent report on the economic cost of abuse perpetrated against women, it was estimated that women in Ghana on average, spent an estimated amount of US$12.6 million on IPV related health issues, damage to property and legal cost 12 months prior to the survey (Asante et al., 2019). These accounts suggest that the consequences of IPV should not be underestimated, and efforts that focus on exploring the root causes of this canker must be considered. For a long time, research on IPV has concentrated on individual and/or family centred approaches in developing countries to the neglect of societal, community, and cultural factors. The practice of bridewealth payment, which is an important cultural component of marriage transactions, potentially deepens gender inequalities by giving men authority over women (Dodoo & Frost, 2008; Frost & Dodoo, 2010). Traditionally, however, bridewealth is exchanged to legitimize union, ensure marital stability and maintain kinship ties (Aborampah, 1999; Ansell, 2001). Overtime, the implications of bridewealth payment have been diverse. Not examining this cultural factor potentially limits our ability to comprehend the root causes of IPV in the sub- region. Needless to say, a woman’s inability to perform her reproductive, sexual and domestic responsibilities, being duties conferred by bridewealth payment, heightens her risk of experiencing violence from her husband. Bridewealth payment has implications for other domains of a woman’s life including her experience of IPV which has little representation in the 9 University of Ghana http://ugspace.ug.edu.gh literature on IPV. Moreover, the studies on the implications of the practice of bridewealth payment on approval of IPV are scant, specifically with regard to bridewealth negotiation status and completeness of bridewealth payment. In Horne et al.’s (2013), study significant positive associations were found for approval of violence against women with full bridewealth payment status compared to those with no payment. The Horne et al. (2013) study is the only exploratory study that examined the implications of completeness of bridewealth payment and attitude toward wife-beating. In addition, there is a dearth of research on bridewealth, whether negotiated or not, and women’s well-being. In a study that examined whether bridewealth has been negotiated or not, the results suggested that women in unions where bridewealth has been negotiated were more likely to have positive reproductive outcomes (protection against HIV) than their married counterparts who stated that bridewealth has not been negotiated (Wojcicki et al., 2010). This suggests that bridewealth payment potentially proffers some protection in certain aspects of women’s lives. However, till date there is little work on the implication of the practice of bridewealth and IPV with regard to bridewealth negotiation status and completeness of bridewealth payment. To understand the association between bridewealth payment and intimate partner violence, studies have used qualitative, ethnographic and anthropological designs (Bawah et al., 1999; Dery & Diedong, 2014; Frost & Dodoo, 2010; Kaye, Mirembe, Ekstrom, Kyomuhendo, & Johnsson, 2005), and quantitative methodologies (Horne et al., 2013). Generally, there are few questions on bridewealth payment in surveys in the sub-region, (Blanc, Wolff, Gage, Ezeh, & Ssekamatte-Ssebuliba, 1996; Horne et al., 2013), although it is an integral part of marriage arrangements, and a predictor of reproductive and health related outcomes among women (Dodoo & Dodoo, 2017; Horne et al., 2013; Rees et al., 2017; Wojcicki et al., 2010). It is 10 University of Ghana http://ugspace.ug.edu.gh difficult for researchers to investigate the relationship between the practice of bridewealth and IPV using quantitative methodologies, which will allow researchers to assess the prevalence of the phenomenon under study in contemporary times, since existing survey data and other quantitative data lack measures of bridewealth payment. More so, there are limited studies that explore the strengths of mixed research method approaches in studying bridewealth payment and male-perpetrated intimate partner violence. In addition, when studies examine intimate partner violence, especially in sub-Saharan Africa, they usually focus on women’s accounts. Compared to women, there have been limited accounts from the perpetrators’ perspectives on IPV (Dunkle et al., 2006; Gass, Stein, Williams, & Seedat, 2011; Mulawa, et al., 2018; Rani et al., 2004; Speizer, 2010; Takyi & Mann, 2006; Uthman et al., 2010; Verduin et al., 2013). Also, issues concerning bridewealth payment and the implications have centred on women’s account to the neglect of the other important party involved, that is, men. But it is important if not necessary, to consider men’s perspectives, even if only because they are the primary perpetrators of IPV suffered by women. Resolving the problem for women could be futile if we do not understand why men perpetrate such violence. Against this background, the influence of bridewealth on men’s attitude and perpetration of violence were examined by the present study. Using a mixed research methods approach should offer new insights and diverging views (Ågerfalk, 2013) into the association of bridewealth payment and IPV from the perpetrator’s perspective. 11 University of Ghana http://ugspace.ug.edu.gh 1.3 Research questions of the study i. What is the association between bridewealth negotiation status and men’s attitudes toward wife-beating? ii. How is completeness of bridewealth payment related to attitudes toward wife-beating among men? iii. What is the association between bridewealth negotiation status and actual intimate partner violence perpetration by men? iv. How is completeness of bridewealth payment associated with intimate partner violence perpetration by men? v. What are participants’ perceptions of the practice of bridewealth payment and intimate partner violence? 1.4 Rationale for the study This exploratory study uses a nationally representative dataset, another population based survey, and qualitative data to examine the relationship between bridewealth payment and intimate partner violence perpetrated by men in Ghana. These datasets complement each other by exploring attitudes about wife-beating and actual perpetration of violence. Examining attitudes about wife-beating is important because attitudes speak to the norms and perceptions of individuals in a given cultural, religious, or socio-economic context concerning violence suffered by women (Horne et al., 2013; Rani et al., 2004; Uthman et al., 2010). Again, socially controlled or motivated behaviours could be associated with approval of, or punishment from close relatives, for example (Ryan, Sheldon, Kasser, & Deci, 1996). Indeed, recent research has shown 12 University of Ghana http://ugspace.ug.edu.gh that bridewealth payment reinforces norms that are associated with approval of wife-beating (Horne et al., 2013). Further, studies have shown that attitudes about wife-beating are significantly associated with actual perpetration (Kishor & Bradley, 2012; Speizer, 2010; Spencer et al., 2017). However, no individual survey dataset explores bridewealth payment, attitudes about wife-beating and actual perpetration of violence by men. Although attitudes about wife-beating and actual perpetration are not examined using the same dataset, the study gives new insights on how these could be related, and the implications for women’s wellbeing and subsequent research. The social rights and responsibilities bridewealth gives to married partners are well recognized and differentiated. With time, studies have shown both negative and positive associations of bridewealth payment with various social and reproductive outcomes. It will be inadequate to examine all married couples when studying violence without examining their bridewealth payment statuses. Very few recent studies have shown that bridewealth weakens women’s reproductive autonomy by reinforcing norms regarding reproductive decisions (Dodoo, Horne, & Biney, 2014; Horne, et al., 2013), and is associated with actual behaviour (Dodoo & Dodoo 2017;Wojcicki et al., 2010). This challenge of not examining bridewealth has persisted because surveys have scarcely explored bridewealth measures, especially among men. In Wojcicki et al. (2011), the results showed that for all married women, those who stated that bridewealth had been negotiated were more likely to report higher ages at first sex and a higher likelihood of ever using male condoms compared to their counterparts who stated that bridewealth has not been negotiated. Furthermore, examining completeness of bridewealth payment and its association with IPV adds to the scant body of knowledge on bridewealth payment and intimate partner violence, specifically, men’s attitudes and perpetration of IPV. 13 University of Ghana http://ugspace.ug.edu.gh Although the ideal expectation is for bridewealth to be fully paid, payment is not always made in full immediately. This has implications for varying power dynamics in unions for different bridewealth payment statuses. In the sub-region, studies that examine how power relations differ in each of the categories listed above, and how these could provoke male-perpetrated violence in intimate relationships are lacking (Horne et al., 2013). To date, the paucity of data has meant that only few studies (Horne et al., 2013) have been able to use quantitative data to explore the implications of completeness of bridewealth payment (fully paid, partially paid and not paid), and these have mostly been in relation to women. Addressing the issue of violence from only the victim’s perspective, without understanding why the perpetrator commits such actions could totally derail our efforts to reduce violence against women. Studies should also examine the perpetrators, and factors related to their behaviour. Therefore, this study examines completeness of bridewealth payment and its association with men’s attitudes toward wife-beating and actual perpetration of physical abuse by men. As researchers and policy makers strive to bridge the socio-economic gap between males and females by increasing access to education, and active labour force participation, especially for women, social and cultural factors that further improve or worsen the relative position of women should also be examined. Some researchers have argued that IPV should be viewed as a state of the woman’s relative power in the household and in the society (Anderson, 1997; Jewkes, Levin, & Penn-Kekana, 2002; Kaukinen, 2004; Panda & Agarwal, 2005; Weitzman, 2014). For example, a study by Ellsberg et al. (2015) has shown that empowering women economically without regard for their cultural and social contexts may not protect them from IPV. Thus, in developing countries, individual and family centred approaches to studying IPV have been criticized mostly because they discount the importance of culture and poverty which 14 University of Ghana http://ugspace.ug.edu.gh are important components of society (Adjei, 2018; Krug et al., 2002). The current study examines cultural factors that fuel male-perpetrated violence against women, due to expectations of marriage, especially when bridewealth is involved. For decades, researchers have suggested that bridewealth payment may fuel male- perpetrated violence against women in the sub-region (Frost & Dodoo, 2010; Horne et al., 2013; Ofei-Aboagye, 1994b; Takyi, 2001). According to Takyi (2001), writing on divorce, women may be unable to leave abusive marriages because their families are incapable of paying back bridewealth they received at the time of marriage. Other studies show that bridewealth payment could be associated with male-perpetrated abuse because the payment could indicate that the man had bought the woman (Ansell, 2001; Ansell et al., 2017; Ofei-Aboagye, 1994b). Takyi & Mann (2006) and Mann & Takyi (2009) assert that the role of culture, demonstrated through bridewealth payment, may contribute to the approval of wife-beating observed in their studies. Aborampah (1999) argues that bridewealth payment establishes marital stability and ensures that couples maintain healthy behaviour. It is worth noting that although these previous studies try to explain the cultural reasons for male-perpetrated intimate partner violence, there has been little effort to empirically test the above mixed assertions from the husband’s perspective. Furthermore, this study contributes methodologically by using both qualitative and quantitative datasets. The quantitative datasets allow us to examine the prevalence of violence in terms of men’s attitudes and actual perpetration in relation to bridewealth payment. This would go a long way to establish key intervention points, and possibly lower the occurrence of male- perpetrated violence suffered by women. The qualitative dataset assesses participants’ perceptions of the implications of the practice of bridewealth payment and intimate partner violence. The mixed research methods approach adopted by this study adds depth to the study by 15 University of Ghana http://ugspace.ug.edu.gh examining the practices of bridewealth payment and male-perpetrated intimate partner violence in Ghana. 1.5 Objectives of the study The general objective of the study is to examine the relationship among bridewealth payment, men’s attitudes toward wife-beating, and male-perpetrated intimate partner violence. The specific objectives are: i. To examine the association between bridewealth negotiation status and men’s attitudes toward wife-beating. ii. To examine the association between completeness of bridewealth payment and men’s attitudes toward wife-beating. iii. To examine the association between bridewealth negotiation status and intimate partner violence perpetration by men. iv. To examine the association between completeness of bridewealth payment and intimate partner violence perpetration by men. v. To explore participants’ perceptions on the practice of bridewealth payment and intimate partner violence perpetration. 1.6 Organization of the study Chapter one, the introductory chapter of this study, is made up of the background to the study, the statement of the problem, research questions, rationale for the study, and the study 16 University of Ghana http://ugspace.ug.edu.gh objectives. In chapter two, relevant literature on intimate partner violence both globally and in sub-Saharan Africa and Ghana, is reviewed. Relevant theories on intimate partner violence and a conceptual framework for the study are further presented in this chapter. The focus of chapter three is to show the study population, methodology and study limitations. Chapter four provides descriptive statistics for the study samples and presents men’s attitudes toward wife-beating, perpetration of abuse, bridewealth payment and other socio-cultural and demographic characteristics of the study population. The bivariate associations between bridewealth payment, respondents’ characteristics and intimate partner violence are presented in chapter five. The sixth chapter provides multivariate results on whether or not bridewealth has been negotiated, completeness of bridewealth payment and men’s attitudes toward wife-beating, while chapter seven examines bridewealth payment and actual perpetration of intimate partner violence by men. In chapter eight, participants’ perceptions about bridewealth payment and the use or non- use of violence by men against their partners in intimate unions are explored. The last chapter of this study, chapter nine, outlines the summary, recommendations and conclusion. 17 University of Ghana http://ugspace.ug.edu.gh CHAPTER TWO LITERATURE REVIEW 2.1 Introduction Intimate partner violence (IPV) is broadly referred to as “physical violence, sexual violence, stalking and psychological aggression (including coercive tactics) by a current or former intimate partner” (Breiding, et al., 2015, p. 11). Also, preventing individuals from contacting family or friends and access to information, and strictly observing their activities infringe on their fundamental human rights (Krug et al., 2002; World Health Organization, 2017). Intimate partner violence occurs in various shapes and forms, and these have negative consequences on women’s wellbeing (Campbell, 2002; Durevall & Lindskog, 2015; Naved & Akhtar, 2008). Generally, a “person with whom one has a close personal relationship with, and can be characterized by emotional connectedness, regular contact, usually an on-going physical contact and/or sexual behaviour, identify as a couple, and familiarity and knowledge about each other’s lives” is referred to as an intimate partner (IP) (Breiding et al., 2015, p. 11). An intimate partner is either one’s spouse, dating partner (boyfriend or girlfriend), or on- going sexual partner, and may or may not share residence (Breiding et al., 2015). A male intimate partner is the most often cited perpetrator of abuse suffered by females (Carney & Barner, 2012; Dobash & Dobash, 1979; Walby & Allen, 2004). In their review that draws on 204 studies, Carney and Barner (2012) found that in about 80 percent of the cases, perpetrators of violence against women were commonly former or current male intimate partners. Some studies suggest that violence in intimate relationships is gender based, and that men have a higher likelihood of perpetrating violence than women (Dobash, Dobash, Wilson, & Daly, 1992; Kishor 18 University of Ghana http://ugspace.ug.edu.gh & Bradley, 2012). Also, men are less likely to be afflicted with injuries and deaths due to violence than women are (Caman, Kristiansson, Granath, & Sturup, 2017; Stöckl et al., 2013; WHO, 2013). This indicates that male-perpetrated violence is prevalent and has dire consequences for women’s wellbeing. In this study, intimate partner violence refers to men’s attitudes toward wife-beating and men’s self-reported accounts of perpetration of violence against a current female partner. 2.2 Extant measures of intimate partner violence Overtime, various measures have been used to identify violent acts that occur between intimate partners (Krug et al., 2002; Straus et al., 1996; Straus, 1979; WHO, 2013). These include physical, sexual, various forms of psychological violence and economic violence (Fawole, 2008; Sanders, 2015; Sedziafa, Tenkorang, Owusu, & Sano, 2017; Usta, Makarem, & Habib, 2013). Although physical and sexual IPV have been studied rather extensively (García- Moreno, et al., 2006; Horne et al., 2013; Kimuna, Tenkorang, & Djamba, 2018; Krug et al., 2002; McCloskey, Williams, & Larsen, 2005; Oduro, Deere, & Catanzarite, 2015; Takyi & Lamptey, 2016; Tenkorang et al., 2013), some studies have shown that psychological IPV may have major deplorable effects on health and general well-being, perhaps even more so physical violence (Começanha, Basto-Pereira, & Maia, 2017; Lawrence, Orengo-Aguayo, Langer, & Brock, 2012). Some studies use the Conflict Tactics Scale or the revised version of the scale (Straus et. al., 1996; Straus, 1979) in examining the various types of IPV (Neil & Scovelle, 2018; Speizer, 2010). Usually, Demographic and Health Surveys have used modified versions of the Conflict Tactics Scale to examine intimate partner violence. These studies have mostly examined physical 19 University of Ghana http://ugspace.ug.edu.gh and/or sexual intimate partner violence (Johnson & Das, 2009; Panda & Agarwal, 2005; Speizer, 2010; Tenkorang et al., 2013; Weitzman, 2014), and emotional violence (Adjah & Agbemafle, 2016; Asiedu, 2016; Fidan & Bui, 2016; Kaukinen, 2004). The sections below describe the types of male-perpetrated IPV identified in the literature. 2.2.1 Physical intimate partner violence The use of intentional physical force with the aim of inflicting injury, disability or death in intimate unions can be classified as physical violence (Breiding et al., 2015). Such acts include but are not restricted to “scratching, pushing, shoving, throwing, grabbing, biting, choking, shaking, pulling hair, slapping, punching, hitting, burning (with fire or acid or other substances) and using weapons (gun, knife, or other object) on an intimate partner” (Breiding et al., 2015, p. 11). Physical violence includes pressuring third parties to perform these physical acts (Breiding et al., 2015). Some studies limit measures of violence to a subset of these acts or to a single act of physical abuse (Fleming et al., 2015; García-Moreno et al., 2006; Johnson & Das, 2009; Neil & Scovelle, 2018; Panda & Agarwal, 2005; Tenkorang et al., 2013; Wagman et al., 2016; Weitzman, 2014). Fleming et al. (2015, p.5) in their Multi-Country Study conceptualized lifetime male-perpetrated physical IPV to include “slapping, pushing, hitting with a fist, kicking/dragging/beating/choking/burning, or threatening or using a weapon” against a female partner. Research has revealed that male-perpetrated physical intimate partner violence has serious health (Frye et al., 2011; Mitchell, Wight, Van Heerden, & Rochat, 2016; Neil & Scovelle, 2018), economic and social consequences on perpetrators themselves, their victims and the society at large (Lawrence et al., 2012). 20 University of Ghana http://ugspace.ug.edu.gh 2.2.2 Sexual intimate partner violence As with physical violence, sexual violence can include various forms of sexual tactics used against a victim’s will. These acts can be actual or attempted, but generally involve the victim not willingly giving consent or being unable to give consent (Breiding et al., 2015). Sexual intimate partner violence consists of contact or penetration of the vagina, mouth, or anus with the penis, finger, or other objects. It includes unsolicited sexual contact, such as touching a person deliberately even when they do not approve. Making uninvited sexual comments, knowingly exposing sexually explicit media, filming, and distributing naked or sexual videos of a partner, and threatening to spread such videos or rumours can be considered as non-contact unwelcomed sexual abuse (Breiding et al., 2015). The use of various forms of intimidation to obtain sexual favours can be considered as sexual abuse (Breiding et al., 2015). Some studies have also shown that using alcohol or other illegal drugs, for example, in these cases is of particular essence as substance abuse is associated with sexual violence (Dunkle et al., 2006; Moore, Easton, & McMahon, 2011). 2.2.3 Psychological/emotional intimate partner violence The “use of verbal and non-verbal forms of communication with the intent to harm an intimate partner mentally or emotionally and/or exert control over her” may be considered as psychologically initiated violence (Breiding et al., 2015, p. 15). Psychological/emotional abuse does not have to involve the use of physical force or acts that can be described as such. These acts may not necessarily be considered as violent because they may mimic everyday confrontations and behaviour. Psychological violence encompasses various forms of controlling 21 University of Ghana http://ugspace.ug.edu.gh behaviours, and compared with other types of abuse, psychological violence has the highest rate in terms of perpetration and victimization (Começanha et al., 2017). Studying male-perpetrated psychological violence in intimate relationships has important ramifications. First, psychological violence is usually a build up to sexual and/or physical violence (Breiding et al., 2015; Murphy & O’leary, 1989). Psychological violence co-occurs with other violent behaviours, such that psychologically violent men also have a higher likelihood of being physically abusive. Usually, the cause of physical abuse is an expression of on-going psychologically violent acts. Secondly, studies have shown that male-perpetrated psychological violence against women may have serious and long lasting effects as physical violence (Começanha et al., 2017; Lawrence et al., 2012). For instance, although all forms of IPV were predictors of post-traumatic stress disorder (PTSD), experiencing psychological violence was the strongest predictor of PTSD among females (Pico-Alfonso, 2005). Male-perpetrated IPV is rooted in the men’s desire to control and over-power their partners (Começanha et al., 2017; Dobash & Dobash, 1992; Whitaker, 2013). 2.2.4 Economic intimate partner violence Men’s control over women’s economic and financial resources has been identified as a way through which abusive male partners ensure women remain dependent, and in subordinate positions to them (Sanders, 2015; Sedziafa et al., 2017). Some studies have shown that economic abuse against women manifests itself in different ways in that male partners use various strategies to keep their female partners economically dependent on them. A case in point is controlling the amount of, and access to financial resources such that women find it difficult to 22 University of Ghana http://ugspace.ug.edu.gh provide for their basic necessities such as food, for themselves, and their dependents (Sanders, 2015; Sedziafa et al., 2017). Economic dependency may also prevent women from leaving abusive relationships for fear of losing support for themselves and their children (Adjei, 2015b; Heise, Ellsberg, & Gottemoeller, 1999; Kim & Gray, 2008; Sanders, 2015). Thus, women who rely on their partners for economic resources are at risk of suffering other types of male- perpetrated intimate partner violence (Cantalupo, Martin, Pak, & Shin, 2006; Goode, 1971; Oduro et al., 2015). Economic abuse usually co-exists with physical, sexual and/or emotional abuse (Sanders, 2015; Sedziafa, Tenkorang, & Owusu, 2016). For example, women have to perform sexual favours for their male partners even when they do not want to in order to receive some money from them (Sanders, 2015; Sedziafa et al., 2017; Usta et al., 2013). This has implications for sexual exploitation, and its accompanying risks (Fawole, 2008). Another way through which economic abuse manifests itself is through the destruction of women’s property, limiting or preventing their access to paid employment (Sanders, 2015; Sedziafa et al., 2017), or other social resources such as education through which women may become gainfully employed (Fawole, 2008; Sanders, 2015). When women’s economic status and their ability to obtain or maintain paid work are significantly restricted by their partners, it increases female dependence, and places women at risk of experiencing violence (Fawole, 2008; Sanders, 2015; Sedziafa et al., 2017). Women are unable to report economic abuse because they may lack vivid or tangible evidence to support such claims (Sedziafa et al., 2017), and also to ensure peace and keep the family intact, women endure economically abusive unions (Usta et al., 2013). Poverty and patriarchy place women at greater likelihood of male-perpetrated violence (Sanders, 2015; Sedziafa et al., 2017). Thus, in resource scarce settings, poor living conditions 23 University of Ghana http://ugspace.ug.edu.gh could exacerbate IPV perpetration. In such cases, economic violence could be prevalent and could lead to sexual exploitation of women, placing them at risk of HIV (Fawole, 2008). In addition to the danger of HIV/AIDS infection, limiting women’s access to resources may put them at risk of maternal health problems such as unsafe abortions and other pregnancy related complications (Heise et al., 1999). Patriarchal norms and control over resources mainly by men, give women limited access to essential resources, increasing female poverty and further deepening their dependence on men. Women’s active labour force participation and access to high income may challenge male power and authority (Weitzman, 2014). Some male partners may use various tactics to prevent women from maintaining such economic independence (Sanders, 2015; Sedziafa et al., 2017). Resource theory suggests that when men have lesser resources compared to their female partners, the odds of using violence as an alternate resource to maintain dominance in relationships increase (Goode, 1971). Economic violence is also evident when male partners totally relegate their responsibility of providing economic support for their wives and children even in cases when they have the economic power (Sedziafa et al., 2017). 2.3 Prevalence of intimate partner violence Male-perpetrated IPV against females is a social and health menace, and a major public health concern. Global estimates suggest that approximately 30 percent of females have experienced male-perpetrated sexual and/or physical IPV (WHO, 2013). Using a systematic review which synthesizes data on intimate and non-intimate partner violence between 2008 and 2011, WHO (2013) estimates suggest that prevalence of physical and/or sexual IPV was lower in 24 University of Ghana http://ugspace.ug.edu.gh high income WHO regions (23%) such as Finland, Japan and Canada. Conversely, in developing WHO regions, for example, Africa, South-East Asia, and Eastern Mediterranean, IPV prevalence was about 37% with the exception of WHO regions in Western Pacific, Europe and the Americas where physical and/or sexual IPV prevalence was at about 24.5%, 25.4% and 29.8% respectively. Further, relying on data from eight low and middle income countries, Fleming et al. (2015) indicated that self-reported lifetime physical intimate violence perpetrated ranged from 17 percent in Mexico to about 45 percent in Democratic Republic of Congo (DRC). These estimates indicate that although the prevalence of male-perpetrated IPV varies across regions, a lot of men all over the world are at risk of perpetrating violence against their intimate female partners. Also, in estimating physical and/or sexual IPV from fifteen sites using ten countries, Garcia-Moreno et al. (2006) documented lifetime male-perpetrated IPV at about 71 percent in Ethiopia (Butajira) compared to 15 percent in Japan (Yokohama). In Bangladesh, the estimates for male-perpetrated intimate partner violence are higher. Men’s lifetime self-reported intimate partner violence perpetration was estimated at 74 percent, and 12 months before the study, about 37 percent of these men stated that they had ever been violent toward their female partners (Johnson & Das, 2009). In another cross-sectional study in north-eastern US, the results suggest that nearly one in three has ever perpetrated physical or sexual violence against their partners who are females (Decker et al., 2009). Studies on global estimates of IPV have mostly concentrated on physical and/or sexual violence (Garcia-Moreno et al., 2006; WHO, 2013). Rarely are other forms of abuse such as economic or emotional/psychological abuse by men against women in intimate unions examined at global levels, more so from men’s perspectives. Not unexpectedly, some studies have shown 25 University of Ghana http://ugspace.ug.edu.gh that these forms of violence (economic or emotional/psychological) co-exist with other forms of violence (Sanders, 2015; Sedziafa et al., 2017). 2.4 Prevalence of IPV in sub-Saharan Africa According to the WHO (2013) report, the documented rate of intimate partner violence perpetration is higher in sub-Saharan Africa, and variations exist across countries in the sub- region (WHO, 2013). It is estimated that women’s reports of lifetime male-perpetrated intimate partner violence in the sub-region is about 37 percent (WHO, 2013). In a more recent study that used Demographic and Health Surveys across 18 African countries, researchers found that the perpetration of violence was rife in the sub-region (Alesina et al., 2016). Nearly a third of women stated being victims of physical or sexual violence since they turned age 15, and twelve months preceding the survey, about 22 percent of women indicated that they had suffered either sexual or physical intimate partner violence (Alesina et al., 2016). Relying on nationally representative data from Uganda, it was estimated that two out of five (40%) men had ever been violent toward their female partners (Speizer, 2010). Using data from Dar es Salaam, Tanzania for example, Mulawa et al. (2018) found that among men 15 years and over, 27 percent reported that they have ever perpetrated at least sexual, physical or psychological IPV against their intimate partners twelve months preceding the survey. Between 2014 and 2015, Kapiga et al. (2017) estimated that lifetime experience of male- perpetrated IPV was high in Mwanza city of Tanzania. Eighty-two percent of females suffered several types of controlling actions from intimate partners. Additionally, 61 percent of females 26 University of Ghana http://ugspace.ug.edu.gh suffered sexual and/or physical IPV, whereas 32 percent were victims of different forms of economic abuse (Kapiga et al., 2017). About 18.7 and 5.1 percent of females have suffered male-perpetrated intimate physical and sexual violence respectively in Ghana using the 2008 Demographic and Health Survey (Tenkorang et al., 2013). In 2016, Fidan & Bui (2016) examined IPV in Zimbabwe among a sample of 4411 women in their reproductive ages relying on the Demographic and Health Survey, 2010. For the three measures of IPV studied, women reported more male-perpetrated physical violence (27.1%) than sexual (14%) or emotional violence (24.4%). In a recent study examining 5206 males and 3106 females aged 18-49 years in four countries, men’s estimated IPV perpetration ranged from 32.5 percent to 80 percent (Jewkes et al., 2017). These accounts suggest the high prevalence of violence, as it cuts across various groups in the sub-region. Male-perpetrated intimate partner violence is predominant in urban (Mulawa, et al., 2018; Mulawa, et al., 2018; Zembe, Townsend, Thorson, Silberschmidt, & Ekstrom, 2015) and rural areas (Dunkle et al., 2006; Hatcher, Colvin, Ndlovu, & Dworkin, 2014; Hatcher et al., 2013; Speizer, 2010), among the rich and the poor (Fleming et al., 2015; Johnson & Das, 2009), among matrilineal and patrilineal groups (Sedziafa et al., 2016), and among those employed (Alesina et al., 2016) and the unemployed (Babcock & DePrince, 2013). Male-perpetrated intimate partner violence cuts across basically every sphere of a woman’s life. In her work with women who had experienced violence, they recounted various reasons why their male spouses were violent (Ofei-Aboagye, 1994b). The following extracts illustrate the range of reasons women reported were the cause of male-perpetrated violence. 27 University of Ghana http://ugspace.ug.edu.gh “He would beat me whenever he was drunk…When I confronted him with evidence of his sleeping with another woman…When I asked him for chop money…When I refused to have sex ....There was no particular reason…Because I was pregnant…He accused me of sleeping with another man…Because my cooking was not to his taste… He said I was rude in public ....I had insulted his mother…I spent too much money…He links giving me maintenance money with whether or not I have sex with him. He's violent with the children…He even attempts to sleep with our daughters…We should know people for a long time before we marry them” (Ofei- Aboagye, 1994b, p. 5). The above accounts show that the reasons why men perpetrate abuse are numerous. A man may abuse his intimate partner because she failed or purported to have failed to execute her domestic duties properly or was unable to behave correctly as a wife in public spheres. He may also use violence if his partner stands up to him and questions his behaviour or fidelity. Socialization norms and gender roles are at work here as they prescribe what a wife/woman is expected to do and not do (Adomako Ampofo, 2001; Blanc et al., 1996; Horne et al., 2013). Sanctions from male partners may apply for deviating from such gender norms/roles (Adjei, 2016; Horne et al., 2013). This is to say women cannot escape from male-perpetrated violence because it is everywhere, and can be perpetrated for a variety of reasons, especially, when norms support the use of violence. 2.5 Some predictors of male-perpetrated intimate partner violence Over the years, several factors have been identified as risk factors to male-perpetrated intimate partner violence. These factors consist of, but are not restricted to attitudes toward 28 University of Ghana http://ugspace.ug.edu.gh violence, male dominance, gender differences in the risk of IPV, poverty/economic dependence of women on men, social support, educational attainment, observing violence as a child, alcohol consumption, religion and the cultural context in which unions are arranged. 2.5.1 Attitudes toward intimate partner violence Societal approval of violence against women is widespread (Amoakohene, 2004; Blanc et al., 1996; Haj-Yahia, 2003), and it is a major concern because it precipitates the use of violence against women (Nwabunike & Tenkorang, 2017; Takyi & Lamptey, 2016). In their study involving seven sub-Saharan African nations, Rani et al. (2004) found that approval of wife- beating was common in all countries, and among males and females. In another study in the sub- region where data from 28 countries for women and 27 for men were used, similar results emerged (Alesina et al., 2016). Approval of violence is usually related to the contravening gender roles prescribed as appropriate behaviour for men and women. For instance, wife-beating is a way of punishing or disciplining a wife for not doing what her husband expects of her (Ofei- Aboagye, 1994b). This shows that gender socialization overtly teaches men and women to behave within certain cultural lines and spaces. Thus, where men perceive women to have ‘transgressed’, they are at higher risk of perpetrating IPV as a way of teaching women to conform to societal norms of good behaviour. According to Nukunya (2003), a wife may be beaten for not cooking on time or for committing adultery. Where violence is the norm (Adjei, 2015b, 2016; Ofei-Aboagye, 1994b), patriarchal traditional settings and socialization further strengthen such norms (Bawah et al., 1999; Haj-Yahia, 2003). These potentially exacerbate the use of violence when females fail to meet these standards (Bawah et al., 1999; Dery & Diedong, 29 University of Ghana http://ugspace.ug.edu.gh 2014). Traditional norms which encourage subordination of women may further exacerbate women’s experience of IPV (Amoakohene, 2004; Namy et al., 2017). In her study on 50 women who suffered some form of IPV in Ghana, Ofei-Aboagye (1994) noted that women considered violence to be private as it was more or less considered as a norm. Similarly, in a study among Libyan migrants, men for example, considered violence against women as a less serious menance, refering to it as a private family issue (El Abani & Pourmehdi, 2018). Women may be socialized to accept male-perpetrated partner violence or even incite wife-beating because it is seen as a show of affection and love (Luke, 2003; Odimegwu, 2001), or as a form of discipline if she behaves as a child or does not perform domestic duties (Namy et al., 2017). Social and cultural approval, and justification of wife-beating (Luke, 2003) are related with IPV (Heise & Kotsadam, 2015). For instance, women who approve of wife-beating were significantly at higher odds of experiencing male-perpetrated violence (Ozaki & Otis, 2017; Tenkorang et al., 2013). Using both male and female students reports, Spencer, et al. (2017) found that men who approve of IPV reported a higher likelihood of committing or being sufferers of physical violence, and perpetrators of sexual violence. Also, females who support the use of IPV against women had higher odds of suffering sexual and psychological IPV. In other regions, and in sub-Saharan Africa, prior studies have reported significant relationships between approval of wife-beating by men and intimate female partner abuse (Fleming et al., 2015; Johnson & Das, 2009; Speizer, 2010). Various factors are related to men’s perception about wife-beating. For example, younger men (Dalal, Lee, & Gifford, 2012; Rani & Bonu, 2009; Takyi & Mann, 2006), rural men (Takyi & Mann, 2006), those with lower levels of education (Haj-Yahia, 2003; Uthman, Lawoko, & Moradi, 2009), men with low occupational status (Dalal et al., 2012), low household wealth 30 University of Ghana http://ugspace.ug.edu.gh (Rani & Bonu, 2009; Takyi & Mann, 2006; Uthman et al., 2009), those in polygynous unions (Rani et al., 2004), and males who adhere to patriarchal norms (Haj-Yahia, 2003; Haj-Yahia & Uysal, 2008) are more likely to approve of beating wives. In addition to the above, Takyi and Mann (2006) found that men who make sole household decisions, and those who have higher parity had a significantly higher likelihood of justifying wife abuse. In this study, men’s attitudes toward wife-beating is operationalized as their approval or acceptability of violence against a woman if she does not perform certain defined duties that traditionally married women are expected to perform. The acceptance of violence is therefore framed within conjugal relationships. This conceptualization of attitudes towards wife-beating has been employed by various studies as indicated (Dalal et al., 2012; Doku & Asante, 2015; Rani et al., 2004; Speizer, 2010), and has been used as standard in exploring acceptability of violence against women in Demographic and Health Surveys. In the Ghana Demographic and Health Survey, men’s attitudes toward wife-beating is examined by exploring their acceptability of wife-beating if a woman, goes out without telling her husband, neglects the children, argues with her husband, refuses to have sex with her husband, or burns the food. While these items explore some measures of attitudes toward wife- beating, they fail to capture extensively men’s justification of violence against women in intimate unions in different contexts, especially with regard to reproduction and contraceptive issues. This is a challenge because in pro-natalist settings, reproduction and issues regarding limitation of birth could be integral points of justification of violence against women, and should be considered in such large surveys. Dalal et al. (2012) using DHS data from selected countries in South Asia, had a measure of attitude toward wife-beating if a woman disrespects the respondent’s parents and senior relatives and not if she burns the food. The item that examines 31 University of Ghana http://ugspace.ug.edu.gh approval of wife-beating if a woman disrespects her in-laws is vital in this setting because of cultural framings regarding in-law relationships in these settings. This suggests that items that measure attitudes toward wife-beating should also consider context. 2.5.2 Male dominance and intimate partner violence Strict adherence to hegemonic masculinity norms has significant consequences for IPV (Hatcher et al., 2013; Santana, et al., 2006). The use of violence against women is related to control and dominance of men over women by reinforcing traditional masculine norms which encompass identity, respect and authority (Adjei, 2016; Courtenay, 2000; El Abani & Pourmehdi, 2018; Heise & Kotsadam, 2015; Jewkes, 2002; Kiss et al., 2012; Klugman, 2017; Namy et al., 2017; Whitaker, 2013). Hegemonic masculinities do not only reflect individual predispositions but actually show societal influence in constructing these identities (Connell, 1987). Reed (1972) argues that in certain situations, individuals who perpetrate violence do not need to be violent; however, they become violent in order to escape penalties and disgrace for not being violent. This clearly shows that a culture that accepts and perceives violence as appropriate for conflict resolution would incite its members to use it even in instances when members think violence is not necessary. Using qualitative data from two areas in Accra and Kumasi, Ghana, Adjei (2016) found that perpetrators beat their wives because their inability to use such acts will mean they are not being ‘manly’ as society expects them to be. Thus, societal expectations drive some men to use violence even when they actually perceive the use of violence as unnecessary. To project a sense of “manliness”, and to maintain control, some men use violence. In another study by Adjei, (2015a), the reason men gave for perpetrating violence 32 University of Ghana http://ugspace.ug.edu.gh was that, men needed to teach women lessons for going ‘wrong’. This reflects an inherent perception that the husband has the power to put a wife in order. In a quote from a focus group discussion, one male respondent made this statement: “I think it is not unusual for a man to slap or abuse his wife to teach her lessons if she goes wrong [. . .]” (Urban perpetrator, FGD) (Adjei, 2015a, p. 9). This statement reflects male authority and societal endorsement of violence used against married women. Gendered socialization further enforces such acts of violence (Frost & Dodoo, 2010). Dominance or control over women is associated with IPV (Haj-Yahia & Uysal, 2008; Horne et al., 2013; Ozaki & Otis, 2017; Whitaker, 2013). In a study that compares university students of European origin to those of Asian origin, the three measures of male dominance examined were authority, restrictiveness, and disparagement (Ozaki & Otis, 2017). The authors argue that male dominance is at the heart of patriarchal societies, and their results showed that dominance over women contributed to severe forms of abuse committed against women by way of controlling and belittling them (Ozaki & Otis, 2017). Also, the use of violence is normalized in ways that make women look weak, and project men as strong and virile. Some previous research suggest that in some contexts, women allow male partners to beat them to show that they are ‘real men’ because “you don’t want to be associated with boys” (Zembe et al., 2015, p. 10). These accounts echo patriarchal and masculine beliefs about wife-beating (Haj-Yahia, 1998, 2003) that are endorsed by both men and women (Rani et al., 2004; Uthman et al., 2010) but put women at risk of greater harm (Fleming et al., 2015; Speizer, 2010; Spencer et al., 2017). Men’s adherence to masculine standards is correlated 33 University of Ghana http://ugspace.ug.edu.gh with unsafe alcohol uses, a major risk factor to intimate partner violence (Jewkes & Morrell, 2010; Morojele et al., 2006; Peralta, Tuttle, & Steele, 2010). 2.5.3 Gender differences and the risk of intimate partner violence Are there gender differences in the experience of abuse? It has been stated that experience of violence cuts across gender, socio-economic, demographic, cultural and religious groups. However, some studies argue that the female gender in itself is a major risk factor for IPV (Campbell, Abrahams and Martin 2008). In both developing and developed countries, women are at a higher risk of male-perpetrated IPV than men are of female-perpetrated IPV (Dobash et al., 1992; Kishor & Bradley, 2012; Tjaden & Thoennes, 2000b). This argument has been advanced by feminist advocates and researchers who suggest that male dominance over women is correlated with their risk of being violent (Dobash & Dobash, 1979; Dobash et al., 1992; Johnson, 2005). Although some studies have argued that women are as likely as men to be violent in unions, usually studies in the United States (Straus, 2008; Walklate, 2013), empirical evidence (specifically in developing regions) show that men are at significantly higher odds of perpetrating IPV, inflicting injuries and deaths on intimate partners than women (Adinkrah, 2008; Caman et al., 2017; Kishor & Bradley, 2012; Stöckl et al., 2013). 1Were et al. (2013) in their study in 7 countries in southern and eastern sub-Saharan Africa, revealed that men were less likely to be victims of female-perpetrated IPV than women. 1 Between 2004 November and 2007 May, the Partners in Prevention HSV/HIV Transmission Study was conducted with 3408 HIV serodiscordant partners from 7 sub-Saharan African countries in Kenya, Rwanda, Tanzania and Uganda, all in East Africa and Botswana, South Africa, Zambia all in Southern Africa. 34 University of Ghana http://ugspace.ug.edu.gh Again, Umubyeyi, Mogren, Ntaganira, & Krantz (2014) explored the occurrence of some forms of intimate partner violence among young individuals aged 20 to 35 years and found similar results. Using a cross-sectional population-based design in Southern Rwanda, and for the 2three types of violence estimated, women were consistently at higher risk of reporting that such acts of violence were perpetrated against them than men were. In another study which concentrated on the experiences of married males and females in two African countries; Ghana in West Africa and in East Africa, Uganda, there was no evidence to support gender symmetry in relation to spousal violence (Kishor & Bradley, 2012). Relying on Demographic and Health Surveys from these two countries, the study concluded that although women were also perpetrators of spousal violence, they were consistently and significantly at higher risk of experiencing all types of spousal violence (sexual, physical and psychological) investigated than their male counterparts. For instance, in Ghana, 18.5 percent and 9.9 percent of women and men respectively reported lifetime experience of intimate partner violence. In Uganda, 19.1 percent of males compared to 47.2 percent of females had ever experienced any form of physical spousal violence. Another important feature of IPV is the gender disparity in relation to frequency, severity (injuries), and deaths due to IPV. The frequency of experiencing violence for example, is higher among women than among men (Kishor & Bradley, 2012; Umubyeyi et al., 2014). Kishor and Bradley (2012) found that the greatest gender disparities were reported for acts such as being “kicked, dragged, or being beaten up on purpose”. While in Ghana females were nearly 7 times 2 78 women, representing 18.8%, reported physical abuse in the last year, compared to 18 men (4.3%). For sexual violence, while 17.4% of women reported (n = 71) such acts against them, 1.5percent (n = 6) of men did. For psychological abuse, the number for males and females were 21.4% (n = 92) and 7.3% (n = 32) respectively. 35 University of Ghana http://ugspace.ug.edu.gh as likely as males to have experienced such acts, females in Uganda were close to 10 times as likely. Also, about 42 percent of females who suffered IPV obtained several forms of injuries (WHO, 2013). These accounts suggest that IPV perpetrated by men has dire effects on women’s well-being. It is estimated that more male-perpetrated intimate partner homicides against women are recorded than those by women against men (Caman et al., 2017; Stöckl et al., 2013). Globally, about thirty eight percent of all homicides are carried out by men against female partners compared to 6 percent committed by women against male intimate partners (Stöckl et al., 2013). Using data from European Statistical Database on Lethal Violence which reported on 1725 solved homicides in Sweden between 1990 and 2013, although there has been significant decrease in intimate partner related homicides, the results indicated that of all homicides, about 57 percent (399) of women were murdered by male partners compared to 7 percent of men by female partners (78). In South Africa for example, of the estimated 3,797 female homicides obtained from mortuary reports in 1999, about half (50.3%) were due to male-perpetrated IPV (Abrahams et al., 2009). Further, several studies suggest that although males are less likely to support beating wives compared to females (Adomako Ampofo, 1993; Uthman et al., 2009, 2010) the odds of abusing women are high for men (Kishor & Bradley, 2012; Speizer, 2010). In Ghana, for instance, though the fraction of men and women who indicated that IPV is acceptable under certain domestic conditions (for example, “ if a wife goes out without telling her husband, burns the food, refuses to have sex with her husband, neglects the children or argues with her husband”) is decreasing, a fair number of people still agree that beating wives is tolerable with regard to the above scenarios; 28 percent of females and 13 percent of males in 2014 (GSS et al., 36 University of Ghana http://ugspace.ug.edu.gh 2015; GSS et al., 2004). Alesina et al. (2016) suggest that fewer men reported justification of violence because of interviewer desirability, and this may be similar for women too. 2.5.4 Poverty/economic dependence on men and intimate partner violence Globally, studies have investigated the possible relationship between poverty and IPV and the results have generally been mixed due to differences in measurement of concepts, sample size and study areas (Abramsky et al., 2011; Bowman, 2003; Sedziafa et al., 2016; Tenkorang et al., 2013). Some studies suggest that resource distribution, and poverty levels within households are associated with IPV (Panda & Agarwal, 2005; Tenkorang et al., 2013). According to the resource theory, when men have more resources than their spouses, such resources are used to command the needed compliance, in that non-compliance may lead to withdrawal of economic support (Goode, 1971). The marital dependency model predicts that women with low income or those in poor households are at a higher risk of IPV (Kimuna et al., 2018; Oduro et al., 2015; Takyi & Lamptey, 2016; Tenkorang et al., 2013; Tolman & Raphael, 2000). This is because economic dependence on male spouses makes them vulnerable to abuse, and may prevent them from leaving unions that are abusive for fear of resources being withdrawn from them and their children (Adjei, 2017; Sanders, 2015; Sedziafa et al., 2017). Further, when women have the opportunity to obtain extra wealth (resources) compared to their male spouses, their risk of experiencing violence especially when their male spouses adhere to traditional gender beliefs of masculinity and control of power in households, is higher (Atkinson, Greenstein, & Lang, 2005; Goode, 1971; Kaukinen, 2004; Weitzman, 2014). The relative resource theory suggests that status reversals possibly increases the odds of violence 37 University of Ghana http://ugspace.ug.edu.gh against women since their status challenge traditional gender scripts (Atkinson et al., 2005; Kaukinen, 2004; Oduro et al., 2015). Earlier studies similarly show that the use of violence may be an alternative power to men who do not have it in the economic sense (Atkinson et al., 2005; Goode, 1971; Hunnicutt, 2009; Panda & Agarwal, 2005; Weitzman, 2014). In their study for example, Kiss et al. (2012) found that although a woman’s neighbourhood has little impact on her experience of IPV, women who fall within the middle range of the socio-economic scale have higher odds of reporting IPV experience than their counterparts who fall in the wealthier ranges. Thus, unequal economic differences may put women at risk of male-perpetrated abuse. In Africa, intimate partner violence is high due to resource inequalities both at the household and community levels (Cools & Kotsadam, 2017). Gender inequality is a major cause of violence against women (Jewkes et al., 2002; Michau et al., 2015). Women’s dependence on men is primarily one of the drivers of violence against women (Bowman, 2003; Cantalupo et al., 2006;Dodoo & Frost, 2008; Oduro et al., 2015). This is illustrated in the WHO report on violence and health which shows that in developing countries where poverty is a major challenge, the implication for violence against women cannot be overemphasised (Krug et al., 2002). Women who live in households within the lowest wealth quintile have higher odds of male-perpetrated abuse than those in the richer wealth quintiles (Kimuna et al., 2018; Oyediran & Isiugo-Abanihe, 2005). Panda & Agarwal (2005) found that women who own properties like houses or land are at a lower risk of male-perpetrated IPV (both long term and short term physical or psychological violence) compared to women who have no property. This emphasizes the protective effect women’s access to wealth and assets has against violence. The authors argue that women with such property have higher bargaining power in unions, and are less tolerant of the use of 38 University of Ghana http://ugspace.ug.edu.gh violence. Some evidence indicates that for women in Ecuador and Ghana, their share of couple’s wealth reduces the odds of experiencing male-perpetrated physical and emotional violence respectively (Oduro et al., 2015). Thus, financially relying on male partners means that women act at their male partners’ behest, placing them at risk of violence. For instance, Zembe et al. (2015) showed that about 46 percent of women were expected to obey their partners’ orders when they were given gifts or money. Although the above studies have documented several significant relationships between women’s access to wealth and male-perpetrated IPV, some studies have not found any statistically significant relationship between men’s self-reported economic resources and actual abuse against their female partners (Fleming et al., 2015; Johnson & Das, 2009; Speizer, 2010). 2.5.5 Social support and intimate partner violence Studies on social support have highlighted the importance of social network, support and control on intimate partner violence control and prevention. Social support has been linked with reducing or preventing victimization, preventing repeat abuse, encouraging victims of abuse to leave abusive unions, and it has also been associated with better health outcomes of victims of intimate partner violence (Coker et al., 2002; Machisa, Christofides, & Jewkes, 2018; Mburia- Mwalili, Clements-Nolle, Lee, Shadley, & Yang, 2010;Mitchell & Hodson, 1983; Zapor, Wolford-Clevenger, & Johnson, 2018). Social support (time, money, gifts, provision of services (child care, household chores), advice, and physical presence) (Heaney & Israel, 2008) from significant individuals in a couple’s life potentially reduces the frequency of intimate partner abuse (Mburia-Mwalili et al., 2010). 39 University of Ghana http://ugspace.ug.edu.gh Social support and networks are also linked with better mental health outcomes among individuals with experience of IPV (Beeble et al., 2009; Costa & Gomes, 2018; Dillon, Hussain, Loxton, & Khan, 2016; Dutton, 2009). The presence or intervention of other relatives in the household may prevent or reduce the incidence and severity of abuse as these family members may advocate on behalf of the victim (Emery, Wu, Kim, Pyun, & Chin, 2017). Receiving assistance in terms of child support, transportation, housework and advice from friends and family members reduces the chances of ever experiencing abuse (Lanier & Maume, 2009). These forms of informal social support and control are essential for preventing violence against women. Further, studies have shown that adequate social support potentially enables or encourages victims of abuse to leave abusive relationships (Barnett, 2001; Ondicho, 2000). 2.5.6 Education and intimate partner violence The importance of education on several social outcomes cannot be overemphasised (Al Riyami, Afifi, & Mabry, 2004; Jejeebhoy, 1995). Education changes people’s attitudes and perceptions, and it is associated with more liberal beliefs (Thornton & Camburn, 1987). In relation to intimate partner violence, many pathways have been identified. While in India men having higher levels of formal teaching were less probable to perpetrate abuse, in DRC, the opposite was true (Fleming et al., 2015). In their multi-country study, the results showed that in DRC, men with post-secondary education were 2.02 times as likely to have ever perpetrated abuse compared with men with no level of schooling or only primary education. In India, however, men with post-secondary school training were 0.47 times as likely as those with no schooling or primary school training to have perpetrated abuse against a female partner. Former 40 University of Ghana http://ugspace.ug.edu.gh studies show that men with increasing levels of schooling have lower odds of perpetrating abuse (Johnson & Das, 2009). In their study in Tanzania, Mulawa et al. (2018) found no significant relationship between men’s level of education and the likelihood of perpetrating abuse. Some studies suggest that for women, education is protective against male-perpetrated IPV (Abramsky et al., 2011; Adjah & Agbemafle, 2016; Alangea et al., 2018; Alesina et al., 2016; Dery & Diedong, 2014; Kapiga et al., 2017; McCloskey et al., 2005; Weitzman, 2014), and may offer women better economic options, thus, buffering them from violence (McCloskey et al., 2005; Ofei-Aboagye, 1994a). Although education is protective, some studies have shown that women’s education may not always proffer such protection because male dominance and norms may rip women off this advantage (Amoakohene, 2004; Bankole & Singh, 1998; DeRose, Dodoo, & Patil, 2002; Dodoo, Horne, & Biney, 2014). Takyi & Lamptey (2016) found that in Ghana, females with secondary school education had a higher likelihood of emotional and sexual violence experience compared with their counterparts who have no education or higher education. These studies show that educated females may also be at risk of suffering certain types of violence than others. Some researchers have argued that in addition to a woman’s schooling, her partner’s level of training/education is a key predictor of abuse committed against her. According to the relative resource thesis, absolute educational status alone does not predict women’s experience of violence (Atkinson et al., 2005; Kaukinen, 2004). Women’s education relative to their partners’ education has significant implications for violence such that women with higher levels of schooling than their partners’ may challenge traditional gender norms (Jewkes, 2002), and this could increase their risk of experiencing violence. Using data from Zimbabwe, Fidan & Bui (2016) found that women whose partners have higher levels of schooling than they have were 41 University of Ghana http://ugspace.ug.edu.gh more probable to experience physical and emotional abuse when likened to those with the same level of training as their partners. Thus, relative status equality tends to be more protective compared with the other two extremes where the husband has more education or the wife has more education (Abramsky et al., 2011; Oduro et al., 2015). Kaukein (2004), for instance, found that women with the same or higher training/education relative to their significant others were at greater harm of suffering emotional abuse. Thus, feminist and resource theorists allude to the fact that access to these resources is symbolic as they relate with gender identities, marital roles and conflict (Kaukinen, 2004; Oduro et al., 2015). Education is also associated with attitudes about beating wives such that, with lower levels of formal education, men are generally more likely to support beating wives (Haj-Yahia, 2003; Rani & Bonu, 2009; Takyi & Mann, 2006; Uthman et al., 2009, 2010; Wang, 2016). The variations in the direction and strength of association between schooling and intimate partner violence could be attributed to differences in measurement, study locations, sample sizes and reference period for violence, that is, lifetime perpetration or perpetration in the last twelve months. 2.5.7 Observing violence as a child and intimate partner violence Within the IPV literature, researchers find strong and consistent links between witnessing violence as a child, and suffering and/or committing such acts later on in life (Heise, 1998; Peralta et al., 2010; World Health Organization, 2017). Witnessing violence as a child is a deplorable experience for many people because it influences the young person’s subsequent behaviour. Children who see violence between their caregivers are more likely to suffer from 42 University of Ghana http://ugspace.ug.edu.gh troubling behavioural challenges as adults (Eriksson & Mazerolle, 2015; Gage, 2016). Witnessing marital violence between caregivers is positively associated with perpetrating or experiencing IPV (Abramsky et al., 2011; Bucheli & Rossi, 2017; Heise, 1998; Panda & Agarwal, 2005; Tenkorang et al., 2013; Yount et al., 2014), and also, childhood victimization is associated with post-traumatic stress disorder (Becker, Stuewig, & McCloskey, 2010). Observing violence as a child is embedded in the general socialization of children, and it also reflects the norms and perceptions of individuals in a given context (Bandura, 1973; Namy et al., 2017). For example, in their mixed method study, respondents who had been recruited as perpetrators of IPV recounted that they conceptualized violence they witnessed in their families as children as normal (Peralta et al., 2010). Thus, as adults they found themselves engaging in such acts against their intimate partners. They also found that in addition to growing up in violent households, community exposure to violence may place individuals at risk of normalizing such acts (Peralta et al., 2010). Observing violence between one’s parents or caregivers as a child works in two ways. First, men who see abuse as children are more likely to commit IPV (Delsol & Margolin, 2004; Gass et al., 2011; Namy et al., 2017; Yount et al., 2014). Gass et al. (2011) established that men who observed parental abuse were more than twice as likely as those who did not witness parental violence to commit IPV in the future. Heise (1998) argues that using violence in adult intimate unions is a response to habits learnt as a child from parents. Secondly, for women, observing inter-parental violence places them at risk of experiencing IPV in their current unions than their counterparts who did not live in violent homes because these women normalize such acts (Namy et al., 2017; Panda & Agarwal, 2005; Weitzman, 2014). 43 University of Ghana http://ugspace.ug.edu.gh Individuals who observe their fathers or male guardians beat or abuse their mothers are more likely to rationalize such acts as normal and legitimate, and may have little ability to avoid such acts (Dery & Diedong, 2014; Gass et al., 2011; Speizer, 2010). This goes back to the issue of violence being normalized and widely accepted in society, thus, this menace may receive less attention or individuals will do little to avoid them. In their qualitative study on domestic violence in the Upper West Region of Ghana, the quote below from a male respondent illustrates clearly how family history of violence shapes an individual’s perception of wife-beating (Dery & Diedong, 2014). "When I was young and anytime my father beats my mother or had any harsh argument with her, she usually unleashed her anger on us the children. So we grew up to understand that the best way to unleash your anger is to off-load it on your wife or children" (Dery & Diedong, 2014, p. 240). The above extract throws some light on social learning theorists’ argument on observing inter- parental violence as a child, since these violent acts are also learnt by observing (Bandura, 1973, 1977). Thus, children who observe such acts of abuse in their homes are likely to develop or learn aggressive and violent forms of exchanges, and may practise such behaviours in their future encounters, and against their intimate partners. 2.5.8 Alcohol use and intimate partner violence Another major risk factor to IPV perpetration is alcohol use (Abramsky et al., 2011; Adebowale, 2018; Adjah & Agbemafle, 2016; Alangea et al., 2018; Cunradi, Caetano, & Schafer, 2002; de Bruijn & de Graaf, 2016; Dery & Diedong, 2014; Fawole, Aderonmu, & 44 University of Ghana http://ugspace.ug.edu.gh Fawole, 2005; Hatcher et al., 2014; Heise, 1998; Jewkes et al., 2002; Kaukinen, 2004; Kimuna et al., 2018; Mulawa, et al., 2018; Takyi & Lamptey, 2016; Tenkorang et al., 2013; Weitzman, 2014). Alcohol use is positively related with IPV, and women whose partners use alcohol are at a higher risk of being subjected to IPV compared to those whose partners do not consume alcohol. In Ghana, for example, women with spouses who use alcohol are two times more likely to suffer abuse (Adjah & Agbemafle, 2016; Tenkorang & Owusu, 2018). Studies in Uganda (Namy et al., 2017; Zablotska et al., 2009), India (Wagman et al., 2016; Weitzman, 2014), South Africa (Morojele et al., 2006) and Australia (Wilson, Graham, & Taft, 2017) have all recorded similar findings. Alcohol use inhibits one’s judgment, and instigates the use of abuse in intimate interactions (Hatcher et al., 2013; Kiss et al., 2012; Morojele et al., 2006). Scholars have debated that the role of alcohol in perpetrating violence is socially learnt (McDonald, 1994) and men use alcohol in order to settle scores, something they would not have been able to do if they had not been under the influence of alcohol. The use of alcohol may provide an excuse for being violent because perpetrators think that they would not be held responsible for their actions due to the fact that they were ‘drunk’ (Dery & Diedong, 2014; Jewkes, 2002). Indeed, victims may even blame violence perpetrated against them on the alcohol their partners consumed, thus “making an unacceptable behaviour acceptable” (Javaid, 2015, p. 83). Consequently, the use of alcohol significantly prevents men from understanding the gravity of violence they perpetrate. Using 12 semi-structured interviews with professionals who work with or have worked with perpetrators and/or victims of IPV, the author found that in addition to alcohol contributing to violence perpetration, it enables perpetrators to give excuses for their behaviour (Javaid, 2015). Although the underlying cause or reason for violence may not often be alcohol, its use 45 University of Ghana http://ugspace.ug.edu.gh stimulates acts of violence. For the above reason, some researchers have questioned the role of alcohol as a cause of wife-beating since men who beat their wives may use alcohol to escape being punished or excuse their misbehaviours (Heise, 1998). Some research have also revealed that the connection between alcohol consumption and IPV is weak (Gil-González, et al., 2006). Nonetheless, as a cause or a facilitator, alcohol use may exacerbate violence perpetrated against women in intimate unions. Furthermore, men are more likely to abuse their female companions who consume alcohol (Adjah & Agbemafle, 2016; Dery & Diedong, 2014; Graham, Bernards, Wilsnack, & Gmel, 2011; Javaid, 2015; Kiss et al., 2012). According to Adjah & Agbemafle (2016), alcohol use among women increased the likelihood of male-perpetrated IPV by 71 percent. In another study, respondents indicated that a woman who consumes excessive alcohol acts like a juvenile and “she needs to be disciplined like a child” (Dery & Diedong, 2014, p. 238). Hence, being intoxicated is a morally unaccepeted behaviour for women, and such acts require sanctioning through the use of violence. In Rakai, Uganda for instance, females who used alcohol before sex had a higher likelihood of being subjected to both physical and sexual IPV (Zablotska et al., 2009), and this puts women at risk of HIV infection. These results show that alcohol use and violence have even greater health consequences. 2.5.9 Religion and intimate partner violence The influence of religion on male-perpetrated violence has been diverse. Religion has been cited as a predictor of violence used against women as some religious doctrines highlight male dominance and female subordination, thus, justifying male-perpetrated violence against 46 University of Ghana http://ugspace.ug.edu.gh women (Deshotels, Forsyth, Earwood, New, & Fulmer, 2019). In their study on refusing wife- beating in Ethiopia, Gurmu & Endale (2017) found that women in rural areas who adhere to Protestant religious beliefs and women who identified with Islam Religion in urban areas were less likely to reject beating wives. In Ghana, for example, Takyi and Lamptey (2016) found that compared to women who identified with Traditional African Religion and those who were not affiliated to any religious groups, Christians and Moslems were significantly less probable to suffer sexual violence. Adherence to traditional norms could heighten the risk of male- perpetrated violence and its acceptance (Amoakohene, 2004; Haj-Yahia & Uysal, 2008; Kaukinen, 2004; Weitzman, 2014). It is important that studies also investigate religious involvement rather than just religious affiliation as studies have shown that religious involvement specifically frequency of attendance and consistency, could also provide some social support needed for intimate partner violence control. For example, for Hispanic males, and African American men and women, religious service attendance shields one from abuse (Ellison, et al., 2007). According to Ellison & Anderson (2001), men who attend religious programs every week were less likely to be violent towards their spouses. Social support and cohesion developed when one participates in religious programs potentially reduce their risk of being violent. Regular religious attendance and participation regulate alcohol intake (Meyers, Brown, Grant, & Hasin, 2017) which is a major risk factor of male-perpetrated violence (Adebowale, 2018; Alangea et al., 2018; Mulawa, et al., 2018; Takyi & Lamptey, 2016). Measuring religiosity as intensity to religious commitment and religious self-regulation, Renzetti, Dewall, Messer, & Pond (2017) found that higher scores on the religiosity scales were positive predictors of male- perpetrated IPV. Other studies have found no significant association between religious affiliation 47 University of Ghana http://ugspace.ug.edu.gh and attitudes towards wife-beating (Okenwa-Emegwa, Lawoko, & Jansson, 2016; Takyi & Mann, 2006). The observed differences between religion, religiosity and intimate partner violence perpetration could be attributed to variations in measuring these concepts, the sample size used, the region of the study and the type of sample. It is important that in exploring concepts that relate to beliefs, culture and ideologies, context-appropriate measures should be prioritized. 2.6 Bridewealth payment in sub-Saharan Africa Bridewealth payment is a customary practice through which marriages are formed and recognized in sub-Saharan Africa (Dodoo, Horne, & Dodoo, 2014; Frost & Dodoo, 2010; Horne et al., 2013; Yarbrough, 2017). Bridewealth payment compensates the bride’s kinsmen for the loss of her productive and reproductive services, such that the man’s family gains these rights (Dodoo & Frost, 2008; Horne et al., 2013). Traditionally, the contract of marriage is primarily between families, and only secondarily between the couple (Fortes et al., 1947; Nukunya, 1969). Marriages are arranged to maintain kinship ties (Dekker & Hoogeveen, 2002), keep family wealth, and expand lineages (Fortes et al., 1947; Goody, 1973; Ngubane, 1987; Nukunya, 1969). Through the payment of bridewealth, couples’ rights, roles and responsibilities are defined (Blanc et al., 1996; Horne et al., 2013). Bridewealth payment transfers rights to women’s sexual, reproductive and domestic services to men (Bawah et al., 1999; Dekker & Hoogeveen, 2002; Dodoo, 1998a; Dodoo & Frost, 2008; Dodoo, Horne, & Biney, 2014; Frost & Dodoo, 2010; Goody, 1973; Horne et al., 2013). 48 University of Ghana http://ugspace.ug.edu.gh It is the exchange of cash, livestock (usually cattle, goats and other small animals etc.) or other tangibles such as land from the family of the groom to the family of the bride (Ansell, 2001). Other negotiated goods that a man’s family gives to a woman’s family include cloth, drinks, beads and fabric (Aborampah, 1999). Studies have shown that while in the past cattle and other local products dominated the exchange process, in recent times, the practice has mainly been monetized (Kaufman et al., 2001; Ngubane, 1987). Bridewealth payment is mostly practised in eastern and western Africa (Blanc et al., 1996; Dodoo, Horne, & Biney, 2014; Dodoo & Dodoo, 2017; Frost & Dodoo, 2010; Horne et al., 2013; Kaye et al., 2005; Kimuna et al., 2018), in southern Africa (Ansell, 2001; Ansell et al., 2017; Dekker & Hoogeveen, 2002; Klomegah, 2008; Rudwick & Posel, 2015), and also in Northern Africa (Al-nouri, 1980). The practice is common among the young (Ansell, 2001) and old (Dodoo, Horne, & Dodoo, 2014), among rural and urban dwellers (Blanc et al., 1996; Frost & Dodoo, 2010; Horne et al., 2013; Yarbrough, 2017), and among matrilineal and patrilineal lineages (Fortes, 1972; Frost & Dodoo, 2010; Horne et al., 2013; Nukunya, 1992; Tait, 1961; Takyi, 2001; Takyi & Gyimah, 2007). Bridewealth payment is predominant in societies where women have prominent role in agriculture, and where polygyny and divorce rates are high (Anderson, 2007; Boserup, 1985; Goody, 1973). In sub-Saharan Africa, marriage is a process (Kaye et al., 2005; Yarbrough, 2017); and as part of the marriage process, bridewealth is negotiated between the groom’s kinsmen and the bride’s kinsmen. The negotiated amount is then paid to the bride’s kinsmen (Horne et al., 2013; Mizinga, 2000; Ngubane, 1987; Wojcicki et al., 2010). Bridewealth may be substantial and it may take the groom and his family several years to make full payment (Allman & Tashjian, 2000). Hence, payments would be made in instalments making marriage a process rather than a 49 University of Ghana http://ugspace.ug.edu.gh one-time event (Allman & Tashjian, 2000; Kaufman et al., 2001; Lesthaeghe, 1989). In some cases, the very last instalment is made when the woman bears her first child, a sign that she is fertile (Lesthaeghe, 1989). Negotiated bridewealth is either fully paid, partially paid or not paid (Blanc et al., 1996; Dodoo, Horne, & Biney, 2014; Horne et al., 2013; Wojcicki et al., 2010). However, it is only after the negotiated amount agreed upon by the two families has been paid in full that all rights (reproductive and domestic services) will be transferred to the man/his family. Bridewealth payment does not only recognize unions but also determines rights and responsibilities couples enjoy thereafter (Horne et al., 2013). These rights and responsibilities are also differentiated by completeness of payment made (Dodoo, Horne, & Biney, 2014;Dodoo & Dodoo, 2017; Horne et al., 2013). When no bridewealth or little bridewealth is paid, for example, husbands have fewer entitlements (Schneider, 1964). Men obtain fewer rights to women’s domestic labour/services but obtain sexual and housekeeping rights where no bridewealth or very little bridewealth payment is made (Schneider, 1964). Conversely, where the bridewealth payment is substantial, men obtain higher rights to women’s labour and the children the woman bears belong to the man and his lineage (Schneider, 1964). Schneider (1964) suggests that in Africa, marriage rights to women are not sold but leased through the bridewealth payment. Schneider describes it as this; marriage payment is more about loaning these services than the outright sale of these rights. Thus, men have rights to the women’s reproductive and productive services including food and housekeeping services for the exchange of marriage payment (Ansell, 2001; Frost & Dodoo, 2010; Horne et al., 2013; Schneider, 1964). In some cases, the number of livestock (bridewealth) the man’s kinsmen gives to the woman’s kinsmen increases as the number of children she bears increases (Dekker & Hoogeveen, 2002; Schneider, 1964). This translates to marriage payments being made in 50 University of Ghana http://ugspace.ug.edu.gh instalments (Frost & Dodoo, 2010; Hakansson, 1990; Horne et al., 2013; Schneider, 1964). Thus, at the beginning of the marriage, a said amount of livestock will be negotiated, and will be paid as long as the marriage holds (Schneider, 1964). In some settings, lineage members contribute towards bridewealth in order for men to marry their first wives (Schneider, 1964). Only the father or guardian to the woman keeps bridewealth received. Contrarily, in some other settings, when lineage members contribute to the bridewealth, they also share in the bridewealth received although a majority of the wealth received goes to the woman’s father (Schneider, 1964), or is maintained and used by the woman’s father’s household (Ngubane, 1987). The significance of marriage, bridewealth payment, and the importance of having children cannot be overemphasized. Bridewealth payment safeguards control over reproduction (Dodoo & Tempenis, 2002; Dodoo & Dodoo, 2017; Horne et al., 2013), especially in places where children are crucial resources (Hunter, 2016). Through the bridewealth payment, older generations gain considerable control over younger ones. Particularly so for women, bridewealth payment gives men, mainly fathers and husbands, control over women; control over their sexuality/reproduction and their labour (Ansell, 2001; Dodoo, Horne, & Biney, 2014;Dodoo & Dodoo, 2017; Horne et al., 2013; Isiugo-Abanihe, 1994b). Thus, norms regarding acceptable behaviour with respect to bridewealth payment have consequences for the way individuals in marriages behave (Horne et al., 2013). As such, going contrary to such norms has implications for women’s reproductive autonomy, and intimate partner violence (Bawah et al., 1999; Dery & Diedong, 2014; Dodoo & Dodoo, 2017; Horne et al., 2013). Significant differences in social organization, marriage formation, and practices exist across various lineages. In some patrilineal societies, for example, payment received does not go to the bride but to her male kin, and most likely to her brother who in turn uses that to pay for the 51 University of Ghana http://ugspace.ug.edu.gh bridewealth of his wife (Goody, 1973; Ngubane, 1987). Bridewealth payment is substantial, and creates a demanding obligation in some cases for the entire village and the woman’s family to ensure that the couple stay married (Fallers, 1957; Goody, 1973; Horne et al., 2013; Takyi, 2001). Thus, in order to dissolve the marriage for example, the woman’s kinsmen would have to return bridewealth received. As Horne et al. (2013, p.11) describe it, “if the exchange breaks down ripple effects extend beyond the two actors to other group members.” Payment potentially supports norms and may provoke disapproval when individuals deviate from norms as enabled by bridewealth payment. In both kinship systems, the woman’s kinsmen are obligated to pay back the bridewealth if the marriage is dissolved (Schneider, 1964). 2.6.1 Bridewealth payment and intimate partner violence (IPV) In sub-Saharan Africa where marriages are near universal, and early (Boserup, 1985; Hertrich, 2017), it is vital to investigate the implications of bridewealth relative to men’s attitudes and perpetration of IPV. Anthropological and ethnographic studies have tried to explore the implications of bridewealth payment on females’ well-being (Bawah et al., 1999; Dery & Diedong, 2014; Frost & Dodoo, 2010; Kaye et al., 2005). Nonetheless, in contemporary times, due to the paucity of data, we have very little knowledge on the association between bridewealth payment on one very important global health problem – male-perpetrated IPV (Horne et al., 2013). Bridewealth potentially affects women’s welfare by stripping women of control over their bodies, and could be associated with the spread of HIV/AIDS and domestic violence (Anderson, 2007; Wendo, 2004). Bridewealth payment compensates the woman’s kinsmen for her reproductive and productive services, and this may indicate to some husbands that their wives 52 University of Ghana http://ugspace.ug.edu.gh have been purchased, and are now the husbands’ properties (Ansell, 2001; Ansell et al., 2017). This has serious implications because individuals may use violence to control women or demand of women to behave in a certain manner because of bridewealth payment. This is echoed in studies by Frost and Dodoo (2010), Dery and Diedong (2014), Bawah et al. (1999) and Ansell (2001), where participants viewed bridewealth payments as giving men authority and control over women and in some cases justify the use of violence to gain such control. Some qualitative studies have indicated that bridewealth legitimizes men’s use of violence against women because some men may view bridewealth as payment for the women’s domestic services, respect, sexual and reproductive rights (Bawah et al., 1999; Blanc et al., 1996; Dery & Diedong, 2014; Phorano, Nthomang, & Ntseane, 2005; Yarbrough, 2017). Completeness of bridewealth payment also has implications for women’s well-being. In situations where bridewealth is substantial (Allman & Tashjian, 2000), or in order to maintain lasting affiliations (Dekker & Hoogeveen, 2002; Yarbrough, 2017), bridewealth payments may be made in instalments. Some studies argue that the number of children a woman has (Dekker & Hoogeveen, 2002; Kaufman et al., 2001), a woman’s location, her status (level of education attained and other factors), the status of prospective suitors may all influence the amount of bridewealth negotiated (Horne et al., 2013; Kaufman et al., 2001). As such, bridewealth payment varies and it is difficult to come up with a uniform payment size because of some of these reasons (Horne et al., 2013). Thus, in estimating bridewealth payment, they examined the completeness of payment of bridewealth as no payment, partial or full payment. In some settings, for example, no payments are made until a woman has her first child (Lesthaeghe, 1989). Dekker and Hogeveen (2002) indicate that in Zimbabwe, for example, payments are made in instalments to maintain affinal relations, and provide economic security for both families. Bridewealth 53 University of Ghana http://ugspace.ug.edu.gh payments being made in instalments or over the course of the marriage are important and rife (Dekker & Hoogeveen, 2002; Horne et al., 2013; Yarbrough, 2017). If completeness of payment translates or is related to higher normative restrictions over women’s reproductive autonomy (Horne et al., 2013), then we expect to observe stronger association between completeness of payment, and attitudes and perpetration of wife-beating, more so from men if women do not meet their expected marriage obligations. That is, men’s approval of wife-beating increases with completeness of payment such that if women do not perform their reproductive and domestic duties when bridewealth has been fully paid, for instance, men will be more likely to be violent, and other individuals will be more likely to support men’s use of abuse. Although this phenomenon has implications for violence perpetration, to the best of my knowledge there is a paucity of literature on completeness of bridewealth payment with respect to intimate partner violence (Horne et al., 2013), specifically so, from men’s perspectives. Where bridewealth is substantial, the woman’s kinsmen have a higher obligation of ensuring the success of the marriage to prevent the return of bridewealth paid to them even if the woman is in an abusive union (Isiugo-Abanihe, 1994a; Takyi, 2001; Wendo, 2004). This implies that in such situations, women may have to remain in abusive relationships if they are unable to repay the bridewealth. For fear of losing contact with children or withdrawal of economic support for themselves and their children, women are more likely to remain with their abusive partners (Adjei, 2015b, 2018). In Frost and Dodoo (2010), the data suggest that, young males anticipated greater authority and obedience/subordination from their future wives due to bridewealth payment. Thus, in their opinion, if their future wives are not respectful or obedient, some young respondents indicated that they would have the right to ‘discipline’ them because 54 University of Ghana http://ugspace.ug.edu.gh bridewealth has been paid. The reasons for partner abuse are rooted in the general socialization of children, where women are expected to be submissive and respectful (Amoakohene, 2004; Dodoo & Frost, 2008; Ofei-Aboagye, 1994b), and bridewealth further strengthens such authority and control over women (Ansell et al., 2017; Frost & Dodoo, 2010; Horne et al., 2013; Yarbrough, 2017). Over time, bridewealth payment in sub-Saharan Africa may have been interpreted to mean women have been purchased. Some studies have, however, suggested that traditionally this was not the purpose of bridewealth payment (Aborampah, 1999). Bridewealth payment traditionally indicates the value of women. Payments were made to maintain kinship ties and to ensure marital stability. Bridewealth payment also provided some economic support for the bride in her new household (Ngubane, 1987). In this study, bridewealth negotiation status refers to whether bridewealth has been negotiated between the man’s family and the woman’s family. According to the 2014 Ghana Demographic and Health Survey Interviewer’s Manual, “bridewealth negotiation means that either the families of both the man and the woman or the couple themselves played a role in arriving at the bridewealth” (Ghana Statistical Service, 2014, p. 99). 2.7 Legislation on domestic violence in Ghana Male-perpetrated intimate partner violence is a social menace that affects many women in Ghana. A recent study by Asante et al. (2019) showed that about 20% of men have ever perpetrated some form of violence against their intimate female partners. In order to reduce or eliminate various forms of violence against women, several governmental interventions through the use of legislation have been enacted in Ghana. These governmental interventions are geared 55 University of Ghana http://ugspace.ug.edu.gh towards protecting the fundamental human rights of the vulnerable, and also achieving international goals of reducing violence against women. Among these are the 1992 Constitution of Ghana, the Criminal Code, 1960 (Act 29) and Domestic Violence Act, 2007 (Act 732). The Criminal Code, 1960 (Act 29) and its subsequent amendments address offenses that violate an individual’s fundamental human rights. Some of the related offenses stipulated in the Criminal Code, 1960 (Act 29) include those associated with homicides, sexual exploitations, damage to property, causing physical harm to an individual, and such similar offenses. It is important to note that if such offenses occur within intimate unions they should be handled as criminal offenses under the Criminal Code, 1960 (Act 29) (Government of Ghana, 1960). Apart from the Criminal Code, 1960 (Act 29), other judicial interventions over the years have been legislated to address issues relating to violence perpetrated against women in Ghana. In October 1998, a specialised unit of the Ghana Police Service named the Women and Juvenile Unit (WAJU) was mandated to handle issues regarding to domestic violence against women and children in Ghana, including violence perpetrated against women by intimate male partners. This special unit was formed due to increasing reported cases of domestic violence related crimes in the country (Amoakohene, 2004). Relying on various aspects of the Criminal Code, 1960 (Act 29), WAJU was mandated to investigate and prosecute violent crimes such as rape, defilement, assault/battery, threatening, and causing harm against women in Ghana. In 2005, the name of this special unit was changed from the Women and Juvenile Unit (WAJU) to Domestic Violence and Victim Support Unit (DOVSSU) in response to arguments that domestic violence is experienced by both females and males. Similarly, DOVVSU is mandated to investigate and prosecute crimes or offenses perpetrated against women, children and men. Further, the unit is expected to refer victims of 56 University of Ghana http://ugspace.ug.edu.gh violence to counselling and medical services, and provide victims with social support by relying on the services of trained clinical psychologists and social workers. Thus, DOVVSU works with various supporting departments such as the Department of Social Work. Although DOVVSU is tasked with investigating and prosecuting domestic violence in Ghana, it is rather difficult to obtain up to date data and information on the operations of the unit. Also, a valid and all- inclusive database on domestic violence in Ghana over the years is lacking. This challenge limits researchers’ ability to track and organize comprehensive data on domestic violence, more so, intimate partner violence in Ghana. In response to addressing issues relating to domestic violence in Ghana through collective activism of various governmental and non-governmental stakeholders, the Domestic Violence Act 732 was passed in 2007. The Domestic Violence Act details various forms of violent acts that constitute threat or harm to a person in a past or current relationship including intimate relationships under the Criminal Code, 1960 (Act 29). These include various forms of physical, sexual, economic, emotional, verbal and psychological abuse. Further, the Act stipulates that harassment including sexual harassment and intimidation qualify as domestic violence. Additionally, the Act states that any behaviour or conduct that: (i) harms or may harm another person, (ii) endangers the safety, health or well-being of another person, (iii) undermines another person's privacy, integrity or security, or (iv) detracts or is likely to detract from another person's dignity and worth as a human being may constitute domestic violence (Government of Ghana, 2007). The Act also states that a Victims of Domestic Violence Support Fund and Victims of Domestic Violence Management Board must be established. The aim of this Fund is to support victims of domestic violence and their families by providing them with the needed material support and 57 University of Ghana http://ugspace.ug.edu.gh training. Also, the Fund is expected to provide shelters for victims of domestic violence, in the various regions and districts across the country among other mandates. The Victims of Domestic Violence Management Board is mandated to engage in any matter concerning domestic violence including, but not limited to conducting research on domestic violence, developing strategies to prevent domestic violence by coordinating with other agencies and organizations, and also managing the Fund. In Adomako Ampofo (2008), detailed accounts are presented on how various Non-Governmental Organizations, civil society groups, state actors and individuals through collective activism and hard work played instrumental roles in the formulation of the Domestic Violence Act 2007 (Act 732). These organizations include the International Federation of Women Lawyers Ghana (FIDA Ghana), Sisters Keepers, Ark Foundation, Women’s Initiative for Self-Empowerment (WISE), Gender Violence Survivors Support Network (GVSSN), Action Aid-Ghana, the UN Gender Systems programme for Promoting Gender Equality in Ghana, and the National Domestic Violence Coalition (Adomako Ampofo, 2008). While the mandates specified in the Act, especially those related to the Victims of Domestic Violence Support Fund and the Victims of Domestic Violence Management Board are essentially towards a worthy cause of protecting the fundamental human rights of individuals, and in this case women in intimate unions, to the best of my knowledge, there is no document to suggest that Fund and the Management Board are currently in operation, nearly 13 years after the Act was passed. Government and non-governmental bodies must maintain the commitment to eliminating all forms of violence against women and girls. As such for this legislation to function effectively, the proposed systems needed to ensure that the Act protects the fundamental human rights all, especially girls and women, must be given the full and continuous consideration from the state. 58 University of Ghana http://ugspace.ug.edu.gh 2.8 Gaps in the Literature Although issues on intimate partner violence perpetrated by men against women have become topical research areas for both policy makers and research scientists, still, there remain substantive gaps in the literature on male-perpetrated violence in sub-Saharan Africa. First, the literature review suggests that in understanding intimate partner violence it is important to take into account the influence of both individual and community level variables. While the literature review shows that studies on IPV mainly focus on individual level variables, and mostly from women’s perspective, they fail to provide context within which violence takes place. In this study, the aim is to examine the association between individual and community level factors on male-perpetrated intimate partner violence in Ghana from men’s perspective. Further, in spite of the fact that bridewealth payment remains a widespread cultural marriage practice in sub-Saharan Africa; the literature review has shown that there is a dearth of research essentially on bridewealth negotiation status and completeness of bridewealth payment, and intimate partner violence. This is largely because traditional surveys have failed to examine this relevant cultural variable. Relying on recent data from Ghana, this study attempts to empirically test the associations between different measures of bridewealth payment and intimate partner violence. Finally, the literature review suggests that studies on intimate partner violence have primarily used only one research approach, either quantitative or qualitative. While separately, these methodologies have their strengths, combining methodologies appropriately adds to our understanding of male-perpetrated violence. Using a mixed-methods approach provides new 59 University of Ghana http://ugspace.ug.edu.gh insights on the relationship between bridewealth payment and male-perpetrated intimate partner violence, a research area that has usually focused on women’s accounts. 2.9 Theories and conceptualization In the study of intimate partner violence, several frameworks have been posited to expound on perpetration of abuse in intimate relationships, whether in previous or current unions. The models that have frequently dominated the discourse of IPV are those theorized by feminist and family violence researchers (Lawson, 2012). These sociological perspectives argue that instead of individual characteristics, the use of violence in intimate relationships should be viewed as the influence of social structures and systems that prevail in one’s environment (Lawson, 2012). In this study, the sociological theories that underpin the study of male- perpetrated intimate partner violence are the integrated ecological framework, resource theory, and the exchange/social control theory. These theories allow for the examination of multiple cultural, socio-economic, and demographic correlates of male-perpetrated IPV. The integrated ecological framework describes how different levels of influences are linked with intimate partner violence (Heise, 1998) while the resource theory focuses on how availability of, and differences in resources fuel intimate partner violence (Goode, 1971). The exchange/social control theory refers to how differences in the cost and benefits of perpetrating violence increase the risk of violence perpetrated (Gelles, 1993; Gelles & Straus, 1979). 60 University of Ghana http://ugspace.ug.edu.gh 2.9.1 Integrated ecological framework The integrated ecological framework incorporates both feminist theory (Dobash & Dobash, 1979) and ecological perspective (Dutton, 2006; Belsky 1980) by family violence theorists (Heise, 1998) in examining violence perpetrated against women. This approach conceptualizes IPV “as multifaceted phenomenon grounded in an inter-play of personal, situational and socio-cultural factors” (Heise, 1998 page, 263). Essential to this assumption is how victimization of women in intimate unions is a function of both individual level factors and immediate and distal environmental factors, and the interaction between these factors (Heise, 1998). Feminist theorists argue that violence used against women is rooted in male domination over women, and being a woman places an individual at risk of experiencing violence (Anderson, 1997; Dobash & Dobash, 1979; Dobash et al., 1992; Johnson, 2005, 2006). Male advantage has an important effect on violence against women in intimate relationships mainly because male dominance strengthens traditional norms (Courtenay, 2000; Jewkes, 2002). Thus, this theory makes a case against patriarchal societies which put women in subordinate positions to men (Dobash & Dobash, 1979), by normalizing various forms of abuse committed against women (Namy et al., 2017). Gendered socialization, economic and power inequalities, and male domination over women (Takyi & Dodoo, 2005) fundamentally increase the risk of violence (Hindin, 2014). In addition, acceptance and justification of abuse against women in intimate unions is rooted in socially accepted patriarchal norms on the use of violence (Haj-Yahia, 2003; Haj-Yahia & Uysal, 2008; Hunnicutt, 2009; Namy et al., 2017). 61 University of Ghana http://ugspace.ug.edu.gh Contrary to family violence theorists who consider the family as the unit of analysis with regard to IPV, feminist theorists argue that gender should be considered as the unit of analysis, and IPV should be examined as a different form of violence from, for example, child or elderly violence. Feminist theorists have also argued that male dominance is the cause of intimate partner violence but fail to expound on how some men are not aggressive towards their partners although all men may have access to social learning on male dominance or advantage (Heise, 1998; Hunnicutt, 2009). Although male dominance is important in explaining violence against women, relying on this single cause limits our ability to understand into detail its etiology, and why the phenomenon has persisted over time (Heise, 1998). Indeed, research suggests that men who lack economic power are more likely to commit abuse against female partners in order to redeem “wounded masculinities” (Hunnicutt, 2009, p. 559). On the other hand, family violence theorists argue that IPV is an inevitable part of family relationships, and that, conflicts and confrontations are some of the methods used to resolve such conflicts, even in intimate interactions (Allen & Straus, 1979; Gelles, 1993; Gelles & Maynard, 1987; Gelles & Straus, 1979). Conflicts are normal within most family settings, and such expressions are not solely based on individual characteristics (Gelles & Straus, 1979). These theories therefore measure violence at the family unit than at the individual level. Dutton (2006) used the ecological framework to examine violence among intimate partners. Dutton proposes a nested ecological framework that stems from the Bronfenbrenner (1979) ecological framework which has been used extensively to study behavioural and other health outcomes (Green, Chesla, Beyene, & Kools, 2018; Kubiak, Brenner, Bybee, Campbell, & Fedock, 2018; Lee et al., 2011). Bronfenbrenner’s (1979) ecological framework assumes that individual behaviour is influenced by multiple factors ranging from one’s characteristics, those 62 University of Ghana http://ugspace.ug.edu.gh of their immediate family, community/societal level variables, and those relating to policy. The ecological framework was initially used by Belsky in 1980 to explain a form of family violence specifically, child abuse and neglect. Using an ecological framework in explaining child abuse enables one to identify multiple points of causality (Belsky, 1980). Heise (1998) therefore adopts Belsky's (1980) four levels of influence to explain intimate partner violence. At the individual level, the core of the framework, (see Figure 2.1) some factors identified to be associated with IPV are observing inter-parental violence as a juvenile, being a target of child abuse, and having an absentee father (Heise, 1998). At the second level, described as the microsystem level, male supremacy and control of family resources, alcohol usage, and marital disagreements or conflicts are some of the factors identified (Heise, 1998). At this level, the influence of male power or supremacy is at play, a situation feminist theorists identify as the source of violence perpetrated against women. 63 University of Ghana http://ugspace.ug.edu.gh Figure 2.1: Integrated ecological model showing the association between the different levels of the ecology and their influence of IPV Source: Heise, 1998 At the exosystem level, low socio-economic status or unemployment, isolation of women, and delinquent peer association have been found to be associated with IPV. Heise (1998, p. 273) observes that exosystem variables are usually the result of “changes taking place in the larger social milieu.” The macrosystem factors are those that relate to broader cultural beliefs and ideals that potentially influence other levels of the ecology (Heise, 1998). Here, the arguments of feminist theorists of gender inequality and subordination of women are at play. Factors such as male entitlement/ownership of women, male aggression, traditional gender norms, acceptance and the use of violence have been identified to exacerbate male-perpetrated 64 University of Ghana http://ugspace.ug.edu.gh violence against women in intimate unions (Heise, 1998; Klugman, 2017; Semahegn & Mengistie, 2015). These traditional norms shape an individual’s attitudes and predict the use of violence. This integrated ecological framework provides a model for examining how multiple factors are associated with IPV in sub-Saharan Africa by identifying vital cultural factors such as bridewealth payment, a macrosystem factor, which could potentially interact with other levels of the social ecology. Since the problem of IPV is complex, an integrated framework allows for the examination of possible causal factors in order to develop key culturally sensitive interventions and policies that fit developing settings in sub-Saharan Africa. 2.9.2 Resource theory Proponents of the resource theory argue that the availability of resources between couples determines the levels of violence perpetrated (Allen & Straus, 1979; Goode, 1971). Resources include income, education, occupation and other social positions of power. According to the resource theory, individuals who have more resources may be less likely to use violence to command respect or obedience in that, their resources are likely to do that for them (Goode, 1971). On the other hand, when resources like income and social status are not available, men may resort to violence as an alternative resource (Goode, 1971). For example, some studies suggest that males who are unemployed or have no income are more likely to commit violence against their intimate companions than those who are employed (Dery & Diedong, 2014; Lawrence et al., 2012; Namy et al., 2017; Panda & Agarwal, 2005; Weitzman, 2014). Thus, with 65 University of Ghana http://ugspace.ug.edu.gh poor economic conditions in society, and in the family, strain and conflicts may be inevitable among couples. When women are economically and socially dependent on their partners they are more likely to suffer, and even accept the use of violence, for fear that resources may be withdrawn (Adjei, 2017; Sanders, 2015; Sedziafa et al., 2017). In relation to relative resources, male partners with fewer resources than their female partners are more likely to use violence to gain compliance or project societal male authority or dominance (Atkinson et al., 2005; Goode, 1971; Kaukinen, 2004). When couples have some relative equality in terms of education, for instance, women are less likely to suffer abuse (Oduro et al., 2015). On attitudes toward beating wives, studies have revealed that access to resources is linked to lower acceptance of violence against women (Cools & Kotsadam, 2017; Doku & Asante, 2015). It is expected that males with advanced levels of schooling, those who are employed, and those in rich households will be less tolerant of IPV. 2.9.3 Exchange/Social control theory According to family violence theorists, once the benefits of violence are greater than the cost, individuals are likely to be violent (Gelles & Straus, 1979). The basic hypothesis is that individuals provide services and assume that their actions will be reciprocated thus, when such expectations are not met, disapproval and conflicts are inevitable. Bridewealth payment is related to the social exchange theory which encompasses the four components of the exchange; actors, resources, structures and processes (Molm, 2003). The actors are primarily the families of the bride and the groom, the man and the woman, and the larger society. Resources are the 66 University of Ghana http://ugspace.ug.edu.gh negotiated amount exchanged between the two families, while the structures involve the rights, responsibilities and arrangements that are negotiated to ensure that the agreements hold. The processes are all the laid down structures that are followed by the actors in the exchange. Bridewealth payment also involves both direct and indirect exchange structures (Horne et al., 2013). Direct exchange is between two actors, and it can be either negotiated or reciprocal. In negotiated exchanges, actors make joint decisions such that they agree on the terms and conditions of the exchange, and these terms are strictly binding (Horne et al., 2013). However, in reciprocal exchanges, actors contribute to the exchange separately without negotiations. It is to be noted that although in negotiated exchanges agreements are made, the outcomes may not always be equal for both parties (Horne et al., 2013; Molm, 2003). Indirect (generalized) exchanges involve more than two actors and can be net-or chain- generalized exchange. While in net-generalized exchanges actors give to a group and receive from that same group, in chain-generalized exchanges although actors give to a group they may not necessarily receive from that same group but from different actors in the network (Molm, 2003). With bridewealth payment, both elements of direct and chain-generalized exchanges are evident (Horne et al., 2013). Horne et al. (2013) argue that bridewealth payment involves a direct exchange between the two families because the groom and his family pay the negotiated amount of wealth to the bride’s kinsmen in exchange for the bride’s domestic, sexual and reproductive services. Bridewealth payment fundamentally triggers some levels of reciprocity. Gelles (1983) argues that the relationship could be broken if there is continuous lack of reciprocity. In this case, if men, after paying bridewealth, do not benefit from the rights to women’s domestic, sexual and reproductive services, it would be expected that the relationship will fall apart. On the contrary, in intimate unions, dissolving the relationship is not always the last resort (Gelles, 1983; Lawson, 67 University of Ghana http://ugspace.ug.edu.gh 2012). Men may use violence to ensure that their female partners perform their duties as bridewealth confers. According to Gelles (1983, p.157), the locus of the exchange/control theory is that “people hit and abuse their family member because they can.” Therefore, with structured exchanges (bridewealth), when one party, that is the man, believes that the woman is not reciprocating, he might be more inclined to be abusive to gain such control. On the other hand, since the woman knows that bridewealth has been exchanged, she will be more likely to perform her marriage duties, thus, averting various forms of violence from her partner. Further, since the actors involve not just the couple, but their families, and the society at large, men may be less likely to behave in ways that put their relationship with the large society at risk. These forms of social control can influence a couple’s behaviour such that they seek to maintain the needed compliance and support from all stakeholders. Some studies have also shown that traditionally, bridewealth payment goes to a male kin of the bride’s family, most likely to her brother who in turn uses that to pay for the bridewealth of his wife (Goody, 1973; Gelfand, 1973 as cited in Ansell, 2001). This form of transaction means that bridewealth payment network/structure is complicated, and does not merely involve the two immediate families but the whole society. As has been discussed above, for a marriage to be dissolved, the family of the bride must return all or part of the payment they received (Horne et al., 2013; Schneider, 1964). It is clear that this becomes difficult or almost impossible if bridewealth obtained from one family has already been exchanged for the services of a partner in another family. Horne et al. (2013) indicate that although bridewealth has been paid, the requirements are not automatically mandatory and there is a high risk that actors in the exchange may not fulfil their part of the bargain, in the case of this study, women may not perform or may be seen not to have performed marriage obligations. This may affect not only the families of the 68 University of Ghana http://ugspace.ug.edu.gh couple but also other members in the community. To prevent the whole network from falling apart, norms are developed such that both individuals and other community members would disapprove of individuals who do not meet their part of the bargain. Indeed, Horne et al. (2013) found that when bridewealth has been fully paid, participants expected that women in the bride’s family, women in the community, and participants themselves would be more permissive towards her husband’s use of violence against her regarding taking autonomous reproductive decisions. Therefore, in this study, it is expected that with completeness of bridewealth payment, men would be likely to approve of beating wives or perpetrate abuse if a woman does not meet her domestic obligations. Family violence theories on intimate partner violence are not without critique. Some of the key critiques of the resource and exchange/social control theories have been from feminist theorists of violence who argue that family violence theorist focus on inter-family violence rather than violence perpetrated against women (Jasinski, 2001). Another critique of family violence theories on intimate partner violence are the theories’ over reliance on data from a single partner. Thus, the information gathered from a single partner may be biased as researchers have no way of verifying misreported information due to social desirability, especially on sensitive matters like intimate partner violence were men under report their use of violence (Anderson, 1997). 2.10 Conceptual framework In this study, an adapted version of the integrated ecological framework (Heise, 1998), the resource theory, and exchange theory are used to examine bridewealth payment on men’s attitudes toward wife-beating and actual perpetration of violence (see Figure 2.2). The initial 69 University of Ghana http://ugspace.ug.edu.gh integrated framework by Heise (1998) is used with several modifications. The study examines two forms of IPV; attitudes toward violence (wife-beating), and actual perpetration of physical violence against an intimate female partner. The first dependent variables are men’s attitudes toward wife-beating if a woman (a) “goes out without telling her husband”, (b) “burns the food”, (c) “argues with her husband”, (d) “neglects the children”, or (e) “refuses to have sex with her husband”. The second dependent variable measures lifetime self-reported actual perpetration of physical violence against an intimate female partner. The conceptual framework shows the association between individual, situational and cultural elements, and approval, and actual perpetration of IPV. The dependent variables and all individual and community level variables refer to the man’s characteristics. The main levels of influence are individual characteristics and community level factors. At the individual level, socio-demographic variables are examined. Socio-demographic characteristics like age, level of schooling, religion and employment status/type of earning, are identified to be associated with IPV. Individuals with higher social and economic resources may be less likely to approve of IPV. Further, personal history of observing violence as a child, number of children born and duration of marriage are also examined. 3Observing violence between one’s caregivers, having no children or having more than 5 children, and shorter duration of marriage may increase women’s risk of IPV. 3 Except for variables labelled with asterisks (*) and pluses (+), all other control variables are found in both datasets. Control variables that have been labelled with asterisks (*) can only be found in the GDHS datasets, and those labelled with pluses (+) can only be found in the Bridewealth Study datasets. All various presented here are all refereeing to men’s characteristics 70 University of Ghana http://ugspace.ug.edu.gh The framework shows community level factors that relate to cultural and societal ideals are examined. These include type of marriage, household wealth, lineage/ethnicity and place of residence. These factors can influence variables in the other levels of the ecology. It is expected that men in polygynous unions, patrilineal men, and those who earn cash only would be more likely to approve of or perpetrate IPV. Further, at the community level, attitudes of wife-beating in a particular context are shaped by traditional norms. These 4perceptions potentially influence the acceptance and use of violence in intimate unions. In this study bridewealth payment, an essential cultural (community level) factor for establishing marriages in sub-Saharan Africa is examined. It is the main independent variable in the study, and it is measured as whether or not bridewealth has been negotiated and completeness of bridewealth payment (full, partial, and not negotiated/paid). 4 Scholars agree that perceptions and attitudes are closely linked (Bahamonde-Birke, Kunert, Link, & Ortúzar, 2017; Pickens, 2005). The research questions and objectives were formulated based on the set of questions that produced the first dependent variable. This variable is a universal concept (known as attitudes towards wife-beating (Hindin 2014; Rani and Bonu 2009; Mann and Takyi 2009)) which assess an individual’s acceptability of violence. 71 University of Ghana http://ugspace.ug.edu.gh Figure 2.2: Conceptual framework indicating the relationship between men’s individual, situational, and cultural factors and IPV Bridewealth practice Individual level factors Community level factors Bridewealth negotiation status Age CEB *Marriage type Negotiated Employment status/type of earning Lineage/ethnicity Not negotiated Level of education *Household wealth Completeness of bridewealth payment Duration of marriage *Place of residence Full payment + Partial payment Religious affiliation Social support/network + No payment +Observed violence as a child Social control Forms of intimate partner violence Men’s attitudes toward wife-beating Physical IPV perpetration Source: Adapted from Heise, 1998 72 University of Ghana http://ugspace.ug.edu.gh CHAPTER THREE DATA SOURCES, SAMPLING PROCEDURE AND METHODOLOGY 3.1 Introduction The study used mixed research methods approach to assess the association between bridewealth and male-perpetrated intimate partner violence in Ghana. The 2014 Ghana Demographic and Health Survey (GDHS), and the Bridewealth Payment and Normative Constraints on Women’s Lives in Ghana (Bridewealth Study) conducted in 2016, and the follow- up qualitative data in 2018 by the Regional Institute for Population Studies, University of Ghana, were employed. The two quantitative datasets complement each other by addressing research questions on men’s attitudes toward wife-beating and actual perpetration of wife-beating with regard to bridewealth payment. The variables under consideration refer solely to the man’s characteristics. The qualitative dataset examined participants’ perceptions on the practice of bridewealth and men’s use of violence in intimate unions. To expand our understanding of intimate partner violence in Ghana, the study explores how different levels of influences, particularly socio-economic, demographic, and cultural variables like bridewealth payment mediate men’s justification and use of violence in intimate unions by relying on critical realism as the philosophical paradigm that underpin this study. Critical realists seek to go beyond the surface of social interactions and understand the social processes and nuances that these interactions present. Further, they argue that our knowledge of the world is fallible and that no one person knows the truth (Mcevoy & Richards, 2006; McEvoy & Richards, 2007). Since human behaviour is complex it would be difficult to understand this 73 University of Ghana http://ugspace.ug.edu.gh complexity by merely objectifying human behaviour. This is because human behaviour is not only a function of personal choices but it is interlinked in a web of multidimensional processes and influences (Mcevoy & Richards, 2006). In view of this, to study male-perpetrated intimate partner violence, it is vital to lean on this philosophical perspective that gives one the ability to employ different methodologies and tap not only into individual choices but also understand the reality of the world around them and how these realities have been shaped by society (Mcevoy & Richards, 2006). Thus, a mixed-method approach adds diversity to phenomenon under study (Ågerfalk, 2013; Lund, 2012) as quantitative data identifies the predictors of male-perpetrated violence and the qualitative adds context and depth. In this study, intimate partner violence is examined at two levels: the first one relates to men’s attitudes toward wife-beating, and the other examines actual perpetration of wife-beating. First, the 2014 Ghana Demographic and Health Survey, a nationally representative dataset, is useful because it addresses questions on only attitudes toward wife-beating, a situation which is pervasive, and research has shown that approval or justification of wife-beating may lead to perpetration of violence (Adomako Ampofo, 1993; Bowman, 2003; Doku & Asante, 2015; Ozaki & Otis, 2017; Speizer, 2010; Spencer et al., 2017; Takyi & Lamptey, 2016). In addition, approval of wife-beating may prevent neighbours and other family members from intervening when there is a violent incident in a household, especially when spousal violence and traditional norms regarding meddling are widespread (Green, Wilke, & Cooper, 2017). As has been mentioned in the previous chapters, bridewealth payment is not always made in full, and studies have shown that this has consequences for women’s limited reproductive autonomy (Dodoo, Horne, & Biney, 2014; Dodoo & Dodoo, 2017; Horne et al., 2013). In the 2014 GDHS, currently married men were asked whether or not bridewealth was negotiated in their current union, and 74 University of Ghana http://ugspace.ug.edu.gh those who indicated that bridewealth was negotiated were further asked whether it was fully, partially or not paid at all. Actual perpetration of wife-beating is examined using the Bridewealth Study dataset. This dataset examines lifetime self-reported physical violence perpetration. In this survey also, all men who were currently in union were asked whether bridewealth was negotiated in their current union, and the completeness of bridewealth payment. Thus, this survey allows for the examination of completeness of bridewealth payment, and physical abuse against female partners. Apart from the Negotiating Reproductive Outcomes study in Uganda (Blanc et al., 1996) conducted more than two decades ago, and a few recent studies which examined some measures of bridewealth (Abramsky et al., 2011; Horne et al., 2013; Wojcicki et al., 2010), to the best of my knowledge, studies scarcely examine bridewealth payment and attitudes, and actual perpetration of wife-beating quantitatively, more so engaging men. Using these datasets allows for the investigation of the association between bridewealth and men’s attitudes toward violence with the DHS dataset, and actual perpetration using the Bridewealth Study dataset, in an area of research that has focused mainly on women even though men pay and gain rights over women’s reproductive and domestic services, and are often reported as primary perpetrators of intimate partner violence. 3.2 The 2014 Ghana Demographic and Health Survey The 2014 Ghana Demographic and Health Survey (GDHS), a nationally representative survey used a two-stage sampling design based on a sampling frame obtained from the 2010 Population and Housing Census (Ghana Statistical Service, 2013b). At the first stage of the 75 University of Ghana http://ugspace.ug.edu.gh sampling process, 427 enumeration areas were obtained with 216 clusters from urban and 211 from rural areas. At the second stage, household listing activities were undertaken in all designated enumeration areas and about 30 households were selected from every cluster randomly (GSS et al., 2015a). At this stage, 12,831 households were obtained, and from these, 12,010 households were occupied. Interviews were successfully conducted in 11,835 households yielding a response rate of 99 percent. Women in their reproductive ages (15-49 years) and in half of these households, men aged 15-59 years were part of the survey. Usual residents of the households or guests who lived in these houses the night before the research were examined. Of the 4,609 eligible men identified in the household interviews, 4,388 were successfully interviewed and of this number 1,967 (unweighted) were married (GSS et al., 2015a). The multi-stage cluster sampling design employed in selecting participants for the 2014 GDHS is a probability sampling design which ensures that participants are given an equal chance of selection. This sampling method is employed is situations where it is generally difficult to obtain an exhaustive list of the members in the target population, and in this case men in their reproductive ages 15-59 years in Ghana. Although this this sampling procedure is efficient and cost-effective, it may be saddled with sampling bias and error challenges. This is likely to occur when clusters or elements selected from the clusters are similar. To reduce some of these biases, it is important to rely on relatively larger sample sizes (Babbie, 2015). For this thesis, the objective required that the data to be analysed be restricted to all currently married men. The data were weighted with the factors prescribed by the GDHS to adjust for sampling errors. Thus, the weighted sample size for this study was 1,893 married men. The Ghana Statistical Service (GSS), the Ghana Health Service (GHS), and the National Public 76 University of Ghana http://ugspace.ug.edu.gh Health Reference Laboratory (NPHRL) of the GHS implemented the survey, while the DHS Program through ICF International provided technical assistance (GSS et al., 2015a). 3.3 Bridewealth Payment and Normative Constraints on Women’s Lives in Ghana (Bridewealth Study) The second dataset employed was the Bridewealth Study data. This is a population-based dataset and for the main study, consenting men aged 518 years and above were examined. This study was conducted in 18 communities in the Eastern Region of Ghana. The Eastern Region is bordered by five other regions (Greater Accra, Volta, Brong-Ahafo, Ashanti and Central regions) and has 26 administrative districts (Ghana Statistical Service, 2013a). The region has a land cover of 19,323 square kilometres making up 8.1 percent of the total land area of Ghana (Ghana Statistical Service, 2013a). According to the 2010 Population and Housing Census (PHC) of Ghana, the population of the Eastern Region is the 3rd largest in Ghana with a population of 2,633,154 (1,342,615 females and 1,290,539 males), and a population density of 136.3 per square kilometre in 2010 (Ghana Statistical Service, 2013a). Of the total population of the region, more than half (56.0%) reside in rural areas (Ghana Statistical Service, 2013b). The dominant ethnic group in the region is the Akan (51.1%), and the Ewe (18.9%), Ga- Dangme (17%) and Guan (5.3%) as the other major ethnic groups (Ghana Statistical Service, 5 In Ghana the median age at first marriage for men is about 26.5 years. According to the 2014 Ghana Demographic and Health Survey, only 10% of men age 25-49 were married/cohabited by age 20 and only 2% of men age 20-24 were married by age 18. The Bridewealth Study was interested in examining norms, perceptions and behaviour in relation to bridewealth payment thus, the study focused on adults 18 years and above. 77 University of Ghana http://ugspace.ug.edu.gh 2013a). The remaining 7.7% represent other minor ethnic groups. The main occupational activities in the region are subsistence agriculture, cash crop farming and bauxite mining. In recent times, gold mining has emerged as one of the major economic activities in the region (Ghana Statistical Service, 2013a). About 40.6 percent and 7.0 percent of the population respectively are married and living together (Ghana Statistical Service, 2013a). According to the 2010 Ghana Population and Housing Census, the total fertility rate (TFR) of the region is 5.1 births per woman, and is higher than the national average of 4.0 (Ghana Statistical Service, 2013b, 2013a; Ghana Statistical Service et al., 2015). The TFR of urban dwellers is 4.6 compared to 5.6 births per woman in the rural areas (Ghana Statistical Service, 2013a). According to the 2010 PHC, household compositions are diverse, with heads of household, spouses, children, grandchildren, and other relatives living in the same household (Ghana Statistical Service, 2013b). This reflects the practice of the extended family system in Ghana, and in the region, and has implications for transfer of cultural values and norms. The region has two types of lineages: matrilineal and patrilineal. Matrilineal Akan, Patrilineal Akan, Matrilineal Guan and Patrilineal Guan groups are the dominant lineage groups in the region, and this allows for the examination of how bridewealth payment affects various aspects of women’s lives in these lineages/ethnic groups. With these differences in bridewealth payment practices and the implications for power relations including violence, the study area is appropriate for examining the effects of bridewealth payment across these ethnic and lineage lines. The Bridewealth Study was conducted in July 2016. In all, 1,152 males and 1248 females were interviewed. For this thesis, the data was restricted to 579 men who were cohabiting or 78 University of Ghana http://ugspace.ug.edu.gh married. Sampling was purposively done across lineage and ethnic lines as studies have shown that bridewealth payment practices vary for matrilineal and patrilineal lineages, and have different implications for intimate partner violence (Frost & Dodoo, 2010; Horne et al., 2013). The study, in the form of a vignette experiment, was conducted with only individuals who identified as indigenes of a given locality (see Appendix A for a sample of the vignette/questionnaire). The main reason was to capture local representations and norms of people who belong to these areas. Of the 18 study sites, eight are Matrilineal Akan communities (Anyasu, Mpakadan, Nnudu, Aboasa, Apeguso, Old Senchi and Akwamufie, Akropong), three Matrilineal Guan communities (Anum, Boso and Tosen), three Patrilineal Akan communities (Mampong, Obosomase and Mamfe), and four Patrilineal Guan communities (Apirede, Awukugua, Dawu and Abiriw) (see Appendix B for the list of communities and sample sizes). The vignette experiment was used to obtain information about bridewealth payment, norms, practices, and reproductive and health outcomes from respondents in the selected communities. Short stories about different bridewealth payment statuses were read to participants. A computer-assisted randomization tool was used to assign participants to various vignette experiments. All interviews were conducted face-to-face by trained interviewers, using tablet computers. Respondents answered questions regarding stories in the vignette experiments randomly assigned to them. In addition, they provided answers to other demographic, social, marriage history, and questions on bridewealth status. Only women and men who indicated that they were in intimate unions were asked for bridewealth payment and physical violence-related information. The survey instruments were pre-tested before the actual survey. 79 University of Ghana http://ugspace.ug.edu.gh 3.4 Qualitative data collection In January 2018, researchers on the Bridewealth Study returned to some of the study locations and conducted 6in-depth interviews with some individuals in these areas (see Appendix C for list of localities and description of interview participants). These interviews were precipitated by some of the responses obtained from the earlier quantitative survey conducted in 2016. The interviews were conducted with key informants (opinion leaders-chiefs, queen mothers, linguists, a priestess) and other elderly individuals in some of these localities. The purpose for interviewing these individuals was to understand and clarify issues relating to the practice of bridewealth payment and the implications for social, sexual, other reproductive outcomes, and IPV. These in-depth interviews unearthed traditional implications of bridewealth practice. In all, 16 interviews were conducted; 8 male and 8 female interviews. These interviews were mostly semi-structured and informal in nature. The interviewers were all graduate students and other fieldworkers who have considerable qualitative data collection experience and skills, and had all been trained in the Bridewealth Study. The principal investigators supervised the data collection process and were involved in interviewing the participants. Interviewers were trained and the interview guide (see Appendix D for interview guide) was piloted. The interview guide was structured such that it allowed the participants to engage with the issue of bridewealth and the implications into details while facilitators probed where needed. As part of the training process, the purpose of the 6 Primarily, all in-depth interviews were designed to be individual interviews. However, in some cases, for example, while interviewing a chief, his linguist may join the discussion. In some cases it impossible to continue with only the primary participant. 80 University of Ghana http://ugspace.ug.edu.gh qualitative data collection was explained to the fieldworkers and the interview guide was discussed. Therefore, challenges regarding the interview guide were addressed. The interviews were conducted in local languages preferred by the participants. The interviews lasted for about 45 minutes to 90 minutes. The interviews were conducted in the participants’ residence or a place they preferred. The interviews explored participants’ perceptions on the implications of bridewealth payment for contraceptive usage, decision making, childbearing, inheritance and intimate partner violence. Ethical approval for the Bridewealth Study was granted by the Ethics Committee for the Humanities (ECH) at the University of Ghana. 3.5 Measurement and description of variables 3.5.1 Objectives 1 and 2: bridewealth negotiation status, completeness of bridewealth payment and attitudes toward wife-beating among men The data relevant to these objectives were from the 2014 Ghana Demographic and Health Survey. 3.5.1.1 Dependent variable 1: attitudes toward wife-beating Men were asked whether in their opinion a husband is justified in hitting or beating his wife if she (a) “goes out without telling him”, or (b) “neglects the children”, or (c) “argues with him”, or (d) “refuses to have sex with him”, or (e) “burns the food”. It should be noted that these 81 University of Ghana http://ugspace.ug.edu.gh scenarios or reasons speak to traditional Ghanaian roles and responsibilities of married women, and that bridewealth payment confers women’s reproductive, sexual and domestic rights to men. These dependent variables examine attitudes toward wife-beating using the five scenarios, and it measures traditional patriarchal norms which suggest that a husband can discipline his wife (Kishor & Bradley, 2012). For example, respondents were asked: “is a husband justified in hitting or beating his wife if she goes out without telling him?” The original responses were coded as “Yes”, “No” and “Don’t know”. For each scenario or reason, a binary outcome variable for attitudes toward wife-beating was created. Respondents who said “No” were coded as “Not approving of wife-beating” (0). However, if they indicated “Yes” or “7Don’t know” they were coded as “Approving of wife-beating” (1) (Table 3.1). A composite variable measuring attitudes toward wife-beating on at least one of the five scenarios was also created. Respondents indicating a “No” or negative responses to all five situations were recoded as “Not approving of wife-beating” (0). On the other hand, respondents answering “Yes” or a positive response or “Don’t know” to at least any of the five situations were coded as “Yes, approving of wife- beating” (1). Past research shows that approval of wife-beating on at least one of the five conditions is considered as unfavourable and it is positively associated with men’s use of violence against their intimate partners (Nwabunike & Tenkorang, 2017; Takyi & Lamptey, 2016; Alesina et al., 2016). It is possible that a man who approves of wife-beating on one of the five reasons may not be the same as someone who approves of wife-beating on all five reasons for example. However, the 7 Only three (3) respondents indicated ‘Don’t know’. They were combined with those who indicated ‘Yes’. This is because if we conceptualize approval of wife-beating as an empowerment issue, we would expect men who disapprove of wife-beating to say ‘No’. All others are considered otherwise. Again, the small number of men who stated ‘Don’t know’ did not significantly affect the results obtained when they were combined with the category, ‘Yes’. 82 University of Ghana http://ugspace.ug.edu.gh proportions of men who approved of wife-beating on the different reasons are somewhat few thus; the study is unable to estimate any robust multinomial logistic regression models relying on such composite measure. The frequency distribution of men who approve wife-beating on more than one of the reasons is presented in Appendix G. Table 3. 1: Description and measurement of variables Variables Description and measurement Forms and categories Dependent variables Attitudes toward wife- Respondent is asked if a husband Categorical variables beating is justified in hitting or beating his Do not approve of wife- wife if she: beating 1. “goes out without telling him” Approve of wife-beating 2. “neglects the children” 3. “argues with him” 4. “burns the food” 5. “refuses to have sex with him” Composite variable for If the respondent agrees to at least Do not approve of wife- attitudes toward wife- any of the above items then he beating beating approves of wife-beating and takes Approve of wife-beating a value of “1”. He does not approve of wife-beating if he does not agree to any of the above items and takes a value of “0” Main independent variables Bridewealth negotiation “Was bridewealth negotiated in Categorical variable status your current union?” Negotiated Not negotiated Completeness of The variable is computed from Categorical variable bridewealth payment two questions: Full payment “Was bridewealth negotiated in Partial payment your current union?” Not negotiated/paid “What is the status of bridewealth in your current union?” 83 University of Ghana http://ugspace.ug.edu.gh 3.5.1.2 Main independent variable 1: bridewealth negotiation status In the 2014 GDHS, currently married men were asked: “was bridewealth negotiated in your current union?” Those who responded in the affirmative to the question were coded as “Yes negotiated” (1) and those who said no were coded as “No, not negotiated”. 3.5.1.3 Main independent variable 2: completeness of bridewealth payment Completeness of bridewealth payment is second main independent variable for the assessment of how bridewealth payment is related to men’s attitudes toward wife-beating. This variable measures the level or extent to which bridewealth has been paid. In the 2014 GDHS, currently married men were asked: “was bridewealth negotiated in your current union?” The responses were: “Yes” (1) and “No” (2). Those who responded affirmatively to the question above were then asked: “what is the status of bridewealth in your current union?” The responses were “paid in full” (0), “partially paid” (1), and “not paid” (2). The small proportion of men who indicated that bridewealth had been negotiated but not paid (22 out of 1964) were combined with those who stated that they were married but bridewealth had not been negotiated (355 out of 1964). This is because the small proportion of men in the category ‘bridewealth negotiated but not paid’ could not allow for any robust statistical analyses. One way to resolve the statistical issue in a logical manner was to combine these two categories since they both indicate no payment statuses. As such the two categories together connote a state in which men should not have the authority that comes from bridewealth payment. Thus, completeness of bridewealth payment was categorized as (0) meaning “fully paid”, (1) meaning “partially paid” and (2) meaning “not negotiated/paid”. 84 University of Ghana http://ugspace.ug.edu.gh 3.5.2 Objectives 3 and 4: whether or not bridewealth has been negotiated, completeness of bridewealth payment and physical violence perpetration by men For objectives 3 and 4, data from the Bridewealth Study were used. 3.5.2.1 Dependent variable 2: physical violence perpetration Men were asked whether they had ever beaten their female partners. The binary responses to the question were “Yes” (1) and “No” (2). Those who indicated that they had ever beaten their partners were coded as having perpetrated abuse (1), while those who indicated that they had never beaten their partners were coded as never having perpetrated abuse (0). 3.5.2.2 Main independent variable1: bridewealth negotiation status The Bridewealth Study asked all men in 8unions whether bridewealth was negotiated in their current union and the binary response Yes/No was elicited. Thus, those who indicated that bridewealth was negotiated, coded (1), were distinguished from those who responded in the negative, coded (0). 8 Both cohabiting and married men were asked whether bridewealth had been negotiated in their current union. Some cohabiting men indicated that bridewealth had been negotiated and had been partially paid. Appendix F shows the proportion of men who indicated that bridewealth had been negotiated in their current union and completeness of payment. It is possible that some men indicated that they were cohabiting because bridewealth had not been fully paid. With regard to the Bridewealth Study, all men currently in unions are analysed in this thesis. 85 University of Ghana http://ugspace.ug.edu.gh 3.5.2.3 Main independent variable 2: completeness of bridewealth payment In the Bridewealth Study, those who indicated that bridewealth was negotiated for their marriages, were further asked the status of the payment, that is whether it was fully paid (0), partially paid (1), or not paid (2). Table 3. 2: Description and measurement of variables Variables Description and measurement Form and categories Dependent variable Categorical variable Physical violence “Have you ever beaten your partner?” No Yes Main independent variables Bridewealth negotiation “Was bridewealth negotiated in your Categorical variable status current union?” Negotiated Not negotiated The variable is computed from two questions: Categorical variable Completeness of “Was bridewealth negotiated in your Full payment bridewealth payment current union?” Partial payment “What is the status of bridewealth in Not negotiated/paid your current union?” All respondents who indicated that bridewealth was not negotiated in their current union were coded “Not negotiated/paid" (2). Respondents who indicated that bridewealth was not negotiated at all in their current unions (60 out of 579) were combined with those who stated that bridewealth was negotiated but not paid (55 out of 579), because the small cases in these two categories did not allow for any robust analyses and they all represent non-payment status The advantage of asking all respondents in union whether bridewealth was negotiated in their unions is that the study is able to capture cohabiting respondents and their bridewealth payment statuses. 86 University of Ghana http://ugspace.ug.edu.gh Again, it must be noted that the fluid nature of unions in the sub-region could mask the statuses of respondents with regard to bridewealth payment. By asking all respondents currently in union about their bridewealth status (completeness of bridewealth payment), the study does not assume that because they indicated that they were cohabiting, bridewealth had not been negotiated at all or nothing had been paid. Some respondents indicated that they were cohabiting possibly because bridewealth had not been paid fully. Asking all respondents in union questions on bridewealth also increases the sample size for those within the various bridewealth statuses. 3.5.3 Control variables Informed by the review of relevant literature as discussed in Chapter Two above, control variables were selected for this study. The study controls for confounding variables that potentially also mediate the association between bridewealth payment and intimate partner violence. The control variables were grouped into individual and community level factors, adapting the integrated ecological framework design. In studying the association between bridewealth payment and intimate partner violence, we adjusted for confounding variables that were also associated with bridewealth payment and IPV (Agresti, 2008). Individual level factors explored were respondent’s religious affiliation, level of education, 9type of earning, age and 10employment status. 9 Type of earning variable is only available in the 2014 Ghana Demographic and Health Survey datasets, and it measures type of remuneration respondents received from working in the last 12 months preceding the survey. 10 Respondent’s employment status is assessed in the Bridewealth Study dataset as employed or not employed. 87 University of Ghana http://ugspace.ug.edu.gh Table 3. 3: Description and measurement of control variables Control variables Description and measurement Form and categories Individual level factors 11*Religious affiliation Respondent’s religious affiliation Categorical variable Christian Moslem 12Other +Religious affiliation Respondent’s religious affiliation Categorical variable 13Orthodox Pentecostal/charismatic Other Level of education Highest level of education attained Categorical variable by respondent No education Primary Middle/Junior secondary school (JSS)/Junior high school (JHS) Secondary/Higher +Level of education Highest level of education attained Categorical variable by respondent No/primary education Middle/Junior secondary school (JSS)/Junior high school (JHS) Secondary/Higher *Type of earning Respondent’s type of earning Categorical variable received in 12 months preceding Not paid/not working the survey Paid in cash only Paid in cash and in-kind +Employment status Whether respondent is employed Categorical variable or not No Yes Age Age of respondent in years Continuous variable Number of children Number of children the respondent Categorical variable has ever had 0 1- 4 children 5 or more children Duration of marriage Measures how long respondent has Continuous variable been in their current union in years +Observed violence Whether or not respondent Categorical variable as a child observed IPV a child between No parents or caregivers Yes 11 ‘*’ indicates variables only in 2014 GDHS; ‘+’ indicates variables only in Bridewealth Study 12 Other religion category includes those who indicated traditional and no religion. 13 The category Orthodox Christians was created by combining Catholics, Presbyterians, Methodists and Anglicans 88 University of Ghana http://ugspace.ug.edu.gh Table 3.3 (Continued): Description and measurement of control variables Control variables Description and measurement Form and categories Community level factors *Household wealth index The wealth index was computed Categorical variable using household assets e.g. floor Poor and roofing material etc. Middle The score was divided into five Rich equal parts to obtain five wealth categories. *Place of residence Whether respondent lives in a Categorical variable rural or urban area Urban Rural *Ethnicity1 Uses respondent’s ethnicity Categorical variable Akan Ga-Dangbe/Ewe Mole-Dagbani 14Other +Ethnicity2 Uses respondent’s ethnicity and Categorical variable lineage Matrilineal Guan Matrilineal Akan Patrilineal Akan Patrilineal Guan *Marriage type Whether respondent has other Categorical variable wives Monogamous Polygynous ‘*’ indicates variables only in 2014 GDHS; + indicates variables only in Bridewealth study Concerning type of earning, respondents who stated that they were “Paid in cash and in kind” are respondents who received both in-cash and in-kind forms of remuneration (Ghana Statistical Service, 2014, p. 86). Only 7 men indicated that they earn only “in-kind” forms of remuneration. This category was combined with the category “paid in cash and in-kind”. Other individual level variables examined were number of children ever born (CEB), duration of marriage and 14 The “Other” ethnic group category from the 2014 Ghana Demographic and Health Survey datasets is consist of individuals who identify as Guan, Grusi, Gurma, Mande and Other ethnic groups. 89 University of Ghana http://ugspace.ug.edu.gh observing violence as a child. Household wealth, place of residence, lineage/ethnicity, and marriage type were some community level factors examined. It is worth noting that individuals belonging to the various ethnic groups also identify as either matrilineal or patrilineal. In Ghana, the Akan are the major matrilineal ethnic group. The Ga- Dangbe and Ewe are two major patrilineal groups in the southern part of the country. Due to the few cases in the cells, these two ethnic groups; Ga-Dangbe and Ewe were combined. The Mole- Dagbani, a patrilineal ethnic group in the northern part of Ghana was also examined. All other ethnic groups were categorized as ‘Other’ (see Table 3.3). Due to the very small proportion of men in the Ga-Dangbe and Ewe ethnic groups (see Table 5.4 in Chapter 5) who approve of “wife-beating if a woman refuses to have sex with her husband”, ethnicity was recoded as Matrilineal Akan (0) and Other (1) while conducting analysis with this dependent variable. In the Bridewealth Study, respondents were grouped as Matrilineal Akan, Matrilineal Guan, Patrilineal Akan and Patrilineal Guan. In Table 3.3 above, a detailed description of the control variables including their measurements and categorizations are presented. 3.6 Methods of analyses 3.6.1 Quantitative data analyses The GDHS data were weighted to adjust for disproportionate sampling and non-response as recommended by Becker & Kalamar (2018). In analysing the men’s dataset, the men’s weighting variable ‘mv005’ was used. A new weight variable was created by dividing mv005 by 1,000,000. This is because although DHS calculates the sampling weights to six decimal points 90 University of Ghana http://ugspace.ug.edu.gh they are presented without the decimals. Since the DHS uses a two-stage sampling design, in weighting the dataset, the sampling design must be taken into account when calculating standard errors for analyses that involve confidence intervals. Therefore, the survey set (svyset) command was specified. For all analyses using the GDHS, the ‘svy’ command was used to adjust for the sampling design. The Stata commands used in weighting the dataset are presented in Appendix E. The survey datasets were initially analysed by describing the variables of interest. Bar graphs and tables were used to present these results. Bivariate analyses were conducted to examine the association between the independent, control, and dependent variables. Chi-square and ANOVA tests were employed to examine the associations between bridewealth and IPV (men’s attitudes and actual perpetration of physical abuse). Binary logistic models were used at the multivariate level to examine the predictors of attitudes toward wife-beating and perpetration of intimate partner violence, adjusting for individual and community level factors. Binary logistic regression models were employed for the analyses, because the dependent variables were binary; ‘approval of wife-beating or not’, and ‘perpetrated IPV or not’. Identifying variables that predict IPV has implications for planning interventions that are likely to reduce or prevent the incidence of IPV (Reed & Wu, 2013). This study predicted the probability of approval of wife-beating or perpetration of violence by assessing the association between bridewealth payment and the outcome (IPV). A dependent variable with two categories can only take on two values, 0 and 1, and the probability of an event occurring ranges from 0 to 1. Thus, the logit transformation model was used. log it P = ln  P   1− P  91 University of Ghana http://ugspace.ug.edu.gh Considering attitudes toward violence for example, P is the probability of approval of wife-beating and 1-P is the probability of not approving wife-beating. The results of the logistic regression models can be stated as log odds of an event occurring in terms of the independent variables, and an error term (Agresti, 2008). The association between all ecological variables and attitudes toward wife-beating and perpetration of IPV were examined in the logistic models. In this study, separate models were presented for (i) attitudes toward wife-beating, and (ii) perpetration of violence. Separate models are presented for attitudes about wife-beating if (a) “a woman goes out without telling her husband”, (b) “neglects the children”, (c) “argues with her husband”, and (d) “refuses to have sex with her husband”. Due to the small number of men who approve of wife-beating “if a woman burns the food”, no robust bivariate and multivariate models could be presented for this dependent variable. A composite variable which assesses attitudes toward wife-beating on at least one of the five scenarios was also treated as another dependent variable. The datasets were all analysed using Stata 13. 3.6.2 Qualitative data analysis All audio-taped interviews were translated and transcribed verbatim from the local dialect (Twi) to English language. The qualitative coding and further analyses were conducted using Atlas.ti Version 7.5.7 qualitative software tool. All transcribed interviews were read thoroughly and initial codes were identified. Thematic analysis was used to analyze the qualitative data and the results were displayed in a thematic network form (Attride-Stirling, 2001). The transcripts were coded using the initial codes. These initial codes are referred to as the basic themes. Various categories of the basic themes which reflect the same idea or concept were grouped into organizing themes. The overarching global theme reflects participants’ 92 University of Ghana http://ugspace.ug.edu.gh perceptions about bridewealth payment and men’s use of violence against their intimate partners. Ethical clearance for the Bridewealth Study was granted by the Ethics Committee for the Humanities (ECH) at the University of Ghana. 3.7 Limitations of the study There are some limitations to this study. Firstly, by measuring perpetration of violence with the question: ‘have you ever beaten your wife?’, the study is unable to explore other aspects of physical abuse such as “slapping, pushing, hitting with a fist kicking/dragging/beating/choking/burning, or threatening or using a weapon” (Fleming et al., 2015, p.5). Indeed, studies have shown that intimate partner violence is not only physical; it could be emotional, economical and sexual (Oduro et al., 2015; Panda & Agarwal, 2005). Further, issues relating to recall bias and social desirability could present some challenges with regard to self-reported retrospective data. Due to the fact that data on actual perpetration of violence and attitudes towards wife-beating are all self-reported responses, some participants may under-report if they have ever been abusive or favour the use of violence against intimate partners. Also, because bridewealth payment is an important marriage practice, men may not accurately state their bridewealth payment status especially when bridewealth has not been fully paid because they may be perceived as not performing this important marriage practice. The study attempted to reduce these biases by taking interviewers through extensive survey training, and the vignettes were also piloted. Interviewers were trained to develop good rapport with respondents when asking questions relating to bridewealth and perpetrating abuse in order to 93 University of Ghana http://ugspace.ug.edu.gh reduce some of these biases. Notwithstanding these, the study provides some new insights into bridewealth payment and wife-beating. Again, with regard to the 15GDHS, only respondents who qualified themselves as currently married were asked questions relating to bridewealth. As indicated earlier, asking only married respondents about bridewealth assumes that bridewealth payment cannot be associated with cohabiting individuals. In making this assumption, the disadvantage is failure to capture individuals who may have reported that they were cohabiting and, therefore, not asked questions on bridewealth at all. In the Bridewealth Study, all consenting adults the in marriage and in cohabiting unions were asked questions on bridewealth. The results showed that some men in cohabiting unions reported that bridewealth had been negotiated and they also reported that bridewealth had been partially paid or not paid (see Appendix F). Despite these setbacks, the study had robust research techniques for data collection and analyses and the results are credible. Also, although previous studies have shown that attitudes towards wife-beating are associated with actual use of violence, this study was unable to examine this link because no single dataset presented data on bridewealth, attitudes towards wife-beating and actual perpetration of abuse. However, the study begins to throw some light on how bridewealth could be associated with attitude about wife-beating and men’s use of violence against women and the importance of exploring this link in future studies. 15 The candidate is registered with the DHS Program as a dataset user. This allowed for the download and use of the 2014 GDHS dataset for this study. The DHS program gives access to survey data for academic research. 94 University of Ghana http://ugspace.ug.edu.gh CHAPTER FOUR BACKGROUND CHARACTERISTICS OF STUDY RESPONDENTS 4.1 Introduction Past research has shown that factors associated with intimate partner violence cut across individual, personal, and household/community level variables (Dutton, 2006; Heise, 1998). This section presents descriptive statistics on men’s attitudes toward wife-beating and actual perpetration of physical abuse, bridewealth payment, socio-demographic and household/community level characteristics of the study respondents. The chapter provides descriptive statistics of male respondents from the 2014 GDHS, and the Bridewealth Study. 4.2 Ghana Demographic and Health Survey, 2014 4.2.1 Men’s attitudes toward wife-beating, GDHS Figure 4.1 shows the proportion of men who approve of wife-beating if a “woman goes out without telling her husband, neglects the children, argues with her husband, refuses to have sex with her husband, or burns the food”. It should be emphasized that these scenarios relate to traditional responsibilities of women in intimate unions. Thus, a woman’s inability or refusal to perform these duties may provoke men’s disapproval. The results show that close to one in ten men approve of wife-beating for at least one of the above reasons (see Figure 4.1). 95 University of Ghana http://ugspace.ug.edu.gh Again, less than 10.0 percent of the men suggested that wife-beating is justified for each of the five reasons. For example, 6.1 percent of men approve of wife-beating if a woman neglects the children. With regard to going out without telling her husband and arguing with her husband, about one in twenty men approve of wife-beating. Clearly, 4.0 percent of the men indicated that wife-beating is justified if a woman refuses to have sex with her husband. Only 1.9 percent of the men approve of wife-beating if a woman 16burns the food. Figure 4. 1: Percentage of men who approve of wife-beating 12 10 9.6 8 6.1 6 5.3 4.8 4.0 4 1.9 2 0 Goes out Negelcts the Argues with Refuses to Burns food Approval of without telling children her husband have sex with wife-beating her husband her husband on at least one reason Approval of wife-beating Source: 2014 Ghana Demographic and Health Survey 16 Due to the small proportion of men 1.9 percent (36 men) who approve of wife-beating if a woman burns the food, no robust analyses could be run at the bivariate and multivariate stages of analyses. Therefore, in subsequent chapters, no results are presented for the dependent variable; attitude toward wife-beating if a woman burns the food. 96 Percentage (%) University of Ghana http://ugspace.ug.edu.gh 4.2.2 Bridewealth negotiation status, GDHS From the 2014 Demographic and Health Survey, a little over eight in ten men indicated that bridewealth has been negotiated in their current union (see Figure 4.2). Figure 4. 2: Percentage distribution of men by bridewealth negotiation status 100 81.3 80 60 40 18.7 20 0 Negotiated Not negotiated Bridewealth negotiation status Source: 2016 Bridewealth Study (RIPS) 4.2.3 Completeness of bridewealth payment, GDHS Results for completeness of bridewealth payment are shown in Figure 4.3. More than six in ten men (68.2%) reported that bridewealth has been fully paid in their current union, and for close to one in five men (19.8%), bridewealth has not been paid. About 11.9 percent of the men stated that bridewealth has been partially paid. 97 Percentage (%) University of Ghana http://ugspace.ug.edu.gh Figure 4. 3: Percentage distribution of men by completeness of bridewealth payment 100 80 68.3 60 40 19.8 20 11.9 0 Fully paid Partially paid Not negotiated/paid Completeness of bridewealth payment Source: 2014 Ghana Demographic and Health Survey 4.2.4 Socio-demographic and economic characteristics of respondents, GDHS The mean age of men in the sample was 41.3 years. More than two-thirds of the study population identified as Christians, compared with a little over one in five who were Moslems (Table 4.1). A large proportion of men (two in five) had middle or junior secondary education, and 18.8 percent had no formal education. Nearly a third (29.2%) had some secondary or tertiary education. Regarding type of earning, 16 men who stated that they were not working at all when the survey was conducted were not asked if they received any forms of remuneration. These 16 respondents were combined with the category ‘not paid’. Again 7 out of 1,893 men stated that they received remunerations in kind only. These 7 men were combined with the category of men 98 Percentage (%) University of Ghana http://ugspace.ug.edu.gh who stated that they earned cash and in-kind as forms of remuneration. More than seven out of ten men stated that they earned cash only, and less than one in ten indicated that they earned no income. About 45.0 percent of the men had one to four children, while only 15.0 percent had no children when the survey was conducted. The mean duration of marriage is 15 years. While 45.0 percent of the men lived in rich households, close to two in five lived in poor households. Nearly equal proportions of men lived in urban (49.1%) and rural (50.9%) areas. Concerning the distribution by ethnicity, a large proportion of men identified as Akan (44%), and 17.5 percent were Mole-Dagbani. Slightly more than nine in ten men were in monogamous unions (see Table 4.1). 99 University of Ghana http://ugspace.ug.edu.gh Table 4. 1: Percentage distribution of men by socio-demographic and economic characteristics Respondent’s characteristics Number Percent (%) Religious affiliation Christian 1,280 68.0 Moslem 398 21.0 Other 214 11.3 Level of education No education 357 18.8 Primary 227 12.0 Junior secondary 755 39.9 Secondary/higher 553 29.2 Type of earning Not working/not paid 149 7.9 Paid in cash only 1464 77.4 Paid in cash and in-kind 280 14.8 Number of children ever born No child 289 15.3 1 to 4 children 852 45.0 5 or more children 751 39.7 Household wealth Poor 724 38.3 Middle 310 16.4 Rich 859 45.4 Place of residence Urban 929 49.1 Rural 964 50.9 Ethnicity1 Akan 838 44.3 Ga-Dangbe and Ewe 156 22.2 Mole-Dagbani 331 17.5 Other 304 16.1 Type of marriage Monogamous 1,724 91.1 Polygynous 169 8.9 Mean Standard deviation Age 41.3 9.1 Duration of marriage 15.0 9.7 Total 1893 100.0 Source: 2014 Ghana Demographic and Health Survey 100 University of Ghana http://ugspace.ug.edu.gh 4.3 2016 Bridewealth Study 4.3.1 Physical violence perpetration, Bridewealth Study This section on wife-beating sought to examine the proportion of men who stated that they had ever beaten their wives. The results show that male-perpetrated violence against women is prevalent. One in five men (20.9%) indicated that they had ever beaten their female partners (see Figure 4.4). Figure 4. 4: Percentage distribution of men who perpetrated physical violence 100 79.1 80 60 40 20.9 20 0 No Yes Physical violence Source: 2016 Bridewealth Study (RIPS) 4.3.2 Bridewealth negotiation status, Bridewealth Study Figure 4.5 shows the proportion of men who indicated whether bridewealth had been negotiated in their union or not. The results show that majority of the men (89.64%) stated that bridewealth had been negotiated in their current union. 101 Percentage (%) University of Ghana http://ugspace.ug.edu.gh Figure 4. 5: Percentage distribution of men by bridewealth negotiation status 100 89.6 80 60 40 20 10.4 0 Negotiated Not negotiated Bridewealth negotiation status Source: 2016 Bridewealth Study (RIPS) 4.3.3 Completeness of bridewealth payment, Bridewealth Study Figure 4.6 displays variations in completeness of bridewealth payment of the sample. More than half (56.0%) of the sample reported that bridewealth had been fully paid. About a fifth (19.9%) stated that bridewealth had not been paid, and for a quarter (24.2%), bridewealth had been partially paid in their current union. 102 Percentage (%) University of Ghana http://ugspace.ug.edu.gh Figure 4. 6: Percentage distribution of men by completeness of bridewealth payment 100 80 60 56 40 24.2 19.9 20 0 Fully paid Partially paid Not negotiated/paid Completenss of bridewealth payment Source: 2016 Bridewealth Study (RIPS) 4.3.4 Socio-demographic and economic characteristics of respondents, Bridewealth Study The percentage distribution of individual characteristics and community level variables is presented in Table 4.2. The mean age of respondents was 47.3 years (minimum age=19years and maximum age 90years). On average, respondents have been married for about 16 years. More than eight in ten men were Christians. Of this, 47.0 percent were Orthodox Christians, and close to two in five were Pentecostals/Charismatics. Majority of the men had some junior secondary education, and one in three had senior secondary or higher education. More than a tenth had no education or only primary education. Slightly over eight in ten (86.0%) were employed. Almost half had two to four children (49.0%). A fifth of the men in the sample stated that they had observed violence between their parents/caregivers as children. The results show that about a 103 Percentage (%) University of Ghana http://ugspace.ug.edu.gh quarter (25.6%) of the sample identified as Matrilineal Guan, and equal proportions of men were Patrilineal Akan (23.7%) and Patrilineal Guan (23.7%). Table 4. 2: Percentage distribution of men by socio-demographic and economic characteristics Respondent’s characteristics Number Percent (%) Religious affiliation Orthodox 272 47.0 Pentecostal/charismatic 226 39.0 Other 81 14.0 Level of education 17No education/primary 76 13.1 Junior secondary 307 53.0 Secondary/higher 196 33.9 Employment status Not employed 81 14.0 Employed 498 86.0 Number of children ever born 0 to1 children 83 14.3 2 to 4 children 284 49.0 5 or more children 212 36.6 Observed violence as a child No 464 80.1 Yes 115 19.9 Ethnicity2 Matrilineal Guan 148 25.6 Matrilineal Akan 157 27.1 Patrilineal Akan 137 23.7 Patrilineal Guan 137 23.7 Total 579 100.0 Mean Standard deviation Age 47.3 14.7 Duration of marriage 15.9 12.9 Source: 2016 Bridewealth Study (RIPS) 17 For level of education, only 26 men had no education and 50 men primary education. These two categories were combined in order to obtain a large enough sample size for that category to run robust regression analyses. 104 University of Ghana http://ugspace.ug.edu.gh 4.4 Socio-demographic characteristics of qualitative study participants, Bridewealth Study In the previous chapter, it was indicated that the qualitative interviews were conducted with key informants specifically, opinion leaders, chiefs, queen mothers, a priestess and elderly individuals in the localities. Characteristics of the qualitative participants are presented in Table 4.3. On average the informants were about 40 years of age and older. The study segmented equal proportions of males and females, and the proportions of individuals who identified as Akan were about 56 percent. Majority of the participants identify as matrilineal (68.7%). In addition, one in four informants reported that they are natives of Anyasu. Table 4. 3: Characteristics of participants in the qualitative study Characteristics Number Percentage (%) Sex Male 8 50.0 Female 8 50.0 Lineage Matrilineal 11 68.7 Patrilineal 5 31.3 Ethnicity Akan 9 56.2 Guan 7 43.8 Locality Mampong 3 18.8 Akwamufie 2 12.5 Tosen 2 12.5 Anum 3 18.7 Anyasu 4 25.0 Abiriw 2 12.5 Age Range 40 years and older - Source: 2018 Bridewealth Study (RIPS) 105 University of Ghana http://ugspace.ug.edu.gh 4.5 Discussion The aim of this chapter was to examine the characteristics of men in the study populations by describing the prevalence of approval of wife-beating, actual perpetration of physical abuse, bridewealth payment and other socio-demographic and economic characteristics of men. The results show that bridewealth payment is widespread, and for the two datasets, a large proportion of men indicated that bridewealth had been negotiated in their current union. In the 2014 GDHS study, about eight in ten men (81.95%) reported that bridewealth had been negotiated in their current union while in the Bridewealth Study; nine out of ten men stated same. This shows that for majority of the unions, families of the couple know one another and have had some discussions about the union in question. In relation to completeness of payment, more than half of the men in the GDHS and Bridewealth Study samples reported that bridewealth has been fully paid in their current union. With regard to the Bridewealth Study dataset, more men, 24.2 percent stated that bridewealth has been partially paid compared to 11.9 percent of the men in the GDHS. Other studies have reported that bridewealth payment is widespread in the sub-region (Anderson, 2007; Blanc et al., 1996; Dodoo, Horne, & Biney, 2014; Frost & Dodoo, 2010; Horne et al., 2013; Oheneba-Sakyi & Takyi, 2006). For example, in Uganda about 79.1 percent of men reported that bridewealth had been negotiated in their current union (Blanc et al., 1996). This underscores the importance of this cultural marriage practice in sub-Saharan Africa. In some settings, however, the practice of bridewealth payment may not be as prevalent as recorded in the above mentioned studies. Among a sample of 417 currently and formerly married women in Zimbabwe, less than half (48.7%) stated that bridewealth has been negotiated, and of these, a little over one in ten (13.8%) reported that bridewealth has been paid fully (Wojcicki et al., 2010). 106 University of Ghana http://ugspace.ug.edu.gh Concerning men’s attitudes toward wife-beating for the five scenarios examined, the results show that approval of wife-beating is less prevalent among men, although it is justified for various reasons. About one in ten men approve of wife-beating on at least one of the five reasons for wife-beating. Studies in sub-Saharan Africa suggest that approval or justification of wife-beating is less prevalent among men (Alesina et al., 2016; Rani et al., 2004; Speizer, 2010). Alesina et al. (2016) noted that fewer men reported justification of wife-beating due to interviewer effect or social desirability. Further, high levels of acceptance of wife-beating in a given context may cloud men’s perceptions on its pervasiveness (Rani et al., 2004). Other studies have shown that attitudes are significantly affected by what individuals perceive as socially right or wrong in a given context. Therefore, if men perceive approval of wife-beating as being socially wrong or as a deviant behaviour they would be less likely to openly approve of it, although their attitudes may be significantly different from their actual behaviour. Relatively, a high proportion of men approve of wife-beating if a woman neglects the children. This could explain the importance placed on child bearing and child care. This is similar to findings from seven sub-Saharan African countries where a high proportion of men stated that neglecting the child could justify men’s use of violence against women (Rani et al., 2004). The issue of approval of violence is rooted in its normalization by many in society (Namy et al., 2017), and is further deepened by gendered norms where child care is mainly the responsibility of women. Very few men (36 out 1893) approve of wife-beating if a woman burns the food. It is possible that majority of the men do not consider burning food as transgressing from traditional gender norms compared to the other reasons for wife-beating. 107 University of Ghana http://ugspace.ug.edu.gh On the contrary, male perpetrated intimate partner violence is prevalent. About one in five men reported having ever beaten their wives. The high incidence of physical violence obtained from the Bridewealth Study is possibly due to its prevalence in the rural study localities. The fewer proportions of men who approve of wife-beating, compared with the high prevalence of actual wife-beating show the disconnect between attitudes and actual behaviour. Although, majority of the men disapprove of wife-beating, this may not always reflect their actual experiences. While the study does not examine attitudes toward wife-beating and actual perpetration of physical violence using the same datasets, these results begin to shed light on bridewealth payment and the implication on male-perpetrated violence against women, more so the disconnect between attitudes about wife-beating and actual perpetration of physical abuse. Concerning individual level variables, majority of the respondents identified as Christians, followed by Moslems, and others representing minor religious groups. This compares with the general Ghanaian population. A high proportion of men in both datasets had some formal education (Ghana Statistical Service, et al., 2015). Majority of the men were employed and earned cash only. On average, men had between one to four children. Like previous studies, family history of violence was also prevalent in the Bridewealth Study sample (Speizer, 2010; Yount et al., 2014). Men in the Bridewealth Study were relatively older than their counterparts in the GDHS dataset. However, the mean duration of marriage of men in both samples was similar. More than two in five men lived in rich households. These results compare with some national level datasets (Ghana Statistical Service, 2007). Almost equal proportions of men in the GDHS lived in rural and urban areas. Although over the past two decades the Ghana Demographic and Health Surveys have shown that the proportion of men in polygynous unions is decreasing, still a sizable number of men are in these unions (GSS and IRD, 1989; GSS et al., 2015). 108 University of Ghana http://ugspace.ug.edu.gh CHAPTER FIVE BRIDEWEALTH PAYMENT, SOCIO-DEMOGRAPHIC AND ECONOMIC CORRELATES OF MEN’S ATTITUDES TOWARD WIFE-BEATING AND ACTUAL PERPETRATION OF PHYSICAL ABUSE 5.1 Introduction This chapter seeks to explain male-perpetrated intimate partner violence by examining the patterns of association between attitudes toward wife-beating and bridewealth payment on one hand, and actual perpetration and bridewealth payment, and other predictors on the other hand. Bivariate results are first presented for men’s attitudes toward “wife-beating if a woman goes out without telling her husband, or neglects the children, or argues with her husband, or refuses to have sex with her husband” (outcome variables) on bridewealth payment and other socio-demographic and economic predictors. The outcome variable measuring attitudes toward wife-beating if a woman burns food is excluded because of the small number of cases. Bivariate results for attitudes toward wife-beating on at least one of the five reasons for wife-beating on bridewealth and socio-demographic predictors are also presented. In Table 5.1, bivariate results are presented for bridewealth negotiation status and attitudes toward wife-beating. Further, results are presented for completeness of bridewealth payment and attitudes about wife-beating. The chapter further examines the association between actual perpetration of physical violence and bridewealth negotiation status, and also, completeness of bridewealth payment. For categorical variables, cross tabulations were used to examine the relationship between attitudes toward wife-beating and actual perpetration of wife-beating on bridewealth payment and other predictor variables. ANOVA was used to compare the mean differences (for age and duration of 109 University of Ghana http://ugspace.ug.edu.gh marriage) between men who approve of wife-beating and those who do not, and those who have ever perpetrated physical violence and those who have not. 5.2 Bivariate results on men’s attitudes toward wife-beating, GDHS 5.2.1 Bridewealth negotiation status and men’s attitudes toward wife-beating This section examines the association between bridewealth negotiation status and men’s attitudes about wife-beating. Table 5.1 shows that generally, for each reason, less than 10 percent of men approve of wife-beating. Concerning approval of wife-beating on at least one reason, 10.01 percent of men who stated that bridewealth has been negotiated in their current union approve of wife-beating. However, there was no significant association between bridewealth negotiation status and men’s attitudes toward wife-beating. Table 5 1: Percentage distribution of men by bridewealth negotiation status and approval of wife-beating Goes out Bridewealth without Neglects Argues Refuses to At least negotiation telling her the with her have sex with one status husband children husband her husband reason % % % % % Negotiated 3.64 4.16 3.02 2.64 7.76 Not negotiated 5.72 6.56 5.51 4.28 10.01 Chi square 2.583 2.999 3.150 2.107 1.736 p-value 0.147 0.112 0.121 0.169 0.247 Source: 2014 Ghana Demographic and Health Survey 110 University of Ghana http://ugspace.ug.edu.gh 5.2.2 Completeness of bridewealth payment and men’s attitudes toward wife-beating, GDHS Figure 5.1 shows the association between completeness of bridewealth payment and men’s approval of wife-beating. Completeness of bridewealth payment was not significantly associated with attitudes toward wife-beating on all reason stated for approval of wife-beating. Generally, a high proportion of men who stated that bridewealth has been partially paid in their current union approve of wife-beating with regards to the measures on attitudes towards wife-beating except for approval of wife-beating if a woman argues with her husband. Concerning approval of wife-beating on at least one of the five measures of attitude toward wife- beating, the results showed that 11.39 percent of men in the partially paid category approve of wife-beating. In the case where a woman argues with her husband however, 3.99 percent of men approve of wife-beating when bridewealth has been partially compared. 111 University of Ghana http://ugspace.ug.edu.gh Figure 5. 1: Percentage distribution of men by completeness of bridewealth payment and attitudes toward wife-beating Fully paid 12 11.39 Partially paid Not negotiate/paid 10 9.43 8.95 9.05 8 6.39 5.98 6 5.43 4.85 5.15 4.36 3.99 4.09 4.24 4.36 4 2.83 2 0 Goes out without Neglects the children Argues with her Refuses to have sex At least one of the telling her husband husband with her husband five reasons Completeness of bridewealth payment Source: 2014 Ghana Demographic and Health Survey 112 Percentage (%) University of Ghana http://ugspace.ug.edu.gh 5.2.2.2 Association between socio-demographic characteristics and men’s attitudes toward wife-beating, GDHS This section examines the association between socio-demographic characteristics and men’s attitudes toward wife-beating on all measures of attitudes toward wife-beating. The results show that some individual level variables were significantly associated with men’s attitudes toward wife-beating (Table 5.2). Religious affiliation was significantly associated with men’s attitudes toward wife-beating on all measures of attitudes toward wife-beating examined. Generally, the proportion of men who stated that they identified with ‘Other’ religions approve of wife-beating. As indicated in Table 5.2, while 3.08 percent of Christians approve of wife- beating if a woman goes out without telling her husband, close to one in five men who identified with ‘Other’ religions (Traditional African religion and no religion) approve of wife-beating (p- value=0.000). A significant association between men’s level of education and their attitudes toward wife-beating was observed. Men’s approval of wife-beating decreased with increasing level of education. From Table 5.2, whereas 10.81 percent of men with no education approve of wife- beating, about 1.25 percent with secondary or higher education approve of wife-beating if a woman refuses to have sex with her husband. It can be observed that a lower proportion of men who earn cash only approve of wife-beating on all measures of approval of wife-beating, and this relationship is statistically significant (p-value<0.05). For example, a lower proportion of men who earned cash only (6.49%) approve of wife-beating on at least one of the five reasons compared to 20.65 percent of the men who earned both cash and in-kind forms of remunerations. 113 University of Ghana http://ugspace.ug.edu.gh Number of children ever born was significantly associated with men’s attitudes toward wife-beating. Generally, the proportions of men who approve of wife-beating increased with increasing number of children ever born with regard to all measures of attitudes toward wife- beating. In the case where a woman argues with her husband, while 2.02 percent of men who indicated that they have no children approve of wife-beating, about 6.35 percent of men who have 5 or more children reported that they approve of wife-beating. As shown in Table 5.2, the Chi square test shows significant association between all household/community level variables and men’s attitudes toward wife-beating on all measures of attitudes toward wife-beating at 95% confidence level. In general, the proportion of men who approve of wife-beating decreased with increasing household wealth. In the instance where a woman goes out without telling her husband, while 9.57 percent of men in the poor household wealth category approve of wife-beating about 2.30 percent of men in rich households approve of wife-beating. Further, the results show that place of residence was significantly associated with men’s attitudes toward wife-beating except in the instance where a woman argues with her husband. The data present significant rural-urban differences in that the proportion of rural men (8.49%) who approve of wife-beating was higher than those in urban areas (3.63%) in the case where the woman neglects the children (p-value<0.05). Ethnicity was significantly associated with men’s attitudes toward wife-beating (p- value<0.05). For example, higher proportions of men who identified with Mole-Dagbani (7.32%) and ‘Other’ ethnic groups (9.09%) approve of wife-beating if a woman argues with her husband compared with men from other ethnic groups. Concerning all the measures on attitudes toward wife-beating, the results show that type of marriage was significantly associated with men’s 114 University of Ghana http://ugspace.ug.edu.gh approval of wife-beating. Largely, a high proportion of polygynous men approve of wife-beating compared to their monogamous counterparts. For instance, less than one in ten men in monogamous unions approve of wife-beating compared to 21.34 percent of men in polygynous unions on at least one of the five measures of men’s attitudes toward wife-beating. Finally, the results show that age and duration of marriage were not significantly associated with men’s attitudes toward wife-beating. 115 University of Ghana http://ugspace.ug.edu.gh Table 5 2: Percentage distribution of the association between socio-demographic characteristics and men’s attitudes toward wife-beating Goes out with Neglects the Argues with her Refuses to have telling her On at least one of husband children husband sex with her husband the five reasons N Respondent’s Yes Chi Yes Chi Yes Chi Yes Chi Chi characteristics (%) square (%) square (%) square (%) square Yes (%) square Religious affiliation 28.32* 29.24*** 11.41*** 28.32*** 22.3*** Christian 3.08 3.7 3.03 2.06 6.23 1280 Moslem 6.15 6.64 6.76 4.93 12.44 398 Other 17.32 19.52 11.75 13.65 24.4 214 Level of education 20.97* 22.52*** 8.83*** 18.89*** 25.9*** No education 13.44 14.88 9.86 10.81 22.03 357 Primary 7.4 8.9 7.12 6.05 13.48 227 Junior secondary 3.4 4.05 3.29 2.12 6.72 755 Secondary/Higher 1.9 2.12 2.66 1.25 3.89 553 Type of earning 28.3* 14.47*** 14.47*** 18.11*** 18.67*** Not working/not paid 9.1 11.52 11.02 6.2 19.28 149 Cash only 2.9 3.97 3.14 2.15 6.49 1464 Paid in cash and in- kind 16 14.42 10.21 12.34 20.65 280 Number of children ever born 7.36** 6.21** 4.36* 5.4** 7.08** No child 2.7 3.18 2.02 2.6 5.2 289 1 to 4 children 4.2 5.22 4.38 2.95 8.73 854 5 or more children 7.65 8.24 6.35 5.66 12.25 751 116 University of Ghana http://ugspace.ug.edu.gh Table 5.2 (Continued): Percentage distribution of the association between socio-demographic characteristics and men’s approval of wife-beating Goes out with telling her Neglects the Argues with her Refuses to have sex On at least one husband children husband with her husband of the five N reasons Respondent’s Yes Chi Yes Chi Yes Chi Yes Chi Yes Chi characteristics (%) square (%) square (%) square (%) square (%) square Household wealth 16.49* 18*** 5.91** 20.54*** 20.92*** Poor 9.57 10.85 7.25 8.01 15.9 724 Middle 3.8 4.82 3.67 1.89 7.7 310 Rich 2.3 2.57 3.14 1.31 4.95 859 Place of residence 8.35** 18*** 2.29 14.59*** 10.24** Urban 3.4 3.63 3.87 1.81 6.65 929 Rural 7.2 8.49 5.71 6.05 12.43 964 Ethnicity 17.14*** 19.35*** 5.25** 17.81*** 16.43*** Akan 2.8 3.15 2.96 2.16 6.08 838 Ga-Dangbe and Ewe 2.5 2.96 3.38 0.6 4.98 419 Mole-Dagbani 6.4 7.41 7.32 4.76 12.52 331 Other 14.9 17.18 9.09 12.74 22.41 304 Type of marriage 26.04*** 25.74*** 15.94*** 23.9*** 17.99*** Monogamous 4.3 5 4.02 3.21 8.43 1724 Polygynous 16 17.38 12.81 11.75 21.43 169 Mean F Mean F Mean F Mean F Mean F Age 40.6 0.58 40.69 0.56 41.73 0.24 41.34 0.00 40.28 2.54 1893 Duration of marriage 16.2 1.75 16.09 1.76 16.41 2.25 16.06 1.06 15.92 2.11 1893 “*p<0.05;**p<0.01;***p<0.001” Source: 2014 Ghana Demographic and Health Survey 117 University of Ghana http://ugspace.ug.edu.gh 5.3 Physical violence perpetration, Bridewealth Study 5.3.1 Bridewealth negotiation status and physical violence perpetration, Bridewealth Study Bivariate results showing the association between bridewealth negotiation status and actual perpetration of physical abuse are shown in the bar chart in Figure 5.2. There is no statistically significant association between bridewealth negotiation status and men perpetrating intimate partner violence (p-value= 0.135). While more than a quarter of the men who indicated that bridewealth has not been negotiated perpetrated physical violence, a fifth of men in unions where bridewealth has been negotiated perpetrated physical violence against their female partners. Figure 5. 2: Percentage distribution of men by bridewealth negotiation status and having ever perpetrated physical abuse 120 No 100 Yes 79.96 80 71.67 60 40 28.33 20.04 20 0 Bridewealth negotiated Not negotiated Bridewealth negotiation status Source: 2016 Bridewealth Study 118 Percentage (%) University of Ghana http://ugspace.ug.edu.gh 5.3.2 Completeness of bridewealth payment and physical abuse perpetration, Bridewealth Study On the contrary, completeness of bridewealth payment is significantly associated with physical violence of women (p-value =0.003). The bar graph below, Figure 5.3, shows that while close to one in three men who indicated that bridewealth had not been paid perpetrated physical abuse, one in five men (20.0%) with partial payment had. For full bridewealth payment, 17.28 percent of men perpetrated abuse. These bivariate results suggest that having ever perpetrated physical abuse decreases with completeness of bridewealth payment. Figure 5. 3: Percentage distribution of men by completeness of bridewealth payment and having ever perpetrated physical abuse No 100 Yes 82.72 80 80 67.83 60 40 32.17 20 17.28 20 0 Fully paid Partially paid Not negotiated/paid Completeness of bridewealth payment Source: 2016 Bridewealth Study 119 Percentage (%) University of Ghana http://ugspace.ug.edu.gh 5.3.3 Association between socio-demographic characteristics and physical abuse, Bridewealth Study This section examines the association between men’s individual characteristics and household/community variables and having ever perpetrated physical abuse (Table 5.3). There is a weak association between religious affiliation and having ever perpetrated wife-beating (p- value 0.098). Similar proportions of men who were affiliated with the Orthodox and ‘Other’ religions perpetrated abuse against their wives. The test of association between level of education and having ever perpetrated abuse was significant (p-value=0.014). About one in four men (25.41%) with junior secondary education perpetrated physical abuse compared with 13.16 percent with no education or primary education. Observing violence between parents/caregivers as a child was strongly associated with having ever perpetrated violence at older ages (p- value=0.000). Clearly, about a third who observed violence between their parents/caregivers when they were children reported that they had ever beaten their female partners. A one way analysis of variance was computed to compare the mean ages of men who had ever beaten their wives and those who had never beaten their wives (Table 5.3). The ANOVA results revealed a statistically significant difference in mean ages for men who had ever beaten their wives and those who had not (F (1, 576) = 14.89, p=0.000). Generally, men who had ever beaten their wives were younger (42.6 years) than those who had never beaten their wives (48.4 years). There is weak association between lineage/ethnicity and actual wife-beating (p-value =0.068). While a little over one in five men (20.95%) who identified as Matrilineal Guan reported that they had ever beaten their wives, 28.47 percent of Patrilineal Guan men had ever beaten their wives. 120 University of Ghana http://ugspace.ug.edu.gh Table 5. 3: Percentage distribution of the association between socio-demographic characteristics and having ever perpetrated physical abuse Respondent’s characteristics Yes (%) Chi square p-value Number Religious affiliations 4.64 0.098 Orthodox 23.53 272 Pentecostal/charismatic 16.37 226 Other 24.69 81 Level of education 8.49 0.014 No education/primary 13.16 76 Junior secondary 25.41 307 Secondary/higher 16.84 196 Employment status 1.34 0.247 Not employed 16.05 81 Employed 21.69 498 Number of children ever born 3.24 0.197 0 to 1 child 21.69 83 2 to 4 23.59 284 5 or more 16.98 212 Observed violence as a child 12.81 0.000 No 17.89 464 Yes 33.04 115 Lineage/ethnicity 7.12 0.068 Matrilineal Guan 20.95 148 Matrilineal Akan 16.56 157 Patrilineal Guan 28.47 137 Patrilineal Akan 18.25 137 Mean F(degrees of freedom) p-value Age 42.6 14.89(1, 576) 0.000 579 Duration of marriage 14.9 1.07(1, 576) 0.302 579 Source: 2016 Bridewealth Study 121 University of Ghana http://ugspace.ug.edu.gh 5.4 Discussion The aim of this chapter was to assess the association between respondents’ background characteristics and their attitudes and actual perpetration of violence. The results suggest that bridewealth (negotiated or not and completeness of payment) were not significantly associated with attitudes toward wife-beating. Regarding having ever perpetrated physical abuse, bridewealth payment negotiated or not was not significantly associated with physical violence. However, completeness of bridewealth payment was significantly associated with the use of physical violence, such that moving from no payment, through partial payment to full payment, the proportions of men who had ever beaten their wives decrease. Relying on the resource theory, it can be argued that men who do not pay bridewealth are potentially men who do not have the needed resources to undertake this traditional marriage rite. Thus, as Goode, (1971) argues such men may rely on violence as an alternative resource to maintain their authority in the union. With increasing education, men were less likely to approve of wife-beating, and other studies have shown similar results (Rani et al., 2004; Takyi & Mann, 2006). However, concerning men’s actual behaviour, that is having ever beaten their partner, the results show that a higher proportion of men with junior secondary education had ever beaten their wives compared with their counterparts with no education/primary or secondary/higher education. For all scenarios relating to attitudes toward wife-beating, fewer men who earned cash only approve of wife-beating compared to those who earned no income or earned both cash and kind. These results suggest that access to resources could be related to men’s attitudes about wife-beating. According to the resource theory, husbands who have resources (and in this case higher levels of 122 University of Ghana http://ugspace.ug.edu.gh education and income) may be less likely to use overt force to gain control or power in their relationships (Goode, 1971). Generally, a higher proportion of men who had more than five children approve of wife- beating. Some studies have shown that men who observe violence between their parents/caregivers are more likely to normalize violence and use it in adult relationships (Dery & Diedong, 2014; Namy et al., 2017; Yount et al., 2014). Similar to these studies, the current study shows that a higher percentage of men in the study (Bridewealth Study) who observed parental violence had ever beaten their female partners. The results further show that household/community level variables were significantly associated with attitudes toward wife-beating. This suggests the significance of examining different levels of influence of IPV. In relation to attitudes toward wife-beating on at least one reason, the results show that fairly a large proportion of men in rural regions, those in poor households, and those in polygynous unions approve of physical abuse. Similar results emerged for all four reasons for justifying wife-beating that is “if a woman goes out without telling her husband, neglects the children, argues with her husband, or refuses to have sex with her husband”. The bivariate results indicate that a large proportion of men in polygynous unions approve of wife-beating. Polygynous unions are characteristic of rural (Reniers & Tfaily, 2008) and less educated populations (Dodoo & Seal, 1994), and has implication for marital disruption (Gage-Brandon, 1992), and approval of violence against women (Rani et al., 2004; Uthman et al., 2010). Some studies have similarly found that men in rich households disapprove of wife- beating compared to their counterparts in poor households (Rani et al., 2004; Takyi & Mann, 2006). On ethnicity, the outcomes suggest that mainly a higher proportion of patrilineal ethnic 123 University of Ghana http://ugspace.ug.edu.gh groups in the northern part of Ghana approve of wife-beating. In addition, higher proportions of men who were identified as Matrilineal Guan and Patrilineal Guan stated that they had ever perpetrated physical abuse. While no statistically significant mean difference in age is observed between men who approve of wife-beating and those who do not, there was a significant mean difference in age for men who had ever perpetrated physical abuse and those who never had. Men who indicated that they had ever perpetrated abuse were significantly younger (42.6 years) than those who had never perpetrated physical abuse (48.4 years). Dalal et al. (2012) argue that young men in rural areas and those who have access to fewer resources (lower levels of education and income) are more likely to justify wife-beating. It is also possible that older men are underreporting their use of violence in intimate unions. 124 University of Ghana http://ugspace.ug.edu.gh CHAPTER SIX ASSOCIATION BETWEEN COMPLETENESS OF BRIDEWEALTH PAYMENT AND MEN’S ATTITUDES TOWARD WIFE-BEATING CONTROLLING FOR OTHER SELECTED CHARACTERISTICS 6.1 Introduction In this chapter, multivariate results of the association between bridewealth payment and men’s attitudes toward wife-beating, controlling for selected socio-demographic and economic predictors are examined. The aim of this chapter was to examine the independent effects of bridewealth negotiation status, and completeness of bridewealth payment on men’s attitudes toward wife-beating after controlling for other relevant predictors of attitudes toward wife- beating informed by the literature review. Logistic regression estimates are presented for attitudes toward “wife-beating if a woman goes out without telling her husband, or neglects the children, or argues with her husband, or refuses to have sex with her husband”. Separate models are presented for bridewealth negotiation status and completeness of bridewealth payment with regard to the dependent variables relying on data from the GDHS. Logistic regression estimates that examined men’s attitudes toward wife-beating for any one of the five scenarios or reasons were also presented. For each dependent variable, regression estimates are presented for the adjusted effects of bridewealth payment on attitudes toward wife- beating. 125 University of Ghana http://ugspace.ug.edu.gh 6.2 Bridewealth negotiation status and men’s attitudes toward wife-beating Table 6.1 shows adjusted logistic regression analyses for 5 models that examine the relationship between whether or not bridewealth has been negotiated and various scenarios for which wife-beating was justified controlling for some selected socio-demographic characteristics. The results show that for the various scenarios examining men’s attitudes toward wife-beating, some control variables show similar associations with the dependent variables. The main independent variable that is bridewealth negotiation status was not significantly associated with men’s attitudes toward wife-beating. Consistently, men who indicated that they were affiliated with ‘Other’ religion were more than twice as likely as their Christian counterparts to approve of wife-beating with regard to all five scenarios. Concerning approval of wife-beating on at least one reason, men who stated that they have no education were 1.7 times as likely as those with junior secondary education to approve of wife-beating. Also, men with no formal education were 2.26 times as likely as those with junior secondary education to have approved of wife-beating if a woman refuses to have sex with her husband. With the exception of attitudes toward wife-beating if a woman refuses to have sex with her husband, type of earning was significantly associated with men’s attitudes toward wife-beating for the other four scenarios under consideration. Men who earned cash only were less likely to approve of wife-beating. As age increases, the odds of supporting wife- beating on at least one reason decrease. There is a significant relationship between ethnicity and men’s attitudes about wife-beating. Compared with Akan men, men who identify with the “Other” ethnic group were more likely to approve of wife-beating in the instance where a woman goes out without telling her husband and neglects the children. 126 University of Ghana http://ugspace.ug.edu.gh Table 6. 1: Odds ratios (and Confidence Intervals) from binary logistic regression models of approval of wife-beating by bridewealth negotiation status and selected characteristics, among men in Ghana Goes out Neglects Argues with Refuses to have Any one without telling the children her husband sex with her reason her husband husband Respondent’s Odds ratios Odds ratios Odds ratios Odds ratios Odds ratios characteristics (CI) (CI) (CI) (CI) (CI) Bridewealth negotiation status [Negotiated] Not negotiated 0.83(0.42-1.65) 0.85(0.45-1.62) 0.68(0.22-1.42) 0.74(0.35-1.54) 0.99(0.60-1.63) Religious affiliation [Christian] Moslem 0.75(0.39-1.46) 0.59(0.30-1.14) 1.14(0.59-2.19) 1.15(0.61-2.14) 0.83(0.48-1.40) Other 2.38**(1.42-3.99) 2.27**(1.36-3.81) 2.38**(1.31-4.34) 2.84*(1.62-4.97) 2.11**(1.35-3.29) Level of education [Junior secondary] No education 1.44(0.73-2.83) 1.43(0.79-2.60) 1.28(0.62-2.65) 2.26*(1.13-4.55) 1.70*(1.02-2.84) Primary 1.22(0.60-2.50) 1.27(0.69-2.36) 1.58(0.76-3.28) 1.80(0.85-3.80) 1.37(0.78-2.39) Secondary/higher 0.64(0.30-1.34) 0.57+(0.29-1.10) 0.88(0.42-1.84) 0.80(0.31-2.10) 0.65(0.36-1.15) Type of earning [Not paid] Cash only 0.34*(0.15-0.78) 0.38*(0.18-0.79) 0.28**(0.13-0.62) 0.62(0.31-1.26) 0.36**(0.20-0.64) Paid in cash and in-kind 1.02(0.47-2.21) 0.56(0.25-1.26) 0.66(0.31-1.42) 1.69(0.77-3.71) 0.62(0.34-1.11) Age 0.97(0.93-1.01) 0.98(0.93-1.13) 1.00(0.96-1.04) 1.01(0.96-1.06) 0.96*(0.92-0.99) Duration of marriage 1.01(0.97-1.06) 1.01(0.97-1.06) 1.00(0.96-1.04) 0.98(0.94-1.02) 1.03(0.99-1.065) Number of children ever born [1 to 4 children] No child 0.69(0.33-1.48) 0.66(0.32-1.38) 0.44(0.16-1.21) 1.07(0.46-2.48) 0.59(0.32-1.05) 5 or more children 1.34(0.79-2.26) 1.04(0.0.63-1.72) 1.08(0.64-1.80) 1.14(0.61-2.13) 1.05(0.69-1.59) “*p<0.05;**p<0.01;***p<0.001” Source: 2014 Ghana Demographic and Health Survey []: Reference category, CI: Confidence Interval 127 University of Ghana http://ugspace.ug.edu.gh Table 6.1 (Continued): Odds ratios (and Confidence Intervals) from binary logistic regression models of approval of wife- beating by bridewealth negotiation status and selected characteristics, among men in Ghana Goes out without Neglects Argues with Refuses to Any one telling her the children her husband have sex with reason husband her husband Respondent’s Odds ratios Odds ratios Odds ratios Odds ratios Odds ratios Characteristics (CI) (CI) (CI) (CI) (CI) Household wealth [Poor] Middle 0.91 (0.39-2.13) 1.04 (0.50-2.17) 1.03 (0.45-2.36) 0.51 (0.20-1.26) 0.88 (0.26) Rich 0.79 (0.28-2.23) 0.80 (0.32-1.98) 1.23 (0.48-3.12) 0.57 (0.25-1.32) 0.77 (0.26) Place of residence [Urban] Rural 0.63 (0.30-1.31) 0.86 (0.044-1.69) 0.76 (0.38-1.50) 1.11 (0.54-2.28) 0.73 (0.19) Ethnicity [Akan] Ga-Dangbe and Ewe 0.83 (0.36-1.88) 0.81 (0.36-1.82) 1.06 (0.49-2.26) I 0.73 (0.25) Mole-Dagbani 1.12 (0.55-2.29) 1.40 (0.65-3.00) 1.25 (0.59-2.65) I 1.01 (0.33) Other 2.32* (1.13-4.77) 3.27** (1.60-6.68) 1.51 (0.64-3.58) 0.90(0.45-1.77) 2.05 (0.62) Type of marriage [Monogamous] Polygynous 1.67(0.88-3.14) 1.79 (0.96-3.34) 1.70 (0.81-3.57) 1.51 (0.82-2.80) 1.33 (0.35) Sample size 1893 1893 1893 1893 1893 “*p<0.05;**p<0.01;***p<0.001” Source: 2014 Ghana Demographic and Health Survey []: Reference category, CI: Confidence Interval. I: included in the ‘Other’ category due to the small cases in the categories. 128 University of Ghana http://ugspace.ug.edu.gh 6.3 Completeness of bridewealth payment and other correlates of men’s attitudes toward wife-beating In Table 6.2, binary logistic regression estimates examining the association between completeness of bridewealth payment and men’s attitudes towards wife-beating in Ghana are presented. Adjusted odds ratios from 5 models that examine the relationship between completeness of bridewealth payment and men’s attitudes toward “wife-beating if a woman goes out without telling her husband, neglects the children, argues with her husband, refuses to have sex with her husband and on least one reason”, adjusting for selected socio-demographic and economic characteristics are shown. The results show that for the various scenarios for approval of wife-beating examined, some control variables show fairly similar associations with the dependent variables. The main independent variable, that is, completeness of bridewealth payment was not significantly associated with men’s attitudes toward wife-beating. For all five models presented, religious affiliation was a significant predictor of men’s attitudes towards wife-beating. Men who indicated that they were affiliated with the “Other’ religious group were more likely to approve of wife-beating. For example, men who indicated that they were affiliated with “Other” religions were 2.42 times as likely to approve of wife- beating if a woman goes out without telling her husband as men who stated that they were Christians. Education was significantly linked with men’s approval of wife-beating if a woman refuses to have sex with her husband and on at least one reason. Men with no formal education were more likely to approve of wife-beating compared to men with junior secondary education if a woman refuses to have sex with her husband (OR=2.25, p-value<0.05) and on at least one reason (OR=1.72, p-value<0.05). 129 University of Ghana http://ugspace.ug.edu.gh With regard to type of earning, the results showed that compared with men who earned no income, earning cash only reduces the likelihood of men supporting wife-beating. For instance compared with men who earned no income, earning cash only reduces the likelihood of men’s approval of wife-beating by 71 percent (p-value <0.01). Type of earning was not significantly associated with men’s attitude toward wife-beating if a woman refuses to have sex with her husband. With every unit increase in age, men were 0.04 times less likely to approve of wife-beating on at least one reason. Ethnicity was significantly associated with attitudes toward wife-beating except for when a woman refuses to have sex with her husband or argues with her husband. Compared with Akan men, men who identified with the ‘Other’ ethnic group were 2.08 times as likely to justify wife- beating on at least one reason. Consistently, number of children ever born, household wealth, place of residence, duration of marriage, and type of marriage were not statistically significant predictors of men’s attitudes toward wife-beating. 130 University of Ghana http://ugspace.ug.edu.gh Table 6. 2: Odds ratios (and Confidence Intervals) from binary logistic regression models of approval of wife-beating by completeness of bridewealth payment and selected characteristics, among men in Ghana Goes out without telling her Neglects the Argues with her Refuses to have sex with her Any one husband children husband husband reason Respondent’s Odds ratios Odds ratios Odds ratios Odds ratios Odds ratios characteristics (CI) (CI) (CI) (CI) (CI) Completeness of bridewealth payment [Fully paid] Partially paid 0.97 (0.53-1.79) 1.27(0.72-2.23) 0.70(0.37-1.35) 0.76(0.36-160) 1.04(0.63-1.72) Not negotiated/paid 1.05(0.57-1.93) 1.09(0.61-1.95) 0.98(0.52-1.82) 0.76(0.37-1.54) 1.25(0.79-1.97) Religious affiliation [Christian] Moslem 0.75(0.39-1.45) 0.60(0.31-1.16) 1.11(0.58-2.12) 1.12(0.31-2.08) 0.83(0.49-1.41) Other 2.42**(1.44-4.08) 2.29**(1.39-3.86) 2.44**(1.33-4.49) 2.85***(1.62-5.02) 2.14**(1.37-3.35) Level of education [JSS] No education 1.45(0.74-2.84) 1.45(0.81-2.62) 1.27(0.62-2.61) 2.25*(1.12-4.50) 1.72*(1.04-2.84) Primary 1.23(0.60-2.52) 1.25(0.67-2.32) 1.61(0.78-3.35) 1.84(0.88-3.85) 1.37(0.78-2.40) Secondary/higher 0.63(0.30-1.34) 0.56(0.29-1.08) 0.86(0.41-1.82) 0.80(0.31-2.0) 0.64(0.36-1.14) Type of earning [Not paid] Cash only 0.34*(0.15-0.78) 0.37**(0.18-0.77) 0.29**(0.13-0.64) 0.63(0.31-1.28) 0.36**(0.20-0.64) Paid in cash and in-kind 1.00(0.46-2.19) 0.55(0.24-1.23) 0.66(0.31-1.39) 1.72(0.78-3.81) 0.60+(0.33-1.10) Age 0.97(0.93-1.01) 0.98(0.93-1.03) 1.00(0.96-1.04) 1.01(0.96-1.06) 0.96*(0.92-0.99) Duration of marriage 1.01(0.97-1.06) 1.02(0.97-1.06) 1.00(0.96-1.04) 0.98(0.94-1.02) 1.03(0.99-1.07) “*p<0.05;**p<0.01;***p<0.001” Source: 2014 Ghana Demographic and Health Survey 131 University of Ghana http://ugspace.ug.edu.gh Table 6.2 (Continued): Odds ratios (and Confidence Intervals) from binary logistic regression models of approval of wife- beating by completeness of bridewealth payment and selected characteristics, among men in Ghana Goes out without Neglects the Argues with her Refuses to have sex Any one telling her husband children husband with her husband reason Respondent’s Odds ratios Odds ratios Odds ratios Odds ratios Odds ratios characteristics (CI) (CI) (CI) (CI) (CI) Number of children ever born [1 to 4 children] No child 0.68(0.32-1.46) 0.65(0.31-1.37) 0.43(0.15-1.18) 1.07(0.47-2.45) 0.58(0.32-1.04) 5 or more children 1.34(0.79-2.27) 1.06(0.64-1.75) 1.07(0.64-1.80) 1.12(0.60-2.10) 1.06(0.70-1.60) Household wealth [Poor] Middle 0.92(0.40-2.13) 1.05(0.51-2.19) 1.02(0.45-2.34) 0.50(0.20-1.24) 0.89(0.50-1.59) Rich 0.80(0.28-2.25) 0.83(0.33-2.08) 1.19(0.47-2.96) 0.55(0.24-1.25) 0.78(0.40-1.51) Place of residence [Urban] Rural 0.62(0.30-1.29) 0.86(0.44-1.68) 0.76(0.39-1.49) 1.11(0.54-2.27) 0.72(0.43-1.22) Ethnicity [Akan] Ga-Dangbe and Ewe 0.83(0.37-1.90) 0.79(0.35-1.79) 1.09(0.51-2.35) I 0.73(0.37-1.44) Mole-Dagbani 1.14(0.56-2.34) 1.40(0.66-3.01) 1.31(0.63-2.75) I 1.02(0.54-1.95) Other 2.37*(1.16-4.87) 3.27**(1.61-6.67) 1.58(0.67-3.73) 0.91(0.46-1.80) 2.08*(1.15-3.77) Type of marriage [Monogamous] Polygynous 1.69(0.89-3.19) 1.83(0.98-3.41) 1.72 (0.81-3.66) 1.50 (0.81-2.81) 1.35(0.80-2.28) Sample size 1893 1893 1893 1893 1893 “*p<0.05;**p<0.01;***p<0.001” Source: 2014 Ghana Demographic and Health Survey []: Reference category, CI: Confidence Interval. I : included to the Other category due to small sample size. 132 University of Ghana http://ugspace.ug.edu.gh 6.4 Discussion The aim of this chapter was to examine the association between bridewealth payment and intimate partner violence measured as men’s attitudes toward wife-beating. Other socio- demographic and economic predictors of men’s attitudes toward wife-beating were controlled for. The specific domains are attitudes toward “wife-beating if a woman goes out without telling her husband, or neglects the children, or argues with her husband, or refuses to have sex with her husband”. These domestic situations are related to married women’s responsibilities as wives, and traditional and gendered socialization norms define these roles for women (Nukunya 2003). Relying on the integrated ecological framework, the results show the importance of examining the implications of both individual and community level variables on men’s attitudes towards wife-beating. In this study, both individual and community level variables significantly predicted the men’s approval of wife-beating. The results show that religious affiliation, age, level of education, type of earning, number of children ever born were some significant predictors of their approval of wife-beating. With regards to community level variables, ethnicity and type of marriage predicted men’s approval of wife-beating. Concerning the two measures of bridewealth (whether or not bridewealth has been negotiated and completeness of bridewealth payment), the results suggest that bridewealth was not statistically significantly associated with men’s attitudes toward wife-beating. The current result is contrary to the only study that examined completeness of bridewealth payment and women’s attitudes toward wife-beating if a woman disregards her husband’s reproductive preferences (Horne et al., 2013). In their study, participants expected that the woman’s kinsmen, women in the locality and participants (women) themselves would approve of wife-beating if the 133 University of Ghana http://ugspace.ug.edu.gh woman took autonomous reproductive decisions. It is possible that the measures of attitudes toward wife-beating in the GDHS do not fully encompass issues that relate to bridewealth payment. Would we expect men to have more permissive attitudes about wife-beating if a woman aborts their pregnancy or uses contraception without telling her husband versus if she burns the food or goes out without telling him? Another plausible explanation is that due to the small proportions of men who reported that wife-beating is acceptable, we are unable to observe any statistical significance with regard to bridewealth payment and attitudes toward wife-beating. Religious affiliation was significantly associated with men’s attitudes toward “wife- beating if a woman goes out without telling her husband, or neglects the children, or argues with her husband, or refuses to have sex with her husband”. Also, religious affiliation predicted attitudes toward wife-beating for any one of the five scenarios. Significantly, men who indicated that they were affiliated with ‘Other’ religions were more likely to approve of wife-beating for all five dependent variables compared to their Christian counterparts. Rani et al. (2004) and Djamba, Kimuna, & Aga (2015) found that in Ethiopia, men who were affiliated with ‘18other’ religions were more likely to approve of wife-beating compared to Catholic and Orthodox men respectively. In Ghana, Takyi and Mann (2006) found no significant association between religion and men’s approval of wife-beating. Adherence to strict conservative views with regard to women’s roles as wives and mothers may increase the likelihood of men believing that wife- beating is justified if a woman contravenes these traditional or religious socially determined roles. 18 "In the Rani and Bonu (2004) study, the "Other" category includes 'traditional', 'no-religion' or 'animist' depending on the country. This is similar to the categorization of the “Other” in the present study. 134 University of Ghana http://ugspace.ug.edu.gh The results show that men who have no education or only primary education were significantly more likely to approve of wife-beating if a woman refuses to have sex with her husband and on at least one of the five measures wife-beating compared to men who have junior secondary school education. Prior studies have also shown that education is inversely associated with men’s approval of wife-beating (Djamba et al., 2015; Rani & Bonu, 2009; Takyi & Mann, 2006; Uthman et al., 2009; Wang, 2016). Possibly, access to formal education tends to change men’s attitudes toward more liberal and egalitarian perspectives than traditional views, thus, lowering their acceptance of violence against women. Contrary to these studies some studies have found no significant association between men’s level of education and their attitudes toward wife-beating (Speizer, 2010). Further, Lawoko (2008) found that while education was negatively associated with men’s approval of wife-beating in Kenya, in Zambia, the results show that with increasing education men were significantly more likely to support wife-beating. The author argues that differences in contexts and perhaps, content of educational materials in these two countries could mediate the observed differences between education and men’s attitudes toward wife-beating. Type of earning was associated with men’s approval of “wife-beating if a woman goes out without telling her husband, argues with her husband, neglects the children, and on at least one of the five reasons for approval of wife-beating”. Men who earned cash only were significantly less likely to approve of wife-beating. Access to paid employment reduces men’s acceptance of violence than not working or not earning cash. Studies in other countries (Benin and Uganda) have shown that compared to men who were not working, men who were engaged in unpaid jobs were less likely to believe that wife-beating is justified (Rani & Bonu, 2009). Having access to jobs that paid cash may suggest that men have resources that would possibly 135 University of Ghana http://ugspace.ug.edu.gh reduce the levels of strain in households, giving them the financial ability to provide for their families. These results show that access to both social resources (higher level of education) and economic resources (earning cash) have implications for men’s approval of wife-beating. According to the resource theory access to resources in this case higher level of education and earning cash potentially reduce men’s approval of violence since they would not have to rely on violence as alternative resources. Access to these resources could imply these men have the resources need to gain the needed compliance from their partners (Goode 1971). Some studies have typically found that younger men are more likely to approve of wife- beating than older men (Lawoko, 2008; Rani & Bonu, 2009; Rani et al., 2004; Speizer, 2010) . Consistent with these studies, and similar studies in Ghana (Takyi & Mann, 2006), the current study argues that with every unit increase in age, men were less likely to approve of wife-beating on at least one reason. Other studies have also found that younger women were significantly more likely to approve of wife-beating than their colleagues who were older (Hindin, 2003; Speizer, 2010; Tayyab, Kamal, Akbar, & Zakar, 2017). To understand why young people have negative attitudes about wife-beating, it is important to examine some other socio-demographic and context specific characteristics, specifically their place of residence and level of education. For example, some studies have found that young persons who reside in rural areas, are less educated, and also, those who belong to poor households are more likely to approve of wife- beating (Dalal et al., 2012). These results are alarming because they suggest that younger people adhere to traditional views about disciplining women when they do not perform traditional domestic duties as wives. With a growing young population in Ghana, and particularly in sub- 136 University of Ghana http://ugspace.ug.edu.gh Saharan Africa, efforts that seek to improve attitudes toward wife-beating especially among young men, should be prioritized. Since the use of violence perhaps occurred earlier in the relationship, older men may forget to report such occurrences or may plainly decide to under- report such violent behaviours. This is because they may not want to be seen as violent old men. On the contrary, some studies have also found that older men are more likely to be abusive (Fleming et al., 2015). These reasons could explain the results obtained on age. Past studies have found different results with regard to number of children ever born and men’s attitudes toward wife-beating, and this may be mainly due to variations in how number of children ever born is measured. Some studies have measured number of children ever born by examining the composition of males and females ever born (Rani & Bonu, 2009), and found no significant association with men’s attitudes toward wife-beating. Other studies have measured number of children ever born using the continuous form of the variable (Takyi & Mann, 2006) or by categorizing the variable (Speizer, 2010). In Ghana, for example, Takyi and Mann (2006) found that with every additional child born, men were more likely to approve of wife-beating. Also, in Uganda, men who have 0 to 2 children were significantly less likely to approve of wife- beating compared with those who have 5 and more (Speizer, 2010). In this study, the results show that there is a marginal association between number of children ever born and men’s attitudes toward wife-beating on at least one reason. Men who have no children are less likely to approve of wife-beating compared to men who have 1 to 4 children. Ethnicity was significantly associated with all scenarios of wife-beating except for approval of wife-beating if a woman argues with her husband. Also, for approval of wife-beating if a woman refuses to have sex with her husband, ethnicity was not significantly associated with 137 University of Ghana http://ugspace.ug.edu.gh wife-beating. Generally, men who belong to the ‘Other’ ethnic group were more likely to approve of wife-beating compared to Akan men. Men who belong to the ‘Other’ ethnic group were typically patrilineal men in some parts of the Northern Region of Ghana and those who identified as Guan. Takyi and Mann (2006) found varying links between ethnicity and men’s approval of wife-beating. In their study, the results show that while Mole-Dagbani men were more likely to approve of wife-beating, men who indicated that they were Ewe were less likely to approve of wife-beating compared to Akan men. They argue that these observed differences potentially reflect ethnic differences in socialization and bridewealth payment practices in various parts of the country. Type of marriage has been found to be significantly associated with men and women’s attitudes toward wife-beating in some countries (Rani et al., 2004). In this study, we find no significant support for this relationship. In all the logistic regression models presented, household wealth, place of residence, and duration of marriage were not statistically significant predictors of men’s attitudes toward wife- beating. 138 University of Ghana http://ugspace.ug.edu.gh CHAPTER SEVEN BRIDEWEALTH PAYMENT AND PERPETRATION OF INTIMATE PARTNER VIOLENCE BY MEN 7.1 Introduction In this chapter, the relationship between bridewealth payment and IPV perpetration by men is examined relying on data from the Bridewealth Study. The incidence of physical abuse perpetration relies on self-reported accounts from men. Selected socio-demographic and economic characteristics of men are controlled for. The measures of bridewealth payment are; (a) bridewealth negotiation status and (b) completeness of bridewealth payment. Table 7.1 and Table 7.2 show odds ratios from binary logistic regression models for bridewealth negotiation status, completeness of bridewealth payment, and having ever perpetrated physical abuse respectively. 7.2 Bridewealth negotiation status and having ever perpetrated intimate partner physical abuse In Table 7.1, results from binary logistic models are presented. Generally, there is no significant association between bridewealth negotiation status and having ever perpetrated physical abuse against one’s female partner (Table 7.1, Models 1). The results show that religious affiliation, level of education, age, duration of marriage, observing violence as a child and lineage were significant predictors of perpetration of physical abuse. The Pseudo R-squared estimate for this model is 0.088. 139 University of Ghana http://ugspace.ug.edu.gh Men who were affiliated with Pentecostal/Charismatic religions were 0.43 times less likely to have ever perpetrated physical abuse compared with Orthodox Christian men. Compared to men with junior secondary school education, men who stated that they had secondary/higher education were 0.39 times less likely to have ever perpetrated physical abuse (p-value<0.05). As age increases, men were significantly less likely to have perpetrated physical violence against their female partners (OR=0.95; p-value=0.000). It appears that duration of marriage was positively associated with having ever perpetrated physical abuse such that, with a unit increase in duration of marriage, men were significantly more likely to have perpetrated abuse. Compared with men who had never observed violence between their parents, men who observed violence between their parents/caregivers had a higher risk of having ever perpetrated violence against an intimate partner (OR=1.89; p-value<0.05). Men who identified with Patrilineal Guan groups were over two times as likely to have ever perpetrated physical abuse compared with those who identified with Matrilineal Akan groups. The results showed that employment status and number of children ever born were not significant predictors of having ever perpetrated physical abuse against one’s female partner. 140 University of Ghana http://ugspace.ug.edu.gh Table 7. 1: Odds ratios (and Confidence Intervals) from binary logistic regression models of perpetrating physical abuse by bridewealth negotiation status and selected characteristics, among men Respondent’s characteristics Model 1 Odds ratios(CI) Bridewealth negotiation status [Negotiated] Not negotiated 1.44(0.71-2.93) Religious affiliation [Orthodox] Pentecostals/Charismatics 0.57*(0.35-0.92) Other 1.18(0.63-2.20) Level of education [Junior secondary] No education 0.52(0.21-1.12) Secondary/higher 0.61*(0.15-0.99) Employment status [Not employed] Employed 1.13(0.57-2.24) Age 0.95***(0.93-0.98) Duration of marriage 1.04*(1.01-1.07) Number of children ever born [2 to 4 children] 0 to1 children 0.59(0.30-1.17) 5 or more 0.93(0.53-1.64) Observed violence as a child [No] Yes 1.89*(1.15-3.10) Lineage [Matrilineal Akan] Matrilineal Guan 1.61(0.87-2.99) Patrilineal Guan 2.12*(1.17-3.85) Patrilineal Akan 1.14(0.60-2.17) Sample size 579 Pseudo R-squared 0.088 “*p<0.05;**p<0.01;***p<0.001” Source: 2016 Bridewealth Study (RIPS) []: Reference category, CI: Confidence Interval 7.3 Completeness of bridewealth payment and having ever perpetrated intimate physical abuse Table 7.2 shows the multivariate logistic regression estimates for completeness of bridewealth payment and having ever perpetrated physical abuse. Model 1 provides the adjusted odds ratios from logistic regression estimates showing the association between completeness of bridewealth payment and having ever perpetrated abuse, controlling for religious affiliation, 141 University of Ghana http://ugspace.ug.edu.gh level of education, employment status, age, duration of marriage, observing violence as a child and ethnicity. The Pseudo R-squared estimate for this model is 0.102. The adjusted logistic regression estimates indicate that when bridewealth has not been paid at all, men were 2.60 times as likely to have ever perpetrated physical abuse against their female partners compared to when it has been fully paid (p-value=0.003). The results show that Pentecostal/Charismatic men were significantly less likely to have ever beaten their wives (OR=0.55; p-value<0.05) compared to Orthodox men (Table 7.2). There was no statistically significant association between level of education and self-reported actual perpetration of violence against women. Relying on data from Uganda, Speizer (2010) found no statistically significant association between education and men’s use of violence in intimate unions. With every unit increase in age, men were 0.95 times as likely to have perpetrated physical abuse. There is a direct relationship between duration of marriage and having ever perpetrated abuse against one’s female partner. Men who reported that they have one child or none were 0.49 times as likely as those with 2 to 4 to have ever perpetrated physical abuse. The odds of physical abuse were higher for men who had ever witnessed violence between their guardians than those who have never. Men who indicated that they had ever witnessed violence between their guardians were 1.79 times as likely to have ever perpetrated physical abuse as the reference group, those who had never observed violence between their guardians. Men who identified as Matrilineal Guan and Patrilineal Guan were 1.92 and 2.31 times as likely as their Matrilineal Akan counterparts to have ever perpetrated physical abuse. The results suggest that employment status does not predict abuse. 142 University of Ghana http://ugspace.ug.edu.gh Table 7. 2: Odds ratios (and Confidence Intervals) from binary logistic regression models of perpetrating physical abuse by completeness of bridewealth payment and selected characteristics, among men in Ghana Respondent’s characteristics Model 1 Odds Ratios(CI) Completeness of bridewealth payment [Full payment] Partial payment 1.22 (0.69-2.16) No payment 2.60** (1.39-4.86) Religious affiliation [Orthodox] Pentecostal/charismatic 0.55* (0.34-0.90) Other 1.19 (0.63-2.22) Level of education [Junior secondary] No education/primary 0.51 (0.24-1.12) Secondary/higher 0.62(0.38-1.01) Employment status [Not employed] Employed 1.22 (0.61-2.44) Age 0.95*** (0.93-0.98) Duration of marriage 1.04** (1.01-1.08) Number of children ever born [2 to 4 children] 0 to1 child 0.49* (0.25-0.99) 5 or more 0.94 (0.53-1.67) Observed violence as a child [No] Yes 1.79* (1.09-2.96) Lineage [Matrilineal Akan] Matrilineal Guan 1.92* (1.01-3.64) Patrilineal Guan 2.31** (1.26-4.23) Patrilineal Akan 1.37 (1.26-4.23) Sample size 579 Pseudo R-squared 0.102 “*p<0.05;**p<0.01;***p<0.001” Source: 2016 Bridewealth Study (RIPS) []: Reference category, CI: Confidence Interval 143 University of Ghana http://ugspace.ug.edu.gh 7.4 Discussion The aim of this chapter was to investigate the relationship between bridewealth payment and physical abuse perpetration by men. The results show that bridewealth negotiation status was not a predictor of physical abuse by men. Previous research has shown that with regard to bridewealth payment, right to women’s sexual and domestic services are only transferred when bridewealth has been paid (Dodoo, Horne, & Biney, 2014; Dodoo & Dodoo, 2017). In the instance where the families of the bride and groom only negotiated bridewealth (that is the families of both the man and the woman or the couple themselves played a role in arriving at the bridewealth (Ghana Statistical Service, 2014, p. 99)), it is possible that there is virtually no binding contract between the couple or their families although they are living together and both families are potentially aware of union. The results, however, show that completeness of bridewealth payment significantly predicts physical abuse perpetration by men. Compared to men who indicated full bridewealth payment, men who stated that bridewealth has not been paid in their current union were more likely to have ever been violent toward their wives. These results suggest that legitimizing marriage through bridewealth payment potentially reduces the risk of perpetrating abuse against wives. Bridewealth was traditionally paid to strengthen kinship ties, protect women and legitimize unions (Aborampah, 1999; Anderson, 2007). Aborampah (1999) argues that bridewealth payments were made to ensure stable unions and good behaviour from all parties involved. Indeed, anecdotal evidence also suggests that at marriage ceremonies grooms are warned never to lay hands on their wives. Members of the bride’s family specifically state that the exchange of bridewealth does not imply that the woman should be abused. Thus, it is possible that these traditional principles of bridewealth payment protect women from physical 144 University of Ghana http://ugspace.ug.edu.gh abuse. The husband to be is informed of the consequences of being violent and this may reduce the chances of violent acts. Further, if full payment is associated with rights transferred, the consequence, then, is that both parties may be more likely to act in ways that support this transfer, by first, acknowledging the authority they have in the union, and secondly, discharging their bridewealth specified roles and responsibilities. When an individual gives to a member of the network, in this case the husband and his family present bridewealth to the bride and her family, the exchange specifies roles and responsibilities. According to the exchange/social control model when the giver does not obtain the needed reciprocal services (in this case rights to the woman’s sexual, reproductive and domestic services), he may be more likely to use violence (Gelles, 1983). The results presented in Table 7.2 potentially indicate that if the receiver performs the needed reciprocal services it could avert various forms of sanction and abuse. It is important to note that this could put a lot of pressure on the bride to perform her bridewealth specified roles and responsibilities. The results of this study suggest that living together as though married, and having no bridewealth paid could rather provoke disagreements between couples and potentially lead to men’s use of violence. Bridewealth payment is necessary and legitimizes unions, recognizing both men and women as socially responsible individuals (Aborampah, 1999). Bridewealth payment requires that men pay some amount of wealth (cash) or other tangible goods to the bride’s kinsmen (Aborampah, 1999). According to this resource theory, if men lack economic resources, they are more likely to use violence as an “alternative resource” (Goode, 1971). It can be implied that men who are unable to pay bridewealth to some extent lack these economic 145 University of Ghana http://ugspace.ug.edu.gh resources and may resort to the use of violence against their partners to stamp their authority. Where no payments are made, family members of the couples, especially the woman’s family and couples themselves may disagree to such arrangements. It is possible that disagreements may arise when women insist that men make bridewealth payment. Societal pressure to perform this cultural practice which ushers individuals into adulthood may increase men’s distress and provoke their use of violence. These results show that the implications of this cultural practice on men’s use of violence in intimate unions are complex, and challenge the linear examination of bridewealth payment and how it is associated with violence perpetration, experience and couples well-being in general. Religious affiliation, level of education, age, duration of marriage, observing violence between parents as a child, and lineage were significantly associated with having ever perpetrated abuse. Concerning actual perpetration of abuse, men who indicated that they were affiliated with ‘Other’ religious groups were more likely to have ever perpetrated abuse. It should be pointed out again that men who belong to ‘Other’ religious groups are mainly traditional religious practitioners and those who stated that they have no religion. Observing violence between one’s caregivers/parents as a child is one of the most consist predictors of perpetrating violence against one’s intimate partner later on in life. Studies have shown that men who witness such forms of violence in their homes as children were more likely to be perpetrators of violence compared to their colleagues who did not witness such forms of family aggression (Abramsky et al., 2011; Bucheli & Rossi, 2017; Dery & Diedong, 2014; Kishor & Bradley, 2012; Peralta et al., 2010; World Health Organization, 2017; Yount et al., 2014). Intergenerational transmission of violent behaviours has implications for men’s use of 146 University of Ghana http://ugspace.ug.edu.gh violence against their intimate partners. Consistent with these studies, the present study shows that men who reported that they had observed violence between their guardians also stated that they had ever beaten their wives compared to those who stated that they had never observed such forms of exchanges between their parents. Observing violence as a child is embedded in the general socialization of children, and it reflects the norms and perceptions of violence acceptance in a given context (Bandura, 1973; Namy et al., 2017). For example, in their mixed method study, respondents who had been recruited as perpetrators of IPV recounted that they conceptualized violence they witnessed in their families as children as normal (Peralta et al., 2010). Thus, as adults they found themselves using such acts against their intimate partners. With regard to men’s level of education, some statistically significant associations with physical abuse were observed. Compared to men with junior secondary education, men with secondary/higher education were less likely to have perpetrated abuse. Prior studies have shown that with increasing education the risk of perpetrating abuse against intimate partner violence decreases (Johnson & Das, 2009). Contrary to these studies, some studies have found no consistent association between education and abuse against women (Fleming et al., 2015). Relying on data from eight developing countries, Fleming et al. (2015) highlighted that while education was not generally significantly associated with physical intimate partner abuse, in some country samples, increasing education seemed to be associated with IPV. In the Democratic Republic of Congo, the results suggest that with increasing education men were more likely to be perpetrators of physical abuse (Fleming et al., 2015). The inconsistent association of men’s schooling and physical intimate violence perpetration is similar to women's access to education and self-reported experience of male-perpetrated physical abuse. While some research have found that women with secondary or higher training are less likely to experience 147 University of Ghana http://ugspace.ug.edu.gh partner abuse (Kapiga et al., 2017; McCloskey et al., 2005; Tenkorang et al., 2013), others found that women with higher levels of schooling are at higher risk of experiencing male-perpetrated abuse (Adebowale, 2018; Nwabunike & Tenkorang, 2017; Takyi & Lamptey, 2016). Further, some studies found no statistically significant link between women’s level of education and IPV (Gibbs, Corboz, & Jewkes, 2018; Klomegah, 2008). These varying results possibly reflect differences in study samples or context and measurement of levels of education in these studies. Reported number of children ever born was significantly associated with physical violence. Men who have one or no child have lower odds of violence perpetration compared with those with 2 to 4 children. The results compare with other studies where perpetration of abuse increases with number of children ever born (Speizer, 2010; Weitzman, 2014). In Uganda, for instance, men who have three or more children were more likely to have perpetrated spousal abuse than men who had less than three children (Speizer, 2010). It is possible that the stress of providing for large family sizes could provoke men’s use of violence against intimate partners (Weitzman, 2014). Age and duration of marriage were significantly associated with having ever perpetrated physical abuse. With every unit increase in age, men were less likely to have reported that they have ever beaten their wives. The results corroborate findings in Bangladesh and Tanzania where researchers found that the youngest male partners were more likely to have reported being violent (Johnson & Das, 2009; Mulawa, et al., 2018). Concerning attitudes toward wife-beating also, the study results indicate that younger persons are more likely to support wife-beating if a woman refuses to have sex with her husband and on at least one of the five reasons for wife- beating. These results show that to some extent young men have more favourable attitudes 148 University of Ghana http://ugspace.ug.edu.gh toward wife-beating and they may also be violent toward their partners in Ghana. It is unclear why young people are more likely to approve of IPV and actually be violent. It is likely that the stress of managing new or young relationships place these young men at some level of distress. Another plausible explanation is that since the use of violence perhaps occurred earlier in the relationship, older men may forget to report such occurrences or may plainly decide to under- report such violent behaviours. This is because they may not want to be seen as violent old men. On the contrary, some studies have also found that older men are more likely to be abusive (Fleming et al., 2015). Duration of union is positively linked with physical abuse against one’s female partner. The relationship between duration of marriage and physical abuse is contrary to what has been documented in other studies (Johnson & Das, 2009). Since the perpetration of violence relates to lifetime perpetration, it is possible that the observed relationship between duration and physical abuse could be experiences that occurred earlier in the relationship than due to the length of stay in the relationship. This is because in this sample, the results show that young men were more likely to have been violent. Finally, the results show that lineage was a significant predictor of physical abuse. Men who identified as Matrilineal Guan and Patrilineal Guan were more likely to have ever perpetrated physical abuse compared with Matrilineal Akan men. Previous studies have shown that lineage is a predictor of male-perpetrated violence against women. Using in-depth interviews from some parts of the Eastern Region of Ghana, Sedziafa, Tenkorang, & Owusu (2016) found that although women who identified with matrilineal and patrilineal kin groups reported various forms of male-perpetrated violence, all patrilineal women reported that they 149 University of Ghana http://ugspace.ug.edu.gh experienced all forms of violence. Studies have shown that differences in social organization and bridewealth payment in these kin groups have implications for women’s well-being (Takyi & Dodoo, 2005; Takyi & Gyimah, 2007). In conclusion, the results in this chapter show that completeness of bridewealth payment is indeed associated with male-perpetrated IPV in these areas. The results also show that individual level variables associated with having ever perpetrated abuse are religious affiliation, level of education, number of children ever born, age, duration of marriage, and observing violence as a child. Lineage is the community level variable significantly associated with intimate partner physical abuse. Employment status was not associated with having ever perpetrated physical abuse. 150 University of Ghana http://ugspace.ug.edu.gh CHAPTER EIGHT PERCEPTION OF BRIDEWEALTH PAYMENT AND INTIMATE PARTNER VIOLENCE AMONG THE PARTICIPANTS 8.1 Introduction This section explores key informants’ representations of various pathways through which the practice of bridewealth payment may protect or put women at risk of male-perpetrated violence using data from the Bridewealth Study. The qualitative data examined the benefits of bridewealth payment, implications of non-payment, partial and full payment, and sanctions applied to men for being abusive. In the first part of this chapter, data on the benefits of bridewealth payment, and the implications for violence in intimate unions are examined. The second part explores the negative and positive consequences of non-payment, partial and full bridewealth payment and men’s use or non-use of violence against women in intimate unions. While bridewealth payment is widespread, and legitimizes unions (Anderson, 2007; Wojcicki et al., 2010), it has been established that payment may not always be made completely (Dodoo & Dodoo, 2017). The implications of completeness of bridewealth payment on traditional perceptions of the risk and protective features of such payment regimes are explored. The final part of the chapter explores sanctions men who have paid bridewealth are likely to face if they are violent toward their partners. It is vital to note that the accounts presented here reflect bridewealth payment and its associated practices from selected communities in the Eastern Region of Ghana (in southern Ghana). 151 University of Ghana http://ugspace.ug.edu.gh 8.2 Reasons for or benefits of bridewealth payment Several reasons have been given for why marriage payments are made. These reasons have mainly focused on the commercialization of marriage payment, and how this could potentially put women at risk of abuse (Wendo, 2004). Bridewealth payment potentially constrains women’s reproductive decision making autonomy (Dodoo, Horne, & Biney, 2014; Dodoo & Dodoo, 2017; Horne et al., 2013). Less attention, however, has been given to the social and moral benefits that marriage payments may offer. The in-depth interviews identified some social, moral, and economic 19benefits of bridewealth payment. It could be inferred that these social and moral benefits proffer some protection from abuse. Figure 8.1 represents the thematic network of six benefits of bridewealth payments identified in the study based on the qualitative data. 8.2.1 Honour to parents, wife and husband Honouring one’s parents is an important feature of adulthood, and infers a sense of responsibility. In sub-Saharan African societies, the significance of family, most importantly of parents, cannot be overemphasised. In this study, some respondents noted that bridewealth payment honours parents, and also, the husband and the wife. In the following quotes, some participants stated that bridewealth payment honours one’s parent: 19 Organizing themes and basic themes for the reasons for and benefits of bridewealth payment are presented in Appendix H. 152 University of Ghana http://ugspace.ug.edu.gh “Hmm! In order that it will honour your family. You see. Not that your parents have given birth to you, but they won’t benefit from you. It honours your parents. This woman’s daughter respects herself. Do you understand?” (R11; elderly woman) Figure 8. 1: Thematic network of the reasons for/ benefits of bridewealth payment Abuse Unlawful Parent Husband and divorce wife Protect Honour Property women Woman is Benefits of bridewealth Right to valued inheritance payment Economic Old age Legitimizes Name support/security social the union security Source: 2018 Bridewealth Study (RIPS) The exchange of bridewealth also honours the wife. As suggested in the quote below, this elderly female respondent indicated that marriage payment honours the wife. “It is honourable for the woman. It is honourable to you the woman that a man will come and marry you because that is your glory. You understand. That is your glory. Excuse me, maybe Akwasi said he is coming to marry you and he doesn’t do this and you move to stay with him, it is not honourable.” (R10; elderly woman) 153 University of Ghana http://ugspace.ug.edu.gh The narratives also suggest that when bridewealth is exchanged, the man is expected to follow laid down principles which honour the woman and protect her from harm. “…the purpose is that when a man marries you the woman, he honours you. If all these things are not done, then it means that the man is just treating you anyhow. He treats you anyhow he wants to but if he came for you from your parents, then there are disciplines he needs to ensure”. (R9; elderly woman) Some respondents also stated that bridewealth payment also honours the man. An elderly woman observed that: … oo marriage is an honourable thing, if a man has a wife it is greatly valued”. (R16; elderly woman) 8.2.2 Bridewealth payment legitimizes the union Similar to findings from other researches on the subject, the respondents stated that bridewealth payments are made to legitimize the union, and join the families of the bride and the groom together (Aborampah, 1999; Anderson, 2007). In the quote below, a respondent reiterated this assertion; “…..The reason why it is important is that, without the marriage rite, he is not able to name the child and he went ahead to marry a different woman and they give birth and he performs the other woman’s marriage rite, the first woman’s child is born out of wedlock, the child of the one whose marriage rite was not performed is born out of wedlock. Okay, so if he the father has any property the child does not inherit any. Yes, that child is born out of wedlock” (R6; elderly man) 154 University of Ghana http://ugspace.ug.edu.gh 8.2.3 Women are valued Bridewealth payment shows that the bride to be is valued, not only by the husband to be but by her family. In the quotes below, this elderly man and elderly woman describe the importance of the woman agreeing to marry the man. “…When the woman agrees, oh you thank her for agreeing to marry you.” (R2: elderly man) “….The reason why the marriage ceremony is significant is because you cannot give a woman out as a gift to a man. I hope you have understood me. The man must know that she is also someone’s precious being”. (R9; elderly woman) 8.2.4 Protect women Bridewealth payment offers some protection against abuse and unlawful divorce. Some participants reported that when bridewealth has been paid, and the union duly legitimized, men are less likely to be abusive. “…You see it (bridewealth payment) is a way by which they use to insulate the men from mishandling them (wives) in our Akan system. That is why they say “do not abuse your wife”. You see. All these are done to insulate the woman that you can’t do her evil. Do you understand?” (R2; elderly man) Again, bridewealth payment would prevent or make it more difficult for men to unlawfully, and without going through the proper channels, divorce their wives. “….However, if you are married it is difficult even if the man or the woman does not like it again it is difficult to sack the person unless those who were present during the 155 University of Ghana http://ugspace.ug.edu.gh marriage rites are present and investigated the matter maybe they can even resolve the issue”. (R8; elderly man) 8.2.5 Right to inheritance Bridewealth payment gives the man right to his children (right to name the children and be recognized as their father), and also gives the children right to the man’s properties. “…If you marry the woman it helps the man. It also helps the woman because you are married and it gives the right to have a stake in the property”. (R8; elderly man) “…Eeerrrr it is because of the inheritance that marriage rite is very important. The reason why it is important is that, without the marriage rite, he is not able to name the child and he went ahead to marry a different woman and they give birth and he performs the other woman’s marriage rite, the first woman’s child is a child born out of wedlock, the child of the one whose marriage rite was not performed is born out of wedlock. Okay, so if he the father has any property the child does not inherit any. Yes, that child is born out of wedlock”. (R6; elderly man) Some evidence from the qualitative interviews also suggests that the bridewealth payment meant that the children born to the couple become their source of inheritance. “…the children are his inheritance. The benefit of him marrying you is the children”. (R11; elderly woman) 156 University of Ghana http://ugspace.ug.edu.gh 8.2.6 Economic support/security Receiving economic support were some of the benefits participants reported that women in particular obtain when bridewealth payments are made. Some respondents indicated that, in some cases, the bridewealth exchanged in cash or kind is given to the bride as start-up capital for a trade. “The father gives his daughter the 2 and 6 (26 pesewas) to go and work with it. Go and use the 2 and 6 (26 pesewas) to work. So he has given you ‘dwetire’ [capital]”. (R1; elderly man) 8.2.7 Discussion While previous studies have shown that bridewealth payment could be detrimental to women’s reproductive and sexual autonomy (Dodoo, Horne, & Biney, 2014; Horne et al., 2013), and may put them at risk of male-perpetrated violence (Bawah et al., 1999; Dery & Diedong, 2014; Takyi, 2001; Wendo, 2004), very few have focused on how bridewealth payment and the benefits obtained (Aborampah, 1999; Anderson, 2007; Ngubane, 1987) thereof, could offer some protection against physical abuse and mal-treatment. These past narratives have echoed the negative implications of bridewealth payment with less emphasis on some of the potential social benefits that bridewealth payment would offer within a given context. Consistent with previous studies, this study has shown that traditionally, bridewealth is exchanged to show the value of the bride, and to ensure marital stability (Aborampah, 1999). These accounts suggest that bridewealth is exchanged to protect the woman from abuse and deflect various forms of mal- treatment. 157 University of Ghana http://ugspace.ug.edu.gh The participants indicated that men who paid bridewealth do so in order to honour their parents, their wives and themselves (Dery, 2015). This implies that the husband could be less likely to behave in ways that dishonour him and the families involved. It also suggests that through the payment of bridewealth, the couple legitimize the union and reduce or eliminate possible conflicts surrounding claims to property, children and other privileges. Consistent with other studies (Kaye et al., 2005; Ngubane, 1987), this study again shows that bridewealth payment has economic benefits. Relying on the resource theory (Goode, 1971), it can be argued that women’s access to such wealth and economic resources possibly reduces dependence on their partners, thereby decreasing the probability of experiencing intimate partner violence. Furthermore, it could be speculated that men who are able to pay bridewealth could be thought of as those who are economically better off, and may not rely on violence to command respect and compliance. 8.3 Implications of completeness of bridewealth payment This section explores links between completeness of bridewealth payment and pathways to men’s use or non-use of violence against their female partners. The analyses explored various consequences of bridewealth payment (non-payment, partial payment and full payment). Themes relating to triggers of conflicts and stress, and those relating to protection and support from abuse were identified. 158 University of Ghana http://ugspace.ug.edu.gh 8.3.1 Implications of non-payment of bridewealth This organizing theme on bridewealth non-payment covers both the social and economic consequences of non-payment of bridewealth, and how these could be pointers of conflict between the couple, and their families. The organizing themes, basic themes and the description of these themes for non-payment of bridewealth are presented in Appendix I. Eight basic themes were identified in relation to the consequences of non-payment of bridewealth. Figure 8.2 shows the thematic network of the implications of non-payment of bridewealth. 8.3.1.1 Breeds conflicts Some respondents stated that non-payment of bridewealth could be a focal point of quarrel and misunderstanding among couples. Since marriage and bridewealth payment are indicators of transition into adulthood, one’s inability to pay would be a reason for quarrels and argument. Non-payment also challenges traditional masculinities, and men’s rights to women in unions. In some cases, women may actually contest men’s claims to these rights. The following quotes show that non-payment could possibly challenge traditional power relations between the couple. These quotes suggest non-payment defies what it means to be a man in such unions. It also supports the assertion that when bridewealth has been paid, both men and women know their relative positions in the union, and women may be less likely to contest their male partners’ authority, something that could breed quarrels and conflicts in unions. “…Yes. If you haven’t performed my marriage rite, would I allow you to control me? When you talk I would also talk. If you are a capable man then you have to perform my marriage rite…” (R14; elderly woman) 159 University of Ghana http://ugspace.ug.edu.gh Fig ure 8. 2: Thematic network of the implications of non-payment of bridewealth Cannot name the child No privileges Physical Children do not abuse/beating Breeds belong to their conflicts biological father Poor parental support Implications of non- Authority payment of bridewealth Non-payment does not guarantee abuse Man has no Man has some Woman has No sanctions authority authority been stolen against the woman Cheats Insults Source: 2018 Bridewealth Study (RIPS) This respondent also indicated that non-payment could mean that the relationship is not ‘proper’, consequently, couples may show no respect or may disregard each other. “..Then the relationship between you and him was not proper. It depends on both of you. There are women if you have not married them, they will keep disregarding you. There are some men who will say “back off”, the same”. (R15; elderly woman) 160 University of Ghana http://ugspace.ug.edu.gh 8.3.1.2 Woman has been “stolen” In addition to the above, non-payment of bridewealth could also mean that the marriage was not organized through the right channels. In some cases, the respondents described this as the woman being ‘stolen’ or living with the man at no cost. “…What shows is that you live with the woman. You have given birth. He has not gone to see the family. That means he has gotten her cheaply. 20~~~ I hope you understand. ~~~. You understand very well…..He has not paid anything. He has not incurred any expenses on her…He has not incurred any expenses on her so he has gotten her cheaply”. (R9; elderly woman) “..The relevance is that if you marry a man and he has not performed your marriage rites then he is not your husband. He is either your friend or he has stolen you. So your family can take him on but if there is understanding then he can come and perform your marriage rites. So that it can be considered as a marriage”. (R16; elderly woman) 8.3.1.3 Physical abuse/beating With regard to physical abuse, some respondents stated that when bridewealth has not been paid, the man can abuse his partner. Respondents indicated that non-payment suggests that the man does not care about his partner. In addition, some respondents noted that the use of family planning without the man’s consent can also put the woman at risk of physical abuse when bridewealth has not been paid. 20 Transcription symbol “~~~” means laughter 161 University of Ghana http://ugspace.ug.edu.gh “… Yes he can beat her...but he already has kids with her so this will also go to the elders who tell them not to repeat that because it has brought about fight”. (R16; elderly woman) “….What about a man who has not performed the woman’s marriage rites… Yes. He can beat her. Why? Many. He does not care”. (R 9; elderly woman) 8.3.1.4 Children do not belong to their biological father Some participants stated that when bridewealth has not been paid, children born by the woman to the man do not belong to their biological father. The children then belong to their maternal family. In many cases, the man would not be allowed to name the children. The children are named by their maternal uncle. Since their maternal uncle assumes the responsibility of the biological father, he will also be responsible for all other major decisions concerning the children. The following quotes from respondents illustrate how non-payment of bridewealth could strip the biological father of these responsibilities and privileges: “…since you have not married, her family can name them, and if you have given birth and the family has named them are the children yours? You can have so many daughters and when they grow someone will also want to marry them, you the father will be there and you will realize that their uncle from their mother’s side will be taking the bridewealth. It is very painful……What is more painful than this, it is very, very painful”. (R8; elderly man) 162 University of Ghana http://ugspace.ug.edu.gh “Um, in the case of a man who has not performed marriage rites for a woman he has a child with, the woman has power over the children. The power you have over the children is that, we the Akuapims, if a man has not performed your marriage rite, you will not be allowed to give your child a name”. (R9, elderly woman) 8.3.1.5 Authority Participants indicated that when bridewealth has not been paid the man may have some authority or may not have any authority. Some studies have shown that bridewealth payment confers some level of authority to the man. Thus, where no payment has been made, respondents related non-payment of bridewealth with the man having no authority in the union. They stated that when bridewealth has not been paid, the man has no authority over the woman in the union. He may even be considered to have stolen the woman, and is more likely to be disregarded by his partner. The quotes below highlight instances of lack of authority: “…No, he doesn’t have authority. He doesn’t have much authority. It means that you have stolen the woman and she is with you…” (R5; elderly man) “… Then you are just beating her in vain. You have not performed her rites but you just get up and beat her by heart Eiiiiii! Then that would not happen! Yes, the person you are with, have you performed her rites? You took her from her home and have not performed her rites but get up to beat her. It would not happen. Would you be able to do that?” (R14; elderly woman) Although some respondents stated that the man has no authority when no bridewealth has been paid, others stated that the man still has authority. The participants suggested that since the 163 University of Ghana http://ugspace.ug.edu.gh woman lives with the man, and the man provides for their basic needs, he has authority in that union. The quotes below illustrate ways in which the man has authority when no marriage payment has been made: “..21I: Can they also deny their husbands of sex? R: Why won’t she allow him? If she will not allow, then she must leave. If a man lives with you without paying your bridewealth, he still provides food for you. He is not taking care of an orphan? If the woman will deny her husband then she can as well leave”. (R1; elderly man) “…. as for that one you haven’t permitted him to marry you so the man can do it to you……yes it means that he is the one taking care of the woman so if you don’t inform him and it becomes a quarrel then because he hasn’t married her so he can’t take it to your father”. (R5; elderly man) 8.3.1.6 No sanctions against the woman Some respondents also stated that when bridewealth has not been paid, the woman cannot be sanctioned if she cheats on her partner or insults him. They indicated that these acts (cheating and insults) generally attract fines and some other forms of sanctions when the couple is married. From the quotes below, participants indicated that without proper marriage payment, the partner cannot claim these. “… You know if you live with a woman and you have not married her and she cheats, you cannot ask her to pay a fine. It can be that you take care of her but if you do not go and 21 I: Interviewer. R: Respondent 164 University of Ghana http://ugspace.ug.edu.gh perform the marriage rites and she goes out to have an affair, the fine cannot be taken. The father will say he does not know you because, you have not married her. So it is someone you have married that you can have that law with”. (R2; elderly man) “…Eeehhee. This is because if I (the man) have not married you (the woman) and you insult me that I am a fool and I report you to the elders, your family would defend you and say that even the man who says our daughter insulted him is not known to us. I hope you understand. Where have you known him? And you say he is your husband, they would ask, when did he marry you? Who gave you the woman that you are reporting yourself that she insulted you that you are a fool? You have even sent yourself. They would take ‘dedua’ (compensation/fine)”. (R6; elderly man) 8.3.1.7 Poor parental support Traditionally, marriage is primarily between families, and only secondarily between the couple (Fortes et al., 1947; Mizinga, 2000; Nukunya, 1969). This implies that parents and the families of the bride and groom also have a high stake in the success of the union. Thus, couples can lean on these forms of social support to prevent or manage any marital misunderstandings. When no payment has been made, and the union is not duly legitimized, respondents indicated that the pool of social support from which the couple can tap diminishes especially support from parents. Parents of both the man and the woman may refuse to help settle any disputes that will be presented to them. The following quotes suggest the poor parental support couples will receive from their parents when dealing with conflicts in their unions. 165 University of Ghana http://ugspace.ug.edu.gh “It can be that you take care of her but if you do not go and perform the marriage rites and she goes out to have an affair, the fine cannot be taken. The father will say he does not know you because, you have not married her. So it is someone you have married that you can have that law with”. (R2; elderly man) “…how will you summon him? If he has not married you and you are there with him and he beats you up, how will you go and summon him that why has he beaten her up? Have I brought you something? I haven’t taken anything from you so you can’t summon him. You yourself can’t even call him”. (R11; elderly woman) 8.3.1.8 Non-payment does not guarantee abuse Some respondents also stated that non-payment of bridewealth does not mean that the man has any authority to beat his wife. Any such act could attract the disapproval from the woman’s family. The following quotes suggest that when bridewealth has not been paid, the man does not have any right at all to be abusive: “I…What about the man who has not paid the woman’s bridewealth? Can you have that power…..R: Even when you have gone to pay her bridewealth, you do not have the right? How much more when you have not? Have heard this before? Ahh!!!!” (R1; elderly man) 8.3.2 Implications of partial bridewealth payment Some participants highlighted some implications of making partial bridewealth payment. Participants stated that partial bridewealth payment was not the ideal form of marriage arrangement in these contexts. Thus, they indicated that making partial payment may have some 166 University of Ghana http://ugspace.ug.edu.gh negative consequences. Seven themes were identified as some of the consequences of partial payment (see Figure 8.3). In Appendix J, organizing themes and basic themes for partial bridewealth payment are displayed. 8.3.2.1 Breeds conflicts Similar to non-payment of bridewealth, when bridewealth has been partially paid, it could be a point of conflict between the husband and the wife. From the following quote, the respondent indicated that when payments are incomplete, couples are more likely to be verbally abusive towards each other on issues regarding the partial payment. “Traditionally, in the past, you cannot make partial payment. You will perform everything. Unless your in-law likes you and considers you to do that. Even in such cases the woman can use that as an insult anytime. The man can also use that as an insult. You see how the situation is? So you must force the man to perform your marriage rites”. (R9; elderly woman) 167 University of Ghana http://ugspace.ug.edu.gh F igure 8. 3: Thematic network of the implications of partial bridewealth payment Woman is not Children do not Man has some belong to the man authority Breeds conflicts No sanction if the Implications of partial woman cheats bridewealth payment Poor parental su pport Type of union Considered as marriage Not considered as marriage Source: 2018 Bridewealth Study (RIPS) 8.3.2.2 Man has some authority Partial bridewealth payment gives the man some level of authority in the union. Since the couple have introduced themselves to each other’s families, and some bridewealth has been exchanged, the man has some level of authority in the union. “…once he has come to plead; you know, it is not always that it is all well for a person. So he has come to do half of it and he has given you time to come and do the rest or “tomorrow I’ll bring the rest.” and you’ve agreed so she is for him. Do you understand 168 University of Ghana http://ugspace.ug.edu.gh what I’m saying?...so whatever he wants is what he will let the woman do. So anytime you need her, you have to seek permission from her husband”. (R11; elderly woman) 8.3.2.3 Woman is not valued Some participants reported that partial bridewealth payment could imply that the groom does not value his partner. Narratives suggest that men who make incomplete bridewealth payment got into the relationship with their partners on a silver platter. The quote below highlights this. “Some men decide to make part payment. All these depend on the woman. For instance, as an educated young lady, if a man pays part of your bridewealth and you allow him to impregnate you, then he got you cheaply”. (R1; elderly man) 8.3.2.4 Children do not belong to the man As with the narratives that relate to non-payment of bridewealth, participants reported that when bridewealth is partially paid, the children do not belong to the man. In the quote below, this chief indicated that if one goes strictly by tradition then a man who has made partial payment would not be allowed to even name the children he bears with the woman. “R: we will allow him name the child. I: ok, in that case are you considering him it is tradition? R: well, I do not know how to but you want to go strictly by tradition I: yes what tradition says, then we will later come to what pertains today. R: if it is strictly by tradition then you do not qualify to name the child”. (R8; elderly man) 169 University of Ghana http://ugspace.ug.edu.gh “That is what I am saying that if the man does not do that, then the children belong to the woman. ~~~”. (R9; elderly woman) 8.3.2.5 Type of union Participants indicated that in these contexts partial payment may be considered as marriage or not depending on the situation. Some participants stated that when bridewealth is partially paid, it is not considered as marriage. In fact, they argued that in the past partial payment was not practised. Participants blamed the emergence of partial payment on how expensive marriage exchanges have become over time. “If a goat is dying and it is not dead yet, you don’t call it a dead goat”…So that serves as a deterrent for the men to marry a woman they take …It is not considered as marriage that was why I said “if a goat is dying and it is not dead yet, we don’t call it a dead goat”. So if you are in the process and you have not completed it, she is not your wife. You have not married her. This thing, the list you are talking about, it is not static”. (R2; elderly man) “Please, please that word ‘he has partially married the woman’, we don’t have to use it at all….YOU HAVEN’T DONE ANYTHING! YOU HAVEN’T DONE ANYTHING!”. (R3; elderly man) “I have not witnessed a case where a man decides to perform the marriage ceremony, pays the bridewealth and does not provide all the items. If some of the items are not provided, then it means the marriage is not complete. This seldom happens. If you have 170 University of Ghana http://ugspace.ug.edu.gh come for the list, then you must prepare adequately before coming. You select a day when you are well prepared and say “I am also coming with my family members”. I have not witnessed a case where you come for the list and not all the items have been provided. Even if you come, you will be asked to go back so the marriage is not complete”. (R15; elderly woman) While some participants suggested that traditionally partial bridewealth payment is not considered as marriage, others indicated that it could be considered as marriage. Partial payment in these cases is due to financial constraints. “R: Oh! But you came for her from the home. So it is complete because you intentionally came to the house to plead so that is not like you’ve ignored that you will not come and marry her. You’ve come home to plead so you’ve done some and left the rest. You’ve done some but it wasn’t complete so you will bring it later. I: so will they say the woman is his wife? R: yes”. (R11; elderly woman) 8.3.2.6 No sanctions if the woman cheats In order to maintain social control, bridewealth payments are made to deter men and women from engaging in activities that may put the union at risk of dissolution. Some of such sanctions relate to paying fines if a woman for example cheats on her husband. Participants argued that since marriage payments are not complete, and the marriage is not considered as one, how would the man claim such fines? 171 University of Ghana http://ugspace.ug.edu.gh “If you do not finish, you the man it is to your disadvantage… if it happens that the woman cheats on you who will follow you to go for your “ayifri” (compensation/fine)”. (R8; elderly man) 8.3.2.7 Poor parental support Since bridewealth payment brings the families of the bride and groom together and legitimizes the union, incomplete payment possibly reduces the support the couple would receive from their parents if they consider the union as not legally arranged. “If you are my daughter and this man has brought two bottles of schnapps and has brought some things maybe 100 Ghana cedis and says I have started performing the rites and Yaa goes to see another man and she goes to have an affair and the man comes to complain I will tell him, “gentleman, but you have not finished paying?” (~~~)….Maybe you may have had a child with her. If I charge you, it does not mean, my daughter is your wife. No, not all”. (R2; elderly man) “…but if you say you are not entirely ready so you cannot perform the rites then she is not your wife so if she goes for another man they will say you are not married to her and you are behaving like that, then you the man will feel embarrassed”. (R15; elderly woman) 172 University of Ghana http://ugspace.ug.edu.gh 8.3.3 Implications of full bridewealth payment Past research has shown that full bridewealth payment gives the man authority in union. Bridewealth payment also specifies couple’s roles and responsibilities. This study identified various consequences of full bridewealth payment. Organizing themes and basic themes for full bridewealth payment are shown in Appendix K. It appears that these consequences overlap and could either be negative or positive. Seven main themes were identified and these are presented in the thematic diagram in Figure 8.4. Figure 8. 4 : Thematic network of the implications of full bridewealth payment Does not Position of Physical guarantee responsibility abuse/beating physical abuse Man has Implications of full authority over bridewealth payment the woman Woman has not been bought Authority to reprimand the woman Social Other relatives support Friends/witnesses Chiefs Queen mothers Parents Source: 2018 Bridewealth Study (RIPS) 173 University of Ghana http://ugspace.ug.edu.gh 8.3.3.1 Position of responsibility Participants indicated that bridewealth payment though gives the man authority in the union; the essence of that authority is for the man to be responsible in the union. It is a position of responsibility. Participants suggested that payment gives the man authority to protect the woman from harm, and provide for the family. These were some of the ways by which the man should be responsible. Participants reported that when the man and the woman are married, they are both responsible for each other’s safety, and are entreated to do all they can to ensure that both of them are safe and protected from harm. The following quotes echo participants’ understanding on how men are expected to be responsible when marriage payments have been made: “Yes. Now, I have come for you from your family, so if anything happens to you, I have to take responsibility. That is the authority he has been given”. (R1; elderly man) “ ..The power you have is that this is your wife and no one can have her. That is the power and when she asks you for permission when she has to go somewhere. That is the power. The power to say “lady I am hungry so cook something for me to eat”. That is the power. If she does not do it, go and report. You cannot beat her!” (R2; elderly man) “…the man having authority means that when he wakes up, he will provide the housekeeping money and everything”. (R12; elderly woman) 174 University of Ghana http://ugspace.ug.edu.gh 8.3.3.2 Man has authority over the woman Participants’ accounts suggested that when bridewealth is paid in full, the man has authority over the woman. Participants referred to the man as the head of the family thus, has control over his wife, and over the affairs of his household. “..It is the man who is the head of the family. He is the head of the family. The only duty of a woman is to manage the family, house. You do the management and the husband provides finance….The husband provides finance but the man has more authority than the woman. That is how the situation is”. (R1; elderly man) “According to tradition the man is the head so what he says must hold….The man has control in a marriage what gives him the control in a marriage? He is the one who has paid your bridewealth”. (R8; elderly man) “Does the man have power over the woman because of the marriage rite performed? R: Yes. If you have not performed her marriage rite, you cannot control her”. (R14; elderly woman) Some participants suggested that when a woman is married, and bridewealth is paid in full, the woman is more likely to respect the husband. “…..the woman too because she is married to you she respects you”. (R8; elderly man) 175 University of Ghana http://ugspace.ug.edu.gh 8.3.3.3 Does not guarantee physical abuse All the participants indicated that although the man has been given authority over the woman in the union when bridewealth has been paid, the essence of that authority is never to be abusive toward the woman. Participants mentioned how men and women are advised during the marital process never to be abusive, and the sanctions that will be meted if the woman is abused. This was to deter the man from abusing the woman. Men are advised to report any form of behaviour from the woman they considered inappropriate to the woman’s family. Participants also indicated that their traditions do not permit a man to physically abuse his wife because bridewealth has been paid. “… eish! As for that I beg of you, it is an abomination in a marriage, in every marriage the elders advise you not to beat a woman, it is not done, if you do that you have violated the law. We do not beat women. For some people if you beat their daughter then the marriage will be dissolved. No matter how much she has wronged you, you bring the matter to me but do not beat her. Our traditional marriages do not allow that”. (R8; elderly man) “Marrying her does not include beating. You cannot beat her. If you beat her and she takes you to the police station, they won’t say because she is your wife so you will be forgiven. You will be dealt with according to what the law says. That does not mean that that power is part. The fact that you come and marry her does not mean you should beat her. No. during the giving of advice, some people even say “make sure you do not beat my daughter. The day I hear you have beaten her, I will take her back”. That is the meaning of performing the marriage rites and when it comes to the giving of advice, “I 176 University of Ghana http://ugspace.ug.edu.gh don’t want a man who will beat her. If I ever hear that you have beaten my daughter, I will take her back. So I do not want that”. So that power does not exist for you to beat her. But the power is for her to do what you want so that there can be understanding”. (R15; elderly woman) Some participants also noted that bridewealth payment does not give the man leeway to abuse his wife and that men are violent because they want to be and because of their attitude. “…I am saying that a man is not supposed to beat his wife but some men have that attitude. You see. I hope you understand what I mean. He can for instance become very angry because his wife has not finished preparing his food when he comes home”. (R9; elderly woman) 8.3.3.4 Woman has not been bought Participants indicated that bridewealth payment does not mean the woman has been bought. They suggested that because the woman has not been bought the man has no right to beat her or abuse her. The following quotes reflect some of these assertions: “You just married her, you haven’t bought her yet. Do you understand? So when you marry the woman (it) does not mean you should trouble and destroy her. Do you understand?”(R 6; elderly man) “.. no you haven’t bought her. Even with this our tradition, when you are coming to marry a woman, we tell you. The day you will beat her up, when we call you, heed to our call…” (R11; elderly woman) 177 University of Ghana http://ugspace.ug.edu.gh 8.3.3.5 Physical abuse/beating While all participants stated that marriage payment does not give the husband any authority to abuse his wife, from the quotes below, a few participants also indicated that the husband can be physically abusive under certain conditions, for example, if his wife does not perform certain duties. Patriarchal traditional norms may put women at risk of male-perpetrated abuse (Haj-Yahia & Uysal, 2008). “….But the man is the head of the house. It is the woman who sees to the cooking of food, marketing, etc. Because of what you have, if the husband must kill himself, he will do that to obtain it. If the woman refuses to have sexual relations, he can even beat her up mercilessly”. (R1; elderly man) In the quote above, although this particular respondent reported in previous narratives that bridewealth payment does not mean the man can beat his partner, he also makes it clear that under certain conditions the man can be abusive. This indicates that although generally individuals perceive bridewealth payment as not a reason for abuse, they also indicate that depending on the situation some level of physical violence can be used to gain compliance. “….if you have married her and she does that (uses family planning) he can beat her”. (R16; elderly woman) Some participants also equated the woman to a child and suggested that sometimes a woman can be disciplined as a child is disciplined. “…You even beat your child. If your child does something wrong, you beat him or her. It doesn’t mean you always descend on the woman to beat her. Sometimes when the woman 178 University of Ghana http://ugspace.ug.edu.gh does something which is wrong eerh but even that if the woman behaves in a certain way, no one will hear. But there is someone who is always descending on his wife to beat her but this is not right”. (R11; elderly woman) 8.3.3.6 Authority to reprimand the woman Respondents also stated that the man has authority to reprimand his wife if she does something wrong or misbehaves. It was also mentioned that a wife could be beaten as a form of warning to prevent her from repeating the same behaviour subsequently. “…no! Even if you will punish her, it’s not good that you beat her”. (R5; elderly man) “…Whether someone’s child misbehaves and your child misbehaves, who will you reprimand?... Your own child…. So there is no issue about that”. (R14; elderly woman) “….Then it means you have to go back to where you came from. If you make yourself proud over your husband you may receive a slap and the slap is just a warning. That is why I said that it shouldn’t be that you the man will go in for the woman to beat her. You get it”. (R3; elderly man) 8.3.3.7 Social support Participants mentioned various sources from which married couples could seek/obtain social support when they experience difficult situations in their unions. Couples are offered pieces of advice at the time of the union on how to manage marital conflicts and stresses that 179 University of Ghana http://ugspace.ug.edu.gh could occur. In this study, participants mentioned five main individuals from whom the couple could obtain social support or from whom they could contact when they have marital challenges. These people include queen mothers, chiefs, parents, other relatives, and witness who are responsible for the married couples, and friends. These sources of social support work in two ways. First, the recognition that the entire community has a stake in the success of the union may deter couples from actively engaging in activities that may jeopardize the success of the union or their relationship with their pool of social support. Secondly, couples are explicitly counselled to tap into all the social support available to them (parents and queen mothers, chiefs, other relatives, and friends) to help resolve their marital challenges. Parents and family members are aware of marital conflicts, and consciously advise couples to deal with the issue in a way that does not put the woman at risk. “..The way you liked her, that is how her family also sees her and they have to see it like that or see some form of improvement that there is peace where she is. That is one thing parents notice or observe about their child’s marriage. They know if the marriage is good or bad. You won’t say anything but they can know whether it is good or bad. When they see your character and the way you are living, they can tell that things are not going well with you. So they can ask you “Why? Akosua are you sick? What is disturbing you? Now you have lost weight. That shows that the man is not handling you well. That is why some parents visit their daughters in their marital homes. When they come they see how things are going on with you. That is why some parents stay long when they pay a visit and the man does not understand. Maybe the daughter did not complain but they have seen it themselves. Uuuhuu”. (R6, elderly man) 180 University of Ghana http://ugspace.ug.edu.gh “…No matter how much she has wronged you, you bring the matter to me (chief) but do not beat her. Our traditional marriages do not allow that”. (R8, elderly man) “…We are often confronted with many of these issues … this will result in fights and after you investigate and you will realise there is more to it.…We can talk to them and there will be understanding”. (R14, elderly woman) 8.3.3.8 Discussion The aim of this section was to explore participants’ perceptions of the implications of completeness of bridewealth payment in these contexts. The section examined narratives that explored the effects of non-payment, partial payment and full payment on the marriage institution, and specifically, on male-perpetrated intimate partner violence. These accounts suggest that non-payment and partial payment of bridewealth could actually breed conflicts between the man and his partner. Non-payment and partial bridewealth payment breed (show) power dynamics within the union that challenge traditional gender roles and masculinities. Participants suggested that men who have not paid bridewealth were more likely to be challenged by their female partners about their claim to authority in the union. Non-payment of bridewealth actually defies one’s position as a ‘man’, and strips him of his authority in the union. It is important to note that these assertions further deepen traditional masculinity scripts placing a lot of pressure on men to fit within these traditional scripts. When the relative position of men is challenged due to lack of economic or social resources, they are more likely to rely on violence as an alternative resource in order to 181 University of Ghana http://ugspace.ug.edu.gh gain control and compliance (Atkinson et al., 2005; Goode, 1971; Panda & Agarwal, 2005; Weitzman, 2014). These results corroborate the quantitative findings of this study which showed that non-payment of bridewealth is significantly associated with actual perpetration of abuse. Thus, these narratives suggest that men who are unable to pay bridewealth, first, possibly lack the economic resources needed to pay bridewealth, and secondly, lack the respect and authority bridewealth payment confers. Respondents also related non-payment and partial payment of bridewealth to the man not appreciating the value of his partner and thus, may be more likely to abuse her. Consistent with previous studies, participants indicated that bridewealth payment suggested the importance of the bride, and the marriage institution as whole (Aborampah, 1999). Bridewealth is exchanged to unite families, maintain kinship relationship, and protect the women from abuse. The involvement of families in uniting the couple, and the pool of social support available to the couple would insulate or reduce the chances of abuse. According to the exchange/social control theory, when members of the nuclear family for example have strong attachment to one another but lack significant social support (both formal and informal) from individuals outside their inner circle then the use of violence may imminent and private (Gelles, 1983). Social support and control developed when bridewealth is exchanged potentially prevents men from being violent in some cases. Contrary to the Western model of marriage contract which involves the bride and the groom, in sub-Saharan Africa, the marriage contract extends beyond the bride and groom. After the actual marriage ceremony, couples are assured of the support of their kinsmen. Queen mothers, chiefs, parents and friends of the bride and groom are key individuals the couple can rely on to address marital challenges. The involvement of all these actors in the social exchange 182 University of Ghana http://ugspace.ug.edu.gh network possibly prevents the couple from behaving in ways that would cause the collapse of the union. Knowledge of the physical presence of families and the interest of all stakeholders in the success of the union may deter male partners from being abusive. In their study, Lanier & Maume (2009) reported that women who stated that they receive assistance in terms of child support, transportation, housework and advice from friends and family members were less likely to report ever experiencing abuse. This reflects in the narratives where participants stated that couples had poor parental support when bridewealth has not been paid but had access to wide range of support when full bridewealth has been exchanged between the families. As with past research, families do not recognize unions where payments have not been made (Mizinga, 2000), and would be less likely to support the couple if they have marital misunderstanding or needed support. The man has no claim to damages if the woman, for example, cheats on him. Suspicions and accusations of infidelity may put women at risk of intimate partner violence (Adomako Ampofo, 1993; Ezeudu et al., 2019; Owusu-Addo et al., 2018). In addition, because marriage is a process rather than a onetime event, couples receive advice, and are constantly in touch with each other’s family. Aborampah (1999) noted that gifts from sons’-in-laws to the brides’ parents were almost customary. These exchanges are made to maintain cordial relationships between families. In their study for example, Mozhdeh, Ghazinour, Nojomi, & Richter (2012) found that the perceived value of support received by pregnant women had a greater protective effect against violence compared with the number of individuals providing support. 183 University of Ghana http://ugspace.ug.edu.gh Contrary to previous studies which suggest that bridewealth payment could imply to some men that the woman has been bought thus, she is more likely to experience male- perpetrated violence (Ansell, 2001; Ansell et al., 2017; Ofei-Aboagye, 1994b), in these contexts, key informant accounts suggest that bridewealth payment does not mean the woman has not been bought (Aborampah, 1999; Mizinga, 2000), and payment does not give the man the right to abuse his wife. These accounts, however, suggest that bridewealth payment though gives the husband authority, that authority reflects a position of responsibility rather than abuse. This new position is to ensure that his family is safe and provided for, and that his female partner is protected from harm. Another plausible explanation for why key informants noted that bridewealth payment actually protects women from abuse is that because couples, especially women, understand their relative positions in the union, they would be less likely to engage in acts that predispose them to abuse. These highlight the essence of reciprocity as indicated in the exchange/social control model. These results could imply that because the couple acknowledge that bridewealth payment requires of them to perform certain roles and responsibilities, they are more likely to conform to these, averting various forms of conflicts and misunderstanding. For example, the man taking a position of responsibility would provide the needed economic resources for his household and the woman, offers domestic, sexual and reproductive services. Further, the study found that the disadvantages associated with not paying bridewealth in full are quite numerous. Discussions about issues regarding non-payment/incomplete payment between couples, and even their families could heighten levels of quarrels between partners. This is because non-payment/incomplete payment of bridewealth has implications which extend beyond the couple. The data show that bridewealth payment has implications for inheritance and belonging. Without bridewealth payment, traditionally, fathers would be unable to name their 184 University of Ghana http://ugspace.ug.edu.gh children, children would not have access to their fathers’ property, and fathers would be less likely to rely on their children in their old age. Also, it takes from him the advantage of benefiting from the share of bridewealth if he had daughters (Mizinga, 2000). These are all crucial for lineage continuation and expansion of kin groups. Both families of the bride and groom, in an attempt to avert these situations would be more likely to frown upon non- payment/incomplete payment, and ensure that when bridewealth has been paid couples live together amicably. They would also be less likely to approve of behaviours that put the couple at risk of harm or the union at risk of dissolution. It is worth mentioning that some participants stated that irrespective of completeness of bridewealth payment, the man has no authority to physically abuse his partner. Although all participants pointed out that bridewealth payment does not mean the woman has been bought, and that the man is not permitted to beat her, a few participants again suggested that when bridewealth has been paid, the man has the right to reprimand his wife, and could actually sometimes physically abuse her under certain conditions. According to the exchange/social control model, individuals provide services and assume that their actions will be reciprocated. When such expectations are not met, disapproval and conflicts are inevitable (Gelles, 1983). Even though the use of violence is generally not accepted, depending on the context and reason, some level of violence may be tolerated. These key informants indicated that a woman can be beaten if she for example, refuses to have sex with her husband or uses family planning without his permission. These results show that approval of wife-beating although may be relatively low among men as shown in this study, it is important that research focuses on areas that men may be more 185 University of Ghana http://ugspace.ug.edu.gh likely to approve of the use of violence against women. The qualitative results show that while these same key informants claimed that men should not physically abuse their wives when bridewealth has been paid, a few also indicated that women could be physically abused if they use family planning without consent from their spouses, for example. She may also be reprimanded in order to gain compliance. These findings are consistent with other research where the use of violence against women is justified when bridewealth has been paid and women make sole reproductive decisions (Dery & Diedong, 2014; Horne et al., 2013). It is important that items that are used to examine attitudes towards wife-beating focus on a broad spectrum of behaviours especially those regarding sexual and reproductive decision making. 8.4 Sanctions/strategies for dealing with abuse Traditionally, some strategies have been put in place to deal with men’s use of violence against women. Participants mentioned six ways by which men may be sanctioned for being abusive (see Appendix L). These are divorce, separation, conflict resolution strategies (fines, goods, summoned, apology/disgracing), arrested by the police, warning and denying the husband sex. These are represented in Figure 8.5. 186 University of Ghana http://ugspace.ug.edu.gh Figure 8. 5: Thematic network representing sanctions or strategies for dealing with abuse Legal/police Separation case Deny husband sex Strategies/sanctions Divorce for dealing with abuse Warning Apology/shaming Conflict resolution Fines (cash or kind) Summoned Source: 2016 Bridewealth Study (RIPS) 8.4.1 Divorce Some participants indicated that in times of abuse, the only option is for the woman to divorce the abusive husband. It was reported that the woman’s parents may even spearhead the divorce process if the man abuses their daughter. “…For some people if you beat their daughter then the marriage will be dissolved”. (R8; elderly man) “…If the woman says she can’t stay married again or the man says he can’t stay married again before the marriage is dissolved. In most cases, I don’t know if I gave you the 187 University of Ghana http://ugspace.ug.edu.gh details of preparation for marriage. If it is properly done, they hardly get to this point”. (R2; elderly man) 8.4.2 Legal or police case While some participants mentioned various informal avenues through which abusive men could be sanctioned, others indicated that in recent times, men who abuse their wives will be arrested and dealt with by the state. The quotes below support these claims: “That is the simplest solution. I hope you understand what I am saying. This time, if you beat a woman, the government will deal with you. That is all”. (R1; elderly man) “In these present times, if you beat up a woman it becomes a police case”. (R16; elderly woman) 8.4.3 Separation Separation from an abusive partner is one of the ways the partner could be sanctioned for being abusive. From quotes below, participants indicated that an abusive husband can be punished by separating the couple for some time. In doing this, he will lack all the support he usually obtains from the wife. Some parents may also encourage such separation. “Some can make their daughters stay with them for a long time to punish you very well”. (R8; elderly man) 188 University of Ghana http://ugspace.ug.edu.gh “Because such people when you beat them they may go to their mothers for like one week. Within this one week that she hasn’t returned, she hasn’t cooked for you so before she will come and prepare the food, she will have to kill a fowl (pacify the woman) before you should eat the food”. (R11; elderly woman) 8.4.4 Conflict resolution Traditionally, efforts are made to amicably resolve marriage disputes in order that the couple would continue to live together as husband and wife. The abusive husband could be asked to compensate his wife if she reports him for being abusive. The data suggest that conflicts resolution could be compensations paid by the husband to the wife in the form of cash or kind. The husband may also be summoned to the chief’s palace to address the conflict. Some participants indicated that the abusive husband may be disgraced or shamed to deter him from repeating such acts. Again the abusive partner will be asked to apologize to his wife and promise to be of good behaviour. “…yes. He also gives the fowl for the woman to cook with it. You understand. If he doesn’t have some, he would have to go and buy it”. (R11; elderly woman) “If you go and mistreat her, you will run at a loss and you’ll come and pay”. (R3; elderly man) “Over here when you do that, we will think that you are a villager. That is the way it is, that you are a villager. It is a villager who beats his wife. And if you are summoned before us, by the time we release you, you will be weak”. (R2: elderly man) 189 University of Ghana http://ugspace.ug.edu.gh 8.4.5 Warning A few participants indicated that the abusive husband will be sternly warned if his wife’s parents are informed of the abuse their daughter is suffering or has suffered. Parents were very cautious of the fact that their children could be victims of abuse, and were quick to warn the husband never to abuse their daughter. “….I got up and nearly beat the man in the church premises. I warned him that if he dares touches my daughter; I will give him two weeks to return my daughter. He saw that a man has also arrived. I told him I was not an old mam…not my daughter”. (R1; elderly man) “We did not ask you to go and beat her. But if you beat her, and she complains, then whatever the father wishes, whatever he wants to tell you, he can”. (R6; elderly man) 8.4.6 Deny the husband sex One participant indicated that some women may sanction their abusive partners by denying them sex. These women took issues into their own hands and refused to have sexual relations with their partners who abuse them. In the quote below, this elderly woman indicated that a woman may even refuse to have sex with her partner for a whole year if he is abusive. “…there are some women that they won’t even sleep with their husbands for a whole year”. (R10; elderly woman) 190 University of Ghana http://ugspace.ug.edu.gh 8.4.7 Discussion The qualitative narratives highlighted ways by which informal and formal social control strategies are employed to protect women from abuse. These strategies are to first, protect women from abuse, prevent subsequent abuse, punish the offender for committing the crime, and hopefully, deter other men from emulating such behaviours. It is important to note that the fear of losing one’s social status or the fear of being labelled a wife beater as Gelles (1983) describes it may prevent men from being violent. According to the social control theory, if the cost of being violent outweighs the benefits then individuals, in this case men will be less likely to use violence against their intimate female partners (Gelles, 1983). Community members rely on various tactics to maintain order and deter members from committing crimes (Sampson, Raudenbush, & Earls, 1997). This is referred to as informal social control. Studies have shown that both formal and informal strategies have implications for violence prevention and for social order (Goodman, Dutton, Weinfurt, & Cook, 2003; Sampson et al., 1997; Wood, Glass, & Decker, 2019). Social support and control developed when bridewealth is exchanged potentially prevent men from being violent. Key informants suggested that informal sanctions/strategies were instrumental in dealing with conflicts between couples. Compared to institutionalized individuals like pastors, the likelihood of other family members knowing about marital conflict is relatively high, and these informal support systems could be most beneficial for violence prevention (Goodman et al., 2003). Similar to this, the narratives show that parents of the bride pay particular attention to the welfare of their daughter and are more likely to spearhead the dissolution of union or offer a safe place for her when they find out that their daughter is in 191 University of Ghana http://ugspace.ug.edu.gh danger. Past research suggests that informal social control offered by family members potentially reduce the severity of IPV (Emery et al., 2017; Wood et al., 2019). Separation from a physically abusive partner is another strategy key informants reported could be used to stop abusive partners (Ragavan, Iyengar, & Wurtz, 2015). Resistant approaches like denying one’s husband sex may be used to address issues of abuse. The aim of this approach is to hopefully give the abusive partner time to reflect on his actions and probably control the abusive partner’s sense of power. However, these resistance approaches have been found to be the least beneficial (Goodman et al., 2003). Another strategy is the use of the formal judicial services in handling issues of abuse. Past research suggests that arrest is correlated with lower levels of subsequent abuse (Maxwell, Garner, & Fagan, 2001). In the narratives, key informants were clear on the use of the formal judicial system in handling issues of male-perpetrated violence. If used properly and promptly, this approach may prevent repeat abuse and discourage other men from being abusive. In traditional settings, however, divorce and arrest may not always be the first solution to abuse. Families may try to resolve marital conflicts by employing some of the conflict resolution strategies mentioned above. Although these strategies could be useful, they may also place the woman at risk of repeated abuse. If sanctions are not preventive or punitive enough, women may be put at greater harm. 192 University of Ghana http://ugspace.ug.edu.gh CHAPTER NINE SUMMARY, RECOMMENDATIONS AND CONCLUSIONS 9.1 Introduction In this study, the overarching objective was to examine the association between bridewealth payment and male-perpetrated intimate partner violence against women using data from Ghana. Specifically, the study examined the association between bridewealth payment (that is, bridewealth negotiation status and completeness of bridewealth payment) on (a) men’s attitudes toward wife-beating and (b) men’s self-reported actual perpetration of violence against their female partners. The study used the integrated ecological framework (which describes how different levels of factors - individual and community level factors are associated with the dependent variables), the resource and exchange/social control theories to explain male- perpetrated intimate partner violence. Using a mixed-methods approach, this study relied on a nationally representative dataset, 2014 Ghana Demographic and Health Survey (sample size 1,893 men), and a population-based study conducted in selected communities in the Eastern Region. The 2014 GDHS is a nationally representative sample survey and is the most recent of the series that have been conducted every five years since 1988. The second population-based study was on Bridewealth Payment and Normative Constraints on Women’s Lives in Ghana (survey sample size 579 men and 16 key informant in-depth interviews). The mixed-methods approach added depth and offered diverging views on the implications (Ågerfalk, 2013) of bridewealth on male-perpetrated intimate partner violence. 193 University of Ghana http://ugspace.ug.edu.gh The 2014 Ghana Demographic and Health Survey examined bridewealth payment and men’s attitudes toward wife-beating while the Bridewealth Payment and Normative Constraints on Women’s Lives in Ghana Study focused on bridewealth payment and men’s self-reported actual perpetration of abuse against their female intimate partners. Further, the study relied on qualitative data gathered from the Bridewealth Payment and Normative Constraints on Women’s Lives in Ghana Study to understand participants’ perceptions of the implications of bridewealth payment and male-perpetrated intimate partner violence. The summary of the findings of the study, the conclusions drawn from it and the recommendations based on the findings are presented in the succeeding sections. 9.2 Summary 9.2.1 Bridewealth payment and men’s attitudes toward wife-beating Men’s attitudes toward wife-beating was examined on five domestic reasons; approval of “wife-beating if a woman goes out without telling her husband, neglects the children, argues with her husband, refuses to have sex with her husband and burns food”. A composite variable which assessed approval of wife-beating on at least one of the five reasons above was also examined. The univariate results showed that men’s approval of wife-beating was relatively prevalent for some reasons but very low regarding others. About 6 percent of men approve of wife-beating if a woman neglects the children while just about 2 percent of men approve of wife-beating if a woman burns food. One in every ten of the men approve of wife-beating on at least one reason. The practice of bridewealth payment is rife. More than eight in ten men stated that bridewealth was negotiated in their current union, and 68 percent stated that bridewealth has been fully paid. 194 University of Ghana http://ugspace.ug.edu.gh The mean age of men in the sample was 41.3 years. More than two-thirds of the study sample reported to be Christians. A large proportion of men (two in five) had middle or junior secondary school (JSS) education, and 18.8 percent had no formal education. More than seven out of ten men stated that they earned cash only, and less than one in ten indicated that they earned no income. About 45.0 percent of the men had one to four children. Nearly equal proportions of men lived in urban and rural areas. A large proportion of men identified as Akan (44%) and about nine in ten men were in monogamous unions. At the bivariate level, there were no statistically significant association between bridewealth negotiation status and men’s attitudes toward wife-beating, and completeness of bridewealth payment and men’s attitudes toward wife-beating. Individual level factors (religious affiliation, level of education, type of earning and number of children ever born), and community level factors (ethnicity, type of marriage and household wealth) were significantly associated with approval of wife-beating for all domestic reasons for approval of wife-beating (p- value<0.05). A high proportion of men who identified with ‘Other’ religions (traditional and no religion) approve of wife-beating. As men’s level of education increased, the proportion who approve of wife-beating decreased. A lower proportion of men who stated that they earn cash only approve of wife-beating. With regard to number of children ever born, a lower proportion of men who had no children approve of wife-beating. The proportion of men who approve of wife- beating significantly decreased with increasing household wealth. Compared with Akan and Ga- Dangbe/Ewe men, a higher proportion of men who belong to the Mole-Dagbani and ‘Other’ ethnic groups approve of wife-beating. Place of residence was not statistically significantly associated with approval of wife-beating if a woman argues with her husband. Age and duration 195 University of Ghana http://ugspace.ug.edu.gh of marriage were not significant predictors of approval of wife-beating on all reasons for wife- beating that measured men’s attitudes toward wife-beating at the bivariate level of analyses. Using binary logistic regression models, the study examined the predictors of men’s approval of wife-beating. There was no statistically significant relationship between bridewealth negotiation status and men’s approval of wife-beating. Similarly, completeness of bridewealth payment was not a statistically significant predictor of men’s approval of wife-beating. Consistently, religious affiliation was associated with men’s approval of “wife-beating if a woman goes out without telling her husband, neglects the children, argues with her husband, refuses to have sex with her husband and on at least one of the five reasons”. The results further show that the men who are affiliated with the ‘Other” religious group were significantly more likely to approve of wife-beating compared to their Christian counterparts. Type of earning was significantly associated with men’s approval of “wife-beating if a woman goes out without telling her husband, neglects the children, argues with her husband, and on at least one of the five reasons”. Compared with men who did not earn any pay/income, men who earned cash only were less likely to approve of wife-beating. According to Rani and Bonu (2009), men who were not working were less likely to believe that wife-beating is justified compared to men who were engaged in unpaid jobs. Men who indicated that they have no formal education were significantly more likely to approve of wife-beating if a woman refuses to have sex with her husband and on at least one reason compared to their counterparts who had junior secondary education. Ethnicity was a significant predictor of approval of wife-beating in three out of five scenarios studied. Men who identified with ‘Other’ ethnic groups were more likely to approve of wife-beating compared with Akan men. With every unit increase in age, men were more likely to approve of wife-beating on at least one reason. 196 University of Ghana http://ugspace.ug.edu.gh 9.2.2 Bridewealth payment and men’s self-reported actual perpetration of violence Self-reported perpetration of violence was prevalent as one in five men stated that they have ever beaten their female partners in selected communities in the Eastern Region. According to the Ghana Family Life and Health Survey 2016 report, the Eastern Region reported the highest prevalence of physical abuse perpetration compared to other regions (Institute of Development Studies, Ghana Statistical Service, & Associates, 2016). Some studies in other countries have also reported higher or lower prevalence rates. In the Democratic Republic of Congo for example, lifetime physical violence perpetration was as high as 45 percent (Fleming et al., 2015) while in Tanzania 13.2 percent of the participants reported physically abusing their partners at least once in the twelve months preceding the survey (Mulawa, et al., 2018). Similar to the 2014 GDHS, the majority (89.9%), of the respondents stated that bridewealth was negotiated and has been fully paid (56.0%) for in their current union. The mean age of respondents was 47.3 years and on average, respondents had been married for about 16 years. Majority of the respondents were Christians (Orthodox Christians, and Pentecostals/Charismatics). More than half the sample had some JHS schooling and a little over eight in ten were employed. Nearly half (49%) of the men had 2 to 4 children. About one in every five men had ever observed violence between their parents or caregivers as children. The bivariate results revealed no statistically significant relationship between bridewealth negotiation status and actual perpetration of violence. However, completeness of bridewealth payment was significantly (p-value<0.05) associated with actual perpetration of physical abuse. The proportion of men who stated that they have ever beaten their wives decreased with completeness of payment. While about one in every three (32.17%) who stated that bridewealth 197 University of Ghana http://ugspace.ug.edu.gh has not been paid have ever beaten their wives, less than one in five (17.28%) who fall within the full bridewealth payment category have ever beaten their wives. Level of education, observing violence as a child and age were significantly associated with actual perpetration of violence. About one in every four of the men who stated that they had JSS education have ever beaten their wives. Further, about one in three men who had observed inter-parental violence have ever beaten their wives. The results revealed a significant mean age difference between men who have ever beaten their wives and those who had never. The multivariate results indicated no significant association between bridewealth negotiation status and actual perpetration of violence. On the other hand, completeness of bridewealth payment was significantly associated with actual perpetration of violence. Compared with men who reported that bridewealth has been paid in full, those who stated that bridewealth has not been paid at all were more than two times as likely to have ever beaten their wives. Individual level factors such as religious affiliation, level of education, number of children ever born, age, observing violence as a child, and duration of marriage were significant predictors of self-reported violence perpetration. Pentecostal/charismatic men were significantly less likely to report that they have ever beaten their wives compared with Orthodox Christian men. Men who had secondary or higher education were less likely to have ever beaten their wives compared to men with junior secondary education. With increasing age, men were less likely to have ever beaten their wives. However, increasing duration of marriage predicted violence perpetration. Having 0 to 1 children reduced the odds of perpetrating abuse compared with having 2 to 4 children. Men who observed violence between their parents or caregivers as children were 1.79 times as likely to have perpetrated violence as those who did not observe such violence. 198 University of Ghana http://ugspace.ug.edu.gh Compared with Matrilineal Akan men, Patrilineal Akan and Matrilineal Guan men were significantly more likely to have ever beaten their wives. The findings from the interviews with 16 key informants indicate that the practice of bridewealth payment has complex implications for violence perpetration in these contexts. Using thematic analysis, the qualitative data explored the benefits of bridewealth payment, implications of full payment, partial payment and no payment, and finally, strategies/sanctions for dealing with abuse. With regard to benefits of bridewealth payment, six main themes emerged. It could be inferred that these social and moral benefits proffer some protection from abuse and reduce the incidence of conflicts between couples, and between their families. For example, bridewealth is paid to honour parents and the couple, and also to legitimize the union. The participants indicated that bridewealth is paid to protect women from abuse and from unlawful divorce. Further, this widespread cultural practice has implications for inheritance through property sharing, naming of children and old age social security for parents. These have consequences for reducing or encouraging conflicts and tensions regarding ownership and property sharing. Bridewealth payment provides some economic support and security for the woman and this could potentially reduce the woman’s dependence on her partner. Again, bridewealth payment shows the importance/value of the woman not only to her family but also to her husband. The implications of completeness of bridewealth payment were both negative and positive. Generally, the qualitative data showed that when bridewealth is not paid or is partially paid, the implications are detrimental. Non-payment and partial payment breed conflicts and could be potential pointers for conflicts between partners and even their families. Non-payment 199 University of Ghana http://ugspace.ug.edu.gh and partial payment imply that the man does not value his partner, the couple are not traditionally and legally married, and the couple may lack parental support. Also, there would be limited social control mechanisms available should the woman cheat or insult her husband. This suggests that when bridewealth is paid, the social control developed potentially reduces women’s ability to indulge in such behaviours which have been shown to be associated with intimate partner violence perpetration. Incomplete bridewealth payment also means that the children do not belong to the man, thus have no share in his property. When bridewealth is not paid, men lack authority in the union, would be disregarded by their partners and may be more likely to beat their partners than those who have paid the bridewealth. The implications of full bridewealth payment are multifaceted. The results show that bridewealth payment puts the man in a position of responsibility, ensuring that his wife and family are provided for and protected rather than gives him the authority to abuse his wife. Thus, full bridewealth payment does not mean that the woman has been bought and does not warrant the use of violence against the woman. On the contrary, the narratives suggested that the man has the authority to reprimand his wife if she does not perform certain duties. Again, although all the participants indicated that bridewealth payment does not mean the man has the authority to abuse his wife, a few participants suggested again that the woman can be beaten by her husband if she refuses to have sex with him. Further, the interviews showed that when bridewealth is paid in full, the social support and social network developed from parents, friends, queen mothers, chiefs and other relatives may prevent men from being violent or give the couple access to individuals they can rely on when they have marital challenges. Access to and support from these individuals suggests that they all have an interest in the success of the union. Some participants stated that irrespective of bridewealth payment, the husband has no authority or right to abuse his wife. 200 University of Ghana http://ugspace.ug.edu.gh The strategies for dealing with abuse in intimate unions were either formal or informal. The formal strategies were divorce from an abusive intimate partner and relying on formal state institutions like the police to arrest the abusive partner. The informal strategies included separation from an abusive partner, warning the abusive partner and denying the husband sex. Traditionally, other conflict resolution tactics would involve paying fines (cash or kind), apologizing to the victim or disgracing/shaming the offender or being summoned by the partner before the elders. 9.3 Recommendations This study empirically tests the association between bridewealth payment and male- perpetrated intimate partner violence using a mixed research methods approach. The findings from the study indicate that there is a need to expand the concept of intimate partner violence. In this study, intimate partner violence was examined at two levels; men’s attitudes toward wife- beating and self-reported actual perpetration of violence. The results showed that in examining attitudes toward wife-beating, the scenarios created should be context-specific and relevant. For example, traditional surveys like the Demographic and Health Surveys have examined attitudes toward wife-beating on the five scenarios studied. Expanding or revising the scope of measures used to examine attitudes towards wife-beating would add to our understanding of the various ways through which attitudes could predict actual use of violence. Future studies could explore more culturally sensitive scenarios like men’s attitudes toward wife-beating if a woman cheats on her husband, aborts their pregnancy or uses contraception without discussing it with her husband. This recommendation is also based on the findings of the qualitative study which show 201 University of Ghana http://ugspace.ug.edu.gh that violence against women could be approved if a woman refuses to have sex with her husband or uses contraception without her partner’s knowledge. Further, examining attitudes toward wife-beating and actual use of violence relying on a single dataset better predicts violence against women. Continuity and consistency of data collection are essential if large surveys like the Demographic and Health Surveys seek to better predict important outcomes like violence used against women, study the phenomena over time and give governments and other relevant institutions better research evidence to formulate more pertinent interventions in sub-Saharan Africa. In the 2008 Ghana Demographic and Health Survey, domestic violence models were included for the first time, in addition to the attitudes toward wife-beating model. However, in the 2014 GDHS, all modules on domestic violence were removed and only attitudes toward wife-beating model were maintained. This suggests that researchers and government institutions are limited in their ability to measure these important outcomes over time. It is recommended that future studies explore these concepts using single datasets. Also, the study limits the definition of actual use of violence to whether a man has ever beaten his wife. Focusing on other expressions of IPV such as emotional or psychological, sexual and economic violence could be relevant in addressing issues of intimate partner violence. Again, different items could be used to gather information on these broad categories of types of violence. For example, physical violence has been conceptualized as “slapping, pushing, hitting with a fist kicking/dragging/beating/choking/burning, or threatening or using a weapon” against a female partner (Fleming et al., 2015; García-Moreno et al., 2006). In addition, the results from the Bridewealth Study suggest that the implications of bridewealth payment are complex. To continue to unpack these implications, first, studies that 202 University of Ghana http://ugspace.ug.edu.gh explore different contexts would be beneficial. This study focuses on a non-representative sample from the southern part of Ghana. Investigating other contexts or national level datasets as the nature and amount of bridewealth paid are different in different contexts could add depth to the study of bridewealth payment and male-perpetrated violence against women. Secondly, in order to have better prediction power, future studies could consider larger sample sizes and longitudinal studies that provide information on intimate partner violence over time. Furthermore, the study shows that bridewealth payment is relevant for studies on masculinities. Focusing on how bridewealth payment or non-payment influences masculinities and how both men and women negotiate these identities could potentially broaden our understanding of the literature on male-perpetrated intimate partner violence in the sub-region. Future studies could also explore both males and female characteristics in the discourse on violence in intimate unions as this will provide more nuances in the study of male-perpetrated intimate partner violence in sub-Sahara Africa. The study shows that the youngest men were more likely to be violent (that is either favour the wife-beating and actually perpetuate abuse). It is important that studies focus on this sample of men in addressing issues of violence and aggression against women in the sub-region. Educational programs that are transformative and challenge the adherence to conservative tenets would be useful in this regard. Policies and interventions on intimate partner violence should focus on how culture frames power dynamics in unions. Also, this study shows that access to both social and economic resources potentially reduce men’s acceptance and use of violence. Interventions that expand access to such transformative resources could eventually curb the use and acceptance of violence against women in intimate unions. 203 University of Ghana http://ugspace.ug.edu.gh 9.4 Conclusion The study concludes that the implications of bridewealth payment for male-perpetrated intimate partner violence are multifaceted. It showed that relying on a modified version of the integrated ecological framework, the resource theory and the exchange/social control theory, and mixing methods was appropriate. This study has demonstrated that context and cultural processes, in addition to other socio-demographic and economic variables are important in understanding intimate partner violence in sub-Saharan Africa. Men’s approval of wife-beating and actual perpetration of violence against women in intimate unions is prevalent. Using the integrated ecological framework, the study concludes that both individual and community level variables predict men’s approval of wife-beating and actual perpetration of violence against their intimate partners. Individual level variables that predict men’s approval of wife-beating include religion, age, type of earning, number of children ever born and education are predictors of men’s attitude toward wife-beating. At the community level variables such as ethnicity further predicts men’s attitude toward wife-beating. Regarding actual perpetration of violence, the study showed that about one in five men had ever beaten their wives. This is consistent with findings from a recent study which reported that the Eastern Region recorded the highest prevalence of intimate partner violence (IDS, GSS, & Associates, 2016). Bridewealth negotiation status is not associated with men’s use of violence against their intimate partners. However, bridewealth non-payment increased the risk of male- perpetrated violence compared to full bridewealth payment. Individual level factors like religious affiliation, number of children ever born, age, observing violence as a child and duration of marriage were significant predictors of men’s use of violence. At the community level, ethnicity 204 University of Ghana http://ugspace.ug.edu.gh and completeness of bridewealth payment significantly predict male-perpetrated intimate partner violence. The study also shows that the resource and social control theories provide context and explain how men’s access to resources potentially avert their use of violence against their intimate partners. Interventions that explore men’s access to both economic/social resources and ethnic and religious underpinnings of violence against women would be paramount in addressing intimate partner violence in Ghana. The study further showed that the struggle over who has legitimate marital power and control could influence spousal conflict. It is important to note that women in traditional societies may contend incomplete bridewealth payment as this has implications for child ownership and property sharing in the instance where their intimate partner dies, for example. These are triggers of conflict as couples may argue about issues regarding legitimizing their unions, and the man’s legitimacy to authority in the union. Non-payment of bridewealth could imply that the man does not have the necessary resources to legitimize the union, and this could have implications for his reduced or unrecognized authority in the union. As the resource theory suggests, this could lead to his use of violence. This is highlighted in the instance where no payment of bridewealth is associated with male-perpetrated intimate partner violence when men rely on violence as an alternative resource to gain control. When bridewealth is paid in full, however, it places the man in a position of responsibility, ensuring that his wife and family are safe from harm. Traditionally, bridewealth payment does not give the man leeway to use violence against his wife. This is because traditionally, the exchange structures and processes associated with bridewealth payment do not give the man the authority to abuse his wife as bridewealth payment should rather honour the 205 University of Ghana http://ugspace.ug.edu.gh families of the couple, the man and his wife. Central to the findings of this study is that the social support and control developed when bridewealth is paid may prevent men from being violent. Community values of ensuring healthy marriages go a long way to potentially reduce actual use of violence against women in intimate unions. Further, the study argues that both men and women are active actors in the exchange process. When bridewealth is paid, women recognizing their position, rights and responsibilities in the union may be less likely to engage in activities that may put them at risk of abuse. The study concludes that in an exchange relationship such as bridewealth payment, the social process and the structures associated with the exchange ensure some social control which regulates the use of violence in among couples. While bridewealth payment proffers some protection from violence, this exchange may still put women at risk of male-perpetrated intimate partner violence. The study demonstrates that on issues of contraceptive use and a woman challenging the authority of his partner when bridewealth has been paid, participants indicated that a husband may be justified in his use of violence against his wife as a form of discipline. This is possible because bridewealth payment in itself gives the man some level of authority over his female partner. This reflects an inherent perception that husbands have power over their wives. This authority is deepened because bridewealth exchange transfers rights to women reproductive, sexual and domestic services to men. Finally, formal and informal sanctions are also key intervention points that could be streamlined to punish or reform abusers, prevent repeat abuse, and also, deter other individuals from being violent. The study concludes that in order for researchers and policy makers to understand and develop programs that seek to reduce intimate partner violence against women in 206 University of Ghana http://ugspace.ug.edu.gh sub-Saharan Africa, it is important to rely on an integrated framework, a mixed-methods design, and provide context by examining how socio-demographic/economic variables and culture frame violence against women. 207 University of Ghana http://ugspace.ug.edu.gh References Aborampah, O.-M. (1999). Systems of kinship and marriage in Africa. In S. L. Browning & R. R. Miller. (Eds.), Till Death do us part: A multicultural anthology on marriage. Stamford: CT: JAI Press. Abrahams, N., Jewkes, R., Martin, L. J., Mathews, S., Vetten, L., & Lombard, C. (2009). Mortality of women from intimate partner violence in South Africa: A national epidemiological study. Violence and Victims, 24(4), 546–556. https://doi.org/10.1891/0886- 6708.24.4.546 Abramsky, T., Watts, C. H., García-Moreno, C., Devries, K., Kiss, L., Ellsberg, M., … Heise, L. (2011). What factors are associated with recent intimate partner violence? Findings from the WHO multi-country study on women’s health and domestic violence. BMC Public Health, 11(1), 109. https://doi.org/10.1186/1471-2458-11-109 Abu, K. (1983). The separateness of spouses: Conjugal resources in an Ashanti town. In C. Oppong (Ed.), Female and male in West Africa (pp. 156–168). Adebowale, A. S. (2018). Spousal age difference and associated predictors of intimate partner violence in Nigeria. BMC Public Health, 18(1), 1–16. https://doi.org/10.1186/s12889-018- 5118-1 Adinkrah, M. (2008). Spousal homicides in contemporary Ghana. Journal of Criminal Justice, 36, 209–216. https://doi.org/10.1016/j.jcrimjus.2008.04.002 Adjah, E. S. O., & Agbemafle, I. (2016). Determinants of domestic violence against women in Ghana. BMC Public Health, 16(368), 1–9. https://doi.org/10.1186/s12889-016-3041-x Adjei, S. B. (2015a). “Correcting an erring wife is normal”: Moral discourses of spousal violence in Ghana. Journal of Interpersonal Violence, 1–22. https://doi.org/10.1177/0886260515619751 Adjei, S. B. (2015b). Partner dependency and intimate partner abuse: A sociocultural grounding of spousal abuse in Ghana. Psychological Studies, 60(4), 422–427. https://doi.org/10.1007/s12646-015-0336-4 Adjei, S. B. (2016). Masculinity and spousal violence: Discursive accounts of husbands who abuse their wives in Ghana. Journal of Family Violence, 31(4), 411–422. https://doi.org/10.1007/s10896-015-9781-z Adjei, S. B. (2017). Sociocultural groundings of battered women’s entrapment in abusive marital relationship in Ghana. Journal of Aggression, Maltreatment & Trauma, 26(8), 879–901. https://doi.org/10.1080/10926771.2017.1284171 Adjei, S. B. (2018). The social intentionality of battered women’s agency in Ghana. Psychology and Developing Societies, 30(1), 1–18. https://doi.org/10.1177/0971333617747320 Adomako Ampofo, A. (1993). Controlling and punishing women: Violence against Ghanaian women. Review of African Political Economy, Challenging Gender Inequalities in Africa, (56), 102–111. 208 University of Ghana http://ugspace.ug.edu.gh Adomako Ampofo, A. (2001). “When men speak women listen”: Gender socialisation and young adolescents’ attitudes to sexual and reproductive issues. African Journal of Reproductive Health, 5(3), 196–212. Adomako Ampofo, A. (2008). Collective activism: The domestic violence bill becoming law in Ghana. African and Asian Studies, 7(4), 395–421. https://doi.org/10.1163/156921008X359597 Ågerfalk, P. J. (2013). Embracing diversity through mixed methods research. European Journal of Information Systems, 22(3), 251–256. https://doi.org/10.1057/ejis.2013.6 Agresti, A. (2008). An introduction to categorical data analysis. Pharmaceutical Statistics (Second Edi, Vol. 7). Hoboken, New Jersey: John Wiley & Sons, Inc. https://doi.org/10.1002/pst.339 Al-nouri, Q. N. (1980). Changing marriage patterns in Libya: Attitudes of university students. Journal of Comparative Family Studies, 11(2), 219–232. Al Riyami, A., Afifi, M., & Mabry, R. M. (2004). Women’s autonomy, education and employment in Oman and their influence on contraceptive use. Reproductive Health Matters, 12(23), 144–154. https://doi.org/10.1016/S0968-8080(04)23113-5 Alangea, D. O., Addo-Lartey, A. A., Sikweyiya, Y., Chirwa, E. D., Coker-Appiah, D., Jewkes, R., & Adanu, R. M. K. (2018). Prevalence and risk factors of physical or sexual intimate violence perpetration amongst men in four districts in the Central Region of Ghana: Baseline findings from a cluster randomised controlled trial. PLoS ONE, 13(3), 1–20. https://doi.org/10.1371/journal.pone.0191663 Alesina, A., Brioschi, B., & Ferrara, E. La. (2016). Violence against women: A cross-cultural analysis for Africa (No. 21901). Massachusetts Avenue, Cambridge. Allen, C., & Straus, M. A. (1979). Resources, power, and husband-wife violence. (M. A. Straus & G. Hotaling, Eds.), Social causes of husband-wife violence. Minneapolis, MN: University of Minnesota Press. Allman, J. M., & Tashjian, V. B. (2000). “I will not eats stone”: A women’s history of colonial Asante. Portsmouth, NH: Heinemann. Amoakohene, M. I. (2004). Violence against women in Ghana: A look at women’s perceptions and review of policy and social responses. Social Science & Medicine, 59, 2373–2385. https://doi.org/10.1016/j.socscimed.2004.04.001 Anderson, K. L. (1997). Gender, status, and domestic diolence: An integration of feminist and family violence approaches. Journal of Marriage and the Family, 59(3), 655. https://doi.org/10.2307/353952 Anderson, S. (2007). The economics of dowry and brideprice. Journal of Economic Perspectives, 21(4), 151–174. https://doi.org/10.1257/jep.21.4.151 Ansell, N. (2001). “Because it’s our culture!” (Re) negotiating the meaning of Lobola in Southern African secondary schools. Journal of Southern African Studies, 27(4), 697–716. 209 University of Ghana http://ugspace.ug.edu.gh https://doi.org/10.1080/0305707012009069 Ansell, N., Hajdu, F., Blerk, L. Van, & Robson, E. (2017). ‘My happiest time’ or ‘my saddest time’? The spatial and generational construction of marriage among youth in rural Malawi and Lesotho. Transactions of the Institute of British Geographers, 1–15. https://doi.org/10.1111/tran.12211 Asante, F., Fenny, A., Dzudzor, M., Chadha, M., Scriver, S., Ballantine, C., & N., D. (2019). Economic and social costs of violence against women and girls in Ghana: Country Technical Report. Galway: NUI Galway. Asiedu, C. (2016). Lineage ties and domestic violence in Ghana: Evidence from the 2008 Demographic and Health Survey. Journal of Family Issues, 37(16), 2351–2367. https://doi.org/10.1177/0192513X14561523 Atkinson, M. P., Greenstein, T. N., & Lang, M. M. (2005). For women, breadwinning can be dangerous: Gendered resource theory and wife abuse. Journal of Marriage and Family, 67(5), 1137–1148. https://doi.org/10.1111/j.1741-3737.2005.00206.x Attride-Stirling, J. (2001). Thematic networks: An analytic tool for qualitative research. Qualitative Research, 1(3), 385–405. Babbie, E. (2015). The practice of social research. Cengage Learning. Babcock, R. L., & DePrince, A. P. (2013). Factors contributing to ongoing intimate partner abuse: Childhood betrayal trauma and dependence on one’s perpetrator. Journal of Interpersonal Violence, 28(7), 1385–1402. https://doi.org/10.1177/0886260512468248 Bahamonde-Birke, F. J., Kunert, U., Link, H., & Ortúzar, J. de D. (2017). About attitudes and perceptions: Finding the proper way to consider latent variables in discrete choice models. Transportation, 44(3), 475–493. https://doi.org/10.1007/s11116-015-9663-5 Bandura, A. (1973). Aggression: A social learning analysis. Oxford, England: Prentice-Hall. Bandura, A. (1977). Social learning theory. Englewood Cliffs, NJ: Prentice-hall. Bankole, A., & Singh, S. (1998). Couples’ fertility and contraceptive decision-making in developing countries: Hearing the man’s voice. International Family Planning Perspectives, 24(1), 15. https://doi.org/10.2307/2991915 Barnett, O. W. (2001). Barnett, O. W. (2001). Why battered women do not leave, part 2: External inhibiting factors—social support and internal inhibiting factors. Trauma, Violence, & Abuse, 2(1), 3-35. Bawah, A. A., Akweongo, P., Simmons, R., & Phillips, J. F. (1999). Women’s fear and men’s anxieties: The impact of family planning on gender relations in Northern Ghana. Studies in Family Planning, 30(1), 54–66. Becker, K. D., Stuewig, J., & McCloskey, L. A. (2010). Traumatic stress symptoms of women exposed to different forms of childhood victimization and intimate partner violence. Journal of Interpersonal Violence, 25(9), 1699–1715. https://doi.org/10.1177/0886260509354578 210 University of Ghana http://ugspace.ug.edu.gh Becker, S., & Kalamar, A. (2018). Sampling weights for analysis of couple data in Demographic and Health Surveys. Demography, 55, 1–15. Beeble, M. L., Bybee, D., Sullivan, C. M., & Adams, A. E. (2009). Main, mediating, and moderating effects of social support on the well-being of survivors of intimate partner violence across 2 years. Journal of Consulting and Clinical Psychology, 77(4), 718–729. https://doi.org/10.1037/a0016140 Belsky, J. (1980). Child maltreatment: An ecological integration. American Psychological Association, 4, 320–335. https://doi.org/10.1037/0003-066X.35.4.320 Beydoun, H. A., Al-Sahab, B., Beydoun, M. A., & Tamim, H. (2010). Intimate partner violence as a risk factor for postpartum depression among Canadian women in the maternity experience survey. Annals of Epidemiology, 20(8), 575–583. https://doi.org/10.1016/j.annepidem.2010.05.011 Beydoun, H. A., Beydoun, M. A., Kaufman, J. S., Lo, B., & Zonderman, A. B. (2012). Intimate partner violence against adult women and its association with major depressive disorder, depressive symptoms and postpartum depression: A systematic review and meta-analysis. Social Science and Medicine, 75(6), 959–975. https://doi.org/10.1016/j.socscimed.2012.04.025 Beydoun, H. A., Williams, M., Beydoun, M. A., M, S., M. A., E., & Zonderman, A. B. (2017). Relationship of physical intimate partner violence with mental health diagnoses in the nationwide emergency department sample. Journal of Women’s Health, 26(2), 141–151. https://doi.org/10.1089/jwh.2016.5840 Blanc, A. K., Wolff, B., Gage, A. J., Ezeh, A. C., & Ssekamatte-Ssebuliba, J. (1996). Negotiating reproductive outcomes in Uganda. Calverton, Maryland: Macro International Inc. and Institute of Statistics and Applied Economics [Uganda]. Bongaarts, J., & Casterline, J. (2013). Fertility transition: Is sub-Saharan Africa different? Population and Development Review, 38(1), 153–168. Boserup, E. (1985). Economic and demographic interrelationships in sub-Saharan Africa. Population and Development Review, 11(3), 383–397. Bowman, C. G. (2003). Domestic violence: Does the African context demand a different approach? International Journal of Law and Psychiatry, 26, 473–491. https://doi.org/10.1016/S0160-2527(03)00082-7 Breiding, M., Basile, K., Smith, S., Black, M., & Mahendra, R. (2015). Intimate partner violence surveillance: Uniform definitions and recommended data elements, version 2.0. Atlanta (GA). https://doi.org/10.2307/1319341 Bronfenbrenner, U. (1979). The ecology of human development. Cambridge, MA: Harvard University Press. Bucheli, M., & Rossi, M. (2017). Attitudes toward partner violence and gender roles in Uruguayan women. Journal of Interpersonal Violence, 32(23), 3693 –3705. 211 University of Ghana http://ugspace.ug.edu.gh https://doi.org/10.1177/0886260515602897 Caman, S., Kristiansson, M., Granath, S., & Sturup, J. (2017). Trends in rates and characteristics of intimate partner homicides between 1990 and 2013. Journal of Criminal Justice, 49, 14– 21. https://doi.org/10.1016/j.jcrimjus.2017.01.002 Campbell, J. C. (2002). Health consequences of intimate partner violence. Lancet, 359(9314), 1331–1336. https://doi.org/10.1016/S0140-6736(02)08336-8 Cantalupo, N. C., Martin, L. V., Pak, K., & Shin, S. (2006). Domestic violence in Ghana: The open secret. Georgetown Journal of Gender and the Law, 7, 531–597. Carney, M. M., & Barner, J. R. (2012). Prevalence of partner abuse: Rates of emotional abuse and control. Partner Abuse, 3(3), 1–48. https://doi.org/10.1891/1946-6560.3.3.e5 Chukwukere, I. (1978). Akan theory of conception. Are the Fante really aberrant? African Journal of the International African Institute, 48(2), 135–148. Coker, A. L., Smith, P. H., Thompson, M. P., Mckeown, R. E., Bethea, L., & Davis, K. E. (2002). Social support protects against the negative effects of partner violence on mental health, 11(5). Começanha, R., Basto-Pereira, M., & Maia, Â. (2017). Clinically speaking, psychological abuse matters. Comprehensive Psychiatry, 73, 120–126. https://doi.org/10.1016/j.comppsych.2016.11.015 Connell, R. (1987). Gender and power: Society, the person, and sexual politics. Stanford, CA: Stanford University Press. Retrieved from http://isbn.im/download/0804714290.pdf Cools, S., & Kotsadam, A. (2017). Resources and intimate partner violence in sub-Saharan Africa. World Development, 95, 211–230. https://doi.org/10.1016/j.worlddev.2017.02.027 Costa, E. C. V, & Gomes, S. C. (2018). Social support and self-esteem moderate the relation between intimate partner violence and depression and anxiety symptoms among Portuguese women. Journal of Family Violence, 33, 355–368. Courtenay, W. H. (2000). Constructions of masculinity and their influnece on men’s well-being: A theory of gender and health. Social Science & Medicine, 50, 1385–1401. Cunradi, C. B., Caetano, R., & Schafer, J. (2002). Alcohol-related problems, drug use, and male intimate partner violence severity among US couples. Alcohol Clinical and Experimental Research, 26(4), 493–500. Dalal, K., Lee, M. S., & Gifford, M. (2012). Male adolescents’ attitudes toward wife beating: A multi-country study in South Asia. Journal of Adolescent Health, 50(5), 437–442. https://doi.org/10.1016/j.jadohealth.2011.09.012 de Bruijn, D. M., & de Graaf, I. M. (2016). The role of substance use in same-day intimate partner violence: A review of the literature. Aggression and Violent Behavior, 27, 142–151. https://doi.org/10.1016/j.avb.2016.02.010 Decker, M. R., Seage, G. R., Hemenway, D., Gupta, J., Raj, A., & Silverman, J. G. (2009). 212 University of Ghana http://ugspace.ug.edu.gh Intimate partner violence perpetration, standard and gendered STI/HIV risk behaviour, and STI/HIV diagnosis among a clinic-based sample of men. Sexually Transmitted Infections, 85, 555–560. https://doi.org/10.1136/sti.2009.036368 Dekker, M., & Hoogeveen, H. (2002). On bride wealth and household security in rural Zimbabwe. Journal of African Economies, 11(1), 114–145. https://doi.org/10.1093/jae/11.1.114 Delsol, C., & Margolin, G. (2004). The role of family-of-origin violence in men’s marital violence perpetration. Clinical Psychology Review, 24(1), 99–122. https://doi.org/10.1016/j.cpr.2003.12.001 DeRose, L. F., Dodoo, F. N.-A., & Patil, V. (2002). Schooling and attitudes on reproductive- related behaviour in Ghana. International Journal of Sociology of the Family, 30(1), 51–66. Dery, I. (2015). Bride-price and domestic violence: Empirical perspectives from Nandom District in the North Western region of Ghana. International Journal of Development and Sustainability, 4(3), 258–271. Dery, I., & Diedong, A. L. (2014). Domestic violence against women in Ghana: An exploratory study in Upper West Region, Ghana. International Journal of Humanities and Social Science, 4(12), 228–244. Deshotels, T. H., Forsyth, C. J., Earwood, S., New, B. A., & Fulmer, J. (2019). For HE tells me so: Techniques of neutralization applied to Christian domestic discipline. Deviant Behavior, 00(00), 1–20. https://doi.org/10.1080/01639625.2018.1519134 Devries, K., Watts, C., Yoshihama, M., Kiss, L., Schraiber, L. B., Deyessa, N., … García- Moreno, C. (2011). Violence against women is strongly associated with suicide attempts: Evidence from the WHO multi-country study on women’s health and domestic violence against women. Social Science and Medicine, 73(1), 79–86. https://doi.org/10.1016/j.socscimed.2011.05.006 Dillon, G., Hussain, R., Loxton, D., & Khan, A. (2016). Rurality and self-reported health in women with a history of intimate partner violence. PLoS ONE, 11(9), 1–12. https://doi.org/10.1371/journal.pone.0162380 Djamba, Y. K., Kimuna, S. R., & Aga, M. G. (2015). Socio-demographic factors associated with men’s attitudes toward wife beating in Ethiopia. In Y. K. Djamba & S. R. Kimuna (Eds.), Gender-based violence: Perspectives from Africa, Middle East and India (pp. 3–16). Switzerland: Springer International Publishing. https://doi.org/10.1007/978-3-319-16670- 4_1 Dobash, R. E., & Dobash, R. P. (1979). Violence against wives: A case against the patriarchy. New York: Free Press. Dobash, R. E., & Dobash, R. P. (1992). Women, violence, and social change. New York: Routledge. Dobash, R. P., Dobash, R. E., Wilson, M., & Daly, M. (1992). The myth of sexual symmetry in 213 University of Ghana http://ugspace.ug.edu.gh marital violence. Social Problems, 39(1), 71–91. https://doi.org/10.2307/3096914 Dodoo, F. N.-A. (1998a). Marriage type and reproductive decisions: A comparative study in sub- Saharan Africa. Journal of Marriage and Family, 60(1), 232–242. Dodoo, F. N.-A. (1998b). Men matter: Additive and interactive gendered preferences and reproductive behavior in Kenya. Demography, 35(2), 229–242. https://doi.org/10.2307/3004054 Dodoo, F. N.-A., & Frost, A. E. (2008). Gender in African population research: The fertility/ reproductive health example. Annual Review of Sociology, 34, 431–452. Dodoo, F. N.-A., Horne, C., & Biney, A. (2014). Does education mitigate the adverse impact of bridewealth on women’s reproductive autonomy? In Genus (Vol. LXX, pp. 77–97). https://doi.org/10.4402/genus-591 Dodoo, F. N.-A., Horne, C., & Dodoo, N. D. (2014). Lab experiments in demographic fieldwork: Understanding gender dynamics in Africa. Demographic Research, 31(47), 1417–1430. https://doi.org/10.4054/DemRes.2014.31.47 Dodoo, F. N.-A., & Seal, A. (1994). Explaining spousal differences in reproductive preferences: A gender inequality approach. Population and Environment, 15(5), 379–394. https://doi.org/10.1007/BF02208319 Dodoo, F. N.-A., & Tempenis, M. (2002). Gender, power, and reproduction: Rural-urban differences in the relationship between fertility goals and contraceptive use in Kenya. Rural Sociology, 67(1), 46–70. Dodoo, N. D., & Dodoo, F. N.-A. (2017). Bridewealth payment and actual contraceptive behavior in Uganda. Sociological Viewpoints, 31(1), 46–60. https://doi.org/10.26908/3112017016 Doku, D. T., & Asante, K. O. (2015). Women’s approval of domestic physical violence against wives: Analysis of the Ghana Demographic and Health Survey. BMC Women’s Health, 1–8. https://doi.org/10.1186/s12905-015-0276-0 Dunkle, K., Jewkes, R., Nduna, M., Levin, J., Jama, N., Khuzwayo, N., … Duvvury, N. (2006). Perpetration of partner violence and HIV risk behaviour among young men in the rural Eastern Cape, South Africa. AIDS, 20(16), 2107–2114. https://doi.org/10.1097/01.aids.0000247582.00826.52 Durevall, D., & Lindskog, A. (2015). Intimate partner violence and HIV in ten sub-Saharan African countries: What do the Demographic and Health Surveys tell us? The Lancet Global Health, 3(1), e34-43. https://doi.org/10.1016/S2214-109X(14)70343-2 Dutton, D. G. (2006). The abusive personality: Violence and control in intimate relationships. Guilford Press. Dutton, M. A. (2009). Pathways linking intimate partner violence and posttraumatic disorder. Trauma, Violence, & Abuse, 10(3). https://doi.org/10.1177/1524838009334451 214 University of Ghana http://ugspace.ug.edu.gh El Abani, S., & Pourmehdi, M. (2018). Gender and educational differences in perception of domestic violence against women among Libyan migrants in Manchester. Journal of Interpersonal Violence, 1–23. https://doi.org/10.1177/0886260518760006 Ellison, C. G., & Anderson, K. L. (2001). Religious involvement and domestic violence among U.S. couples. Journal for the Scientific Study of Religion, 40(2), 269–286. https://doi.org/10.1111/0021-8294.00055 Ellison, C. G., Trinitapoli, J. A., Anderson, K. L., & Johnson, B. R. (2007). Race/ethnicity, religious involvement, and domestic violence. Violence Against Women, 13(11), 1094– 1112. Ellsberg, M., Arango, D. J., Morton, M., Gennari, F., Kiplesund, S., Contreras, M., … Washington, G. (2015). Prevention of violence against women and girls: what does the evidence say? The Lancet, 385(9977), 1555–1566. https://doi.org/10.1016/S0140- 6736(14)61703-7 Ellsberg, M., Jansen, H. A., Heise, L., Watts, C. H., & García-Moreno, C. (2008). Intimate partner violence and women’s physical and mental health in the WHO multi-country study on women’s health and domestic violence: An observational study. The Lancet, 371, 1165– 1172. https://doi.org/10.1016/s0140-6736(08)60522-x Emery, C. R., Wu, S., Kim, O., Pyun, C., & Chin, W. W. (2017). Protective family informal social control of intimate partner violence in Beijing. Psychology of Violence, 7(4), 553– 562. https://doi.org/10.1037/vio0000063 Eriksson, L., & Mazerolle, P. (2015). A cycle of violence? Examining family-of-origin violence, attitudes, and intimate partner violence perpetration. Journal of Interpersonal Violence, 30(6), 945–964. https://doi.org/10.1177/0886260514539759 Ezeudu, C. C., Akpa, O., Waziri, N. E., Oladimeji, A., Adedire, E., Saude, I., … Fawole, O. I. (2019). Prevalence and correlates of intimate partner violence, before and during pregnancy among attendees of maternal and child health services, Enugu, Nigeria: Mixed method approach, January 2015. The Pan African Medical JournaL, 32(Supp 1), 1–6. https://doi.org/10.11604/pamj.supp.2019.32.1.13287 Fallers, L. A. (1957). Some determinants of marriage stability in Busoga: A reformulation of Gluckman’s hypothesis. Journal of the International African Institute, 27(2), 106–123. Fawole, O. I. (2008). Economic violence to women and girls. Is it receiving the necessary attention? Trauma, Violence and Abuse, 9(3), 167–177. https://doi.org/10.1177/1524838008319255 Fawole, O. I., Aderonmu, A. L., & Fawole, A. O. (2005). Intimate partner abuse: Wife beating among civil servants in Ibadan, Nigeria. African Journal of Reproductive Health, 9(2), 54. https://doi.org/10.2307/3583462 Fidan, A., & Bui, H. N. (2016). Intimate partner violence against women in Zimbabwe. Violence Against Women, 22(9), 1075–1096. https://doi.org/10.1177/1077801215617551 215 University of Ghana http://ugspace.ug.edu.gh Fleming, P. J., McCleary-Sills, J., Morton, M., Levtov, R., Heilman, B., & Barker, G. (2015). Risk factors for men’s lifetime perpetration of physical violence against intimate partners: Results from the International Men and Gender Equality Survey (IMAGES) in eight countries. PLoS ONE, 10(3), 1–18. https://doi.org/10.1371/journal.pone.0118639 Fortes, M. (1950). Kinnship and marriage among the Ashanti. In M. Fortes, A. R. Radcliffe- Brown, & D. Forde (Eds.), African systems of kinship and marriage. African systems of kinship and marriage (pp. 252–284). London, Oxford: University Press. Fortes, M. (1972). Kinship and the social order: The legacy of L.H. Morgan. Current Anthropology, 13(2), 285–296. Fortes, M., Steel, R. W., & Ady, P. (1947). Ashanti survey, 1945-46: An experiment in social research. The Geographical Journal, 110(4/6), 149–177. Frost, A. E., & Dodoo, F. N.-A. (2009). Men are missing from African family planning. Contexts, 8(1), 44–49. https://doi.org/10.1525/ctx.2009.8.1.44 Frost, A. E., & Dodoo, F. N.-A. (2010). “The man comes to marry the woman”: Exploring adolescent boys’ gendered expectations for bridewealth and marriage among the Akwapim of southern Ghana. Marriage and Family Review, 49(1–2), 41–59. https://doi.org/10.1080/01494921003648563 Frye, V., Ompad, D., Chan, C., Koblin, B., Galea, S., & Vlahov, D. (2011). Intimate partner violence perpetration and condom use-related factors: Associations with heterosexual men’s consistent condom use. AIDS and Behavior, 15(1), 153–162. https://doi.org/10.1007/s10461-009-9659-2 Gage-Brandon, A. J. (1992). The polygyny-divorce relationship: A case study of Nigeria. Journal of Marriage and the Family, 54(2), 285–292. Gage, A. J. (2016). Exposure to spousal violence in the family, attitudes and dating violence perpetration among high school students in Port-au-Prince. Journal of Interpersonal Violence, 31(14), 2445–2474. https://doi.org/10.1177/0886260515576971 García-Moreno, C., Jansen, H. A., Ellsberg, M., Heise, L., & Watts, C. H. (2006). Prevalence of intimate partner violence: Findings from the WHO multi-country study on women’s health and domestic violence. Lancet, 368, 1260–1269. https://doi.org/10.1016/S0140- 6736(06)69523-8 García-Moreno, C., Zimmerman, C., Morris-Gehring, A., Heise, L., Amin, A., Abrahams, N., & Oswaldo, M. (2015). Violence against women and girls 5-Addressing violence against women: A call to action. Lancet, 383, 1685–1695. https://doi.org/10.1016/S0140- 6736(14)61830-4 Gass, J. D., Stein, D. J., Williams, D. R., & Seedat, S. (2011). Gender differences in risk for intimate partner violence among South African adults. Journal of Interpersonal Violence, 26(14), 2764–2789. https://doi.org/10.1177/0886260510390960 Gelles, R. J. (1983). An exchange/social control theory. In D. Finkelhor, R. J. Gelles, G. T. 216 University of Ghana http://ugspace.ug.edu.gh Hotaling, & M. A. Straus (Eds.), The dark side of families: Current family violence research (pp. 151–165). Beverly Hills: CA: Sage. Gelles, R. J. (1993). Through a sociological lens: Social structure and family violence. In R. J. Gelles & D. R. Loseke (Eds.), Current controversies on family violence (pp. 31–46). Newbury Park: CA: Sage. Gelles, Richard. J, & Maynard, P. E. (1987). A structural family systems approach to intervention in cases of family violence. Family Relations, 36(3), 270–275. Gelles, Richard. J, & Straus, M. A. (1979). Determinants of violence in the family: Toward a theoretical integration. (W. R. Burr, R. Hill, I. F. Nye, & I. L. Reiss, Eds.). New York: Free Press. Ghana Statistical Service, (GSS). (2007). Ghana Multiple Indicator Cluster Survey, 2010. Accra, Ghana. Ghana Statistical Service, (GSS). (2013a). 2010 Population and housing census, regional analytical report-Eastern Region. Accra, Ghana. Ghana Statistical Service, (GSS). (2013b). Population and housing census, 2010. Accra, Ghana. Ghana Statistical Service, (GSS)., Ghana Health Service, (GHS)., & ICF-International. (2015). Ghana Demographic and Health Survey, 2014. Rockville, Maryland, USA:GSS,GHS, and ICF International. Ghana Statistical Service, (GSS)., Ghana Health Service, (GHS)., & ICF-Macro. (2009). Ghana Demographic and Health Survey, 2008. Accra, Ghana: GSS, GHS, and ICF Macro. Ghana Statistical Service, (GSS)., Institute for Resource Development, (IRD)., & Macro Systems, (Inc). (1989). Ghana Demographic and Health Survey. Columbia, Maryland USA: Accra. Ghana Statistical Service, (GSS)., & Macro-International-Inc, (MI). (1999). Ghana Demographic and Health Survey, 1998. Calverton, Maryland: GSS and MI. Ghana Statistical Service, (GSS)., Noguchi Memorial Institute for Medical Research, (NMIMR)., & ORC-Macro. (2004). Ghana Demographic and Health Survey, 2003. Calverton, Maryland: GSS, NMIMR, and ORC Macro. Ghana Statistical Service, (GSS). (2014). Interviewer’s Manual-Ghana Demographic and Health Survey, 2014. Accra, Ghana. Gibbs, A., Corboz, J., & Jewkes, R. (2018). Factors associated with recent intimate partner violence experience amongst currently married women in Afghanistan and health impacts of IPV: A cross sectional study. BMC Public Health, 18(1), 1–11. https://doi.org/10.1186/s12889-018-5507-5 Gil-González, D., Vives-Cases, C., Álvarez-Dardet, C., & Latour-Pérez, J. (2006). Alcohol and intimate partner violence: Do we have enough information to act? European Journal of Public Health, 16(3), 278–284. https://doi.org/10.1093/eurpub/ckl016 217 University of Ghana http://ugspace.ug.edu.gh Gluckman, M. (1950). Kinship and marriage among the Lozi of Northern Rhodesia and the Zulu of Natal. In A. Radcliffe-Brown & D. Forde (Eds.), African systems of kinship and marriage. London: Oxford University Press. Goode, W. J. (1971). Force and violence in the family. Journal of Marriage and the Family, 33(4), 624–636. https://doi.org/10.2307/349435 Goodman, L., Dutton, M. A., Weinfurt, K., & Cook, S. (2003). The intimate partner violence strategies index: Development and application. Violence Against Women, 9(2), 163–186. https://doi.org/10.1177/1077801202239004 Goody, J. (1973). “Bridewealth and dowry in Africa and Eurasia.” In J. Goody & S. J. Tambiah. (Eds.), Bridewealth and Dowry (pp. 1–58). Cambridge, UK: Cambridge University Press. Government of Ghana, (GoG). Criminal Code, 1960 (Act 29) (1960). Government of Ghana, (GoG). Domestic Violence Act, 2007 (Act 732) (2007). Government Printer. Assembly Press, Accra. GPC/AI47/300/4/2007. Graham, K., Bernards, S., Wilsnack, S. C., & Gmel, G. (2011). Alcohol may not cause partner violence but it seems to make it worse: A cross national comparison of the relationship between alcohol and severity of partner violence. Journal of Interpersonal Violence, 26(8), 1503–1523. https://doi.org/10.1177/0886260510370596 Green, D. P., Wilke, A., & Cooper, J. (2017). Reducing intimate partner violence through informal social control: A mass media experiment in rural Uganda. Green, E., Chesla, K., Beyene, Y., & Kools, S. (2018). Ecological factors that impact adjustment processes and development of Ugandan adolescent immigrant females. Journal of Child and Family Studies, 27(1), 34–46. https://doi.org/10.1007/s10826-017-0867-5 Gurmu, E., & Endale, S. (2017). Wife beating refusal among women of reproductive age in urban and rural. BMC International Health and Human Rights, 17(6), 1–12. https://doi.org/10.1186/s12914-017-0115-5 Haj-Yahia, M. M. (1998). A patriarchal perspective of beliefs about wife beating among Palestinian men from the West Bnak and the Gaza Strip. Journal of Family Issues, 19(5), 595–621. Haj-Yahia, M. M. (2003). Beliefs about wife beating among Arab men from Israel: The influence of their patriarchal ideology. Journal of Family Violence, 18(4). https://doi.org/10.1177/0959353516647071 Haj-Yahia, M. M., & Uysal, A. (2008). Beliefs about wife beating among medical students from Turkey. Journal of Family Violence, 23(2), 119–133. https://doi.org/10.1007/s10896-007- 9134-7 Hakansson, N. T. (1990). Descent bridewealth and terms of alliance in East African societies. Research in Economic Anthropology, 12, 149–173. Hatcher, A. M., Colvin, C. J., Ndlovu, N., & Dworkin, S. L. (2014). Intimate partner violence 218 University of Ghana http://ugspace.ug.edu.gh among rural South African men: Alcohol use, sexual decision-making, and partner communication. Cult Health Sex, 16(9), 1023–1039. https://doi.org/10.1080/13691058.2014.924558 Hatcher, A. M., Romito, P., Odero, M., Bukusi, E. A., Onono, M., & Turan, J. M. (2013). Social context and drivers of intimate partner violence in rural Kenya: Implications for the health of pregnant women. Cult Health Sex, 15(4). https://doi.org/10.1080/13691058.2012.760205.Social Heaney, C. A., & Israel, B. A. (2008). Social networks and social support. In K. Glanz, B. K. Rimer, & K. Viswanath (Eds.), Health behaviour and health education. Theory, research and practice (4th ed., p. 590). San Francisco, US: Jossey-Bass. Heise, L. (1998). Violence against women: An integrated, ecological framework. Violence Against Women, 4(3), 262–290. https://doi.org/10.1177/1077801298004003002 Heise, L., Ellsberg, M., & Gottemoeller, M. (1999). Ending violence against women (Vol. 27(4)). Population Reports. Heise, L., & Kotsadam, A. (2015). Cross-national and multilevel correlates of partner violence: An analysis of data from population-based surveys. The Lancet Global Health, 3(6), e332– e340. https://doi.org/10.1016/S2214-109X(15)00013-3 Hertrich, V. (2017). Trends in age at marriage and the onset of fertility transition in sub-Saharan Africa. Population and Development Review, 43, 112–137. https://doi.org/10.1111/padr.12043 Hindin, M. J. (2003). Understanding women’s attitudes towards wife beating in Zimbabwe. Bulletin of the World Health Organization, 81(7), 501–508. Hindin, M. J. (2014). Adolescent childbearing and women’s attitudes towards wife beating in 25 sub-Saharan African countries. Maternal and Child Health Journal, 18(6), 1488–1495. https://doi.org/10.1007/s10995-013-1389-4 Horne, C., Dodoo, F. N.-A., & Dodoo, N. D. (2013). The shadow of indebtedness: Bridewealth and norms constraining female reproductive autonomy. American Sociological Review, 78(3), 503–529. https://doi.org/10.1177/0003122413484923 Hunnicutt, G. (2009). Varieties of patriarchy and violence against women. Resurrecting “patriarchy” as a theoretical tool. Violence Against Women, 15(5), 553–573. https://doi.org/10.1177/1077801208331246 Hunter, M. (2016). Introduction: New insights on marriage and Africa. Africa Today, 62(3), vii. https://doi.org/10.2979/africatoday.62.3.vii Institute of Development Studies, (IDS)., Ghana Statistical Service, (GSS)., & Associates. (2016). Domestic violence in Ghana: Incidence, attitudes, determinants and consequences. Ids. Retrieved from http://opendocs.ids.ac.uk/opendocs/handle/123456789/12168 Isiugo-Abanihe, U. C. (1994a). Consequences of bridewealth changes on nuptiality patterns among the Ibo of Nigeria. In C. Bledsoe & G. Pison (Eds.), Nuptiality in sub-Saharan 219 University of Ghana http://ugspace.ug.edu.gh Africa. (pp. 74–91). Oxford: Clarendon Press. Retrieved from http://www.africabib.org/rec.php?RID=W00078760 Isiugo-Abanihe, U. C. (1994b). Reproductive motivation and family-size preferences among Nigerian men. Studies in Family Planning, 25(3), 149. https://doi.org/10.2307/2137941 Jasinski, J. L. (2001). Theoretical explanations for violence against women. In & R. K. B. C. M. Renzetti, J. L. Edleson (Ed.), Sourcebook on violence against women (pp. 5–22). Thousand Oaks, CA: Sage. Javaid, A. (2015). The role of alcohol in intimate partner violence: Causal behaviour or excusing behaviour? British Journal of Community Justice, 13(1), 75–92. Jejeebhoy, S. J. (1995). Women’s education, autonomy, and reproductive behaviour: experience from developing countries. OUP Catalogue. Retrieved from http://ideas.repec.org/b/oxp/obooks/9780198290339.html Jewkes, R. (2002). Intimate partner violence: Causes and prevention. Lancet, 359(9315), 1423– 1429. https://doi.org/10.1016/S0140-6736(02)08357-5 Jewkes, R., Fulu, E., Tabassam Naved, R., Chirwa, E. D., Dunkle, K., Haardörfer, R., & García- Moreno, C. (2017). Women’s and men’s reports of past-year prevalence of intimate partner violence and rape and women’s risk factors for intimate partner violence: A multicountry cross-sectional study in Asia and the Pacific. PLoS Medicine, 14(9), 1–20. https://doi.org/10.1371/journal.pmed.1002381 Jewkes, R., Levin, J., & Penn-Kekana, L. (2002). Risk factors for domestic violence: Findings from a South African cross-sectional study. Social Science & Medicine, 55, 1603–1617. Jewkes, R., & Morrell, R. (2010). Gender and sexuality: Emerging perspectives from the heterosexual epidemic in South Africa and implications for HIV risk and prevention. Journal of the International AIDS Society, 13(1), 1–11. https://doi.org/10.1186/1758-2652- 13-6 Johnson, K. B., & Das, M. B. (2009). Spousal violence in Bangladesh as reported by men: Prevalence and risk factors. Journal of Interpersonal Violence, 24(6), 977–995. https://doi.org/10.1177/0886260508319368 Johnson, M. P. (2005). Domestic violence: It’s not about gender — Or is it? Journal of Marriage and Family, 67(December), 1126–1130. Johnson, M. P. (2006). Conflict and control: Gender symmetry and asymmetry in domestic violence. Violence Against Women, 12(11), 1003–1018. https://doi.org/10.1177/1077801206293328 Kapiga, S., Harvey, S., Muhammad, A. K., Stöckl, H., Mshana, G., Hashim, R., … Watts, C. (2017). Prevalence of intimate partner violence and abuse and associated factors among women enrolled into a cluster randomised trial in northwestern Tanzania. BMC Public Health, 17(1), 1–11. https://doi.org/10.1186/s12889-017-4119-9 Kaufman, C. E., Wet, T. De, & Stadler, J. (2001). Adolescent pregnancy and parenthood in 220 University of Ghana http://ugspace.ug.edu.gh South Africa. Studies in Family Planning, 32(2), 147–160. Kaukinen, C. (2004). Status compatibility, physical violence, and emotional abuse in intimate relationships. Journal of Marriage and Family, 66, 452–471. Kaye, D. K., Mirembe, F., Ekstrom, A. M., Kyomuhendo, G. B., & Johnsson, A. (2005). Implication of bride price on domestic violence and reproductive health in Wakiso District, Uganda. African Health Sciences, 5(4). Kim, J., & Gray, K. A. (2008). Leave or stay? Battered women’s decision after intimate partner violence. Journal of Interpersonal Violence, 23(10), 1465–1482. https://doi.org/10.1177/0886260508314307 Kimuna, S., Tenkorang, E. Y., & Djamba, Y. K. (2018). Ethnicity and intimate partner violence in Kenya. Journal of Family Issues, 1–24. https://doi.org/10.1177/0192513X18766192 Kishor, S., & Bradley, S. E. K. (2012). Women’s and men’s experience of spousal violence in two African countries : Does gender matter? DHS analytical studies No. 27. Calverton, Maryland, USA. Kiss, L., Schraiber, L. B., Heise, L., Zimmerman, C., Gouveia, N., & Watts, C. (2012). Gender- based violence and socioeconomic inequalities: Does living in more deprived neighbourhoods increase women’s risk of intimate partner violence? Social Science and Medicine, 74(8), 1172–1179. https://doi.org/10.1016/j.socscimed.2011.11.033 Klomegah, R. Y. (2008). Intimate partner violence (IPV) in Zambia: An examination of risk factors and gender perceptions. Journal of Comparative Family Studies, 39(4), 557–569. Klugman, J. (2017). Governance and the Law: Gender based violence and the law. Background paper for World Development Report. Retrieved from http://documents.worldbank.org/curated/en/611461487586627507/pdf/112914-WP- PUBLIC-WDR17BPGenderbasedviolenceandthelaw.pdf Krug, E. G., Dahlberg, L. L., Mercy, J. A., Zwi, A. B., & Lozano, R. (2002). World report on violence and health. Geneva, World Health Organization. Kubiak, S. P., Brenner, H., Bybee, D., Campbell, R., & Fedock, G. (2018). Reporting sexual victimization during incarceration: Using ecological theory as a framework to inform and guide future research. Trauma, Violence, and Abuse, 19(1), 94–106. https://doi.org/10.1177/1524838016637078 Lanier, C., & Maume, M. O. (2009). Intimate partner violence and social isolation across the rural/urban divide. Violence Against Women, 15(11), 1311–1330. https://doi.org/10.1177/1077801209346711 Lawoko, S. (2008). Predictors of attitudes toward intimate partner violence: A comparative study of men in Zambia and Kenya. Journal of Interpersonal Violence, 23(8), 1056–1074. Lawrence, E., Orengo-Aguayo, R., Langer, A., & Brock, R. L. (2012). The impact and consequences of partner abuse on partners. Partner Abuse, 3(4), 406–428. https://doi.org/10.1891/1946-6560.3.4.406 221 University of Ghana http://ugspace.ug.edu.gh Lawson, J. (2012). Sociological theories of intimate partner violence. Journal of Human Behavior in the Social Environment, 22(5), 572–590. https://doi.org/10.1080/10911359.2011.598748 Lee, R. E., Medina, A. V., Mama, S. K., Reese-Smith, J. Y., O’Connor, D. P., Brosnan, M., … Estabrooks, P. A. (2011). Health is power: An ecological, theory-based health intervention for women of color. Contemporary Clinical Trials, 32(6), 916–923. https://doi.org/10.1016/j.cct.2011.07.008 Lesthaeghe, R. J. (1989). Reproduction and social organization in sub-Saharan Africa. Berkeley: University of California Press. Li, Y., Marshall, C. M., Rees, H. C., Nunez, A., Ezeanolue, E. E., & Ehiris, J. E. (2014). Intimate partner violence and HIV infection among women: A systematic review and meta-analysis. Journal of the International AIDS Society, 17. https://doi.org/10.7448/IAS.17.1.18845 Luke, N. (2003). Age and economic asymmetries in the sexual relationships of adolescent girls in sub-Saharan Africa. Studies in Family Planning, 34(2), 67–86. https://doi.org/10.1111/j.1728-4465.2003.00067.x Lund, T. (2012). Combining qualitative and quantitative approaches: Some arguments for mixed methods research. Scandinavian Journal of Educational Research, 56(2), 155–165. https://doi.org/10.1080/00313831.2011.568674 Machisa, M. T., Christofides, N., & Jewkes, R. (2018). Social support factors associated with psychological resilience among women survivors of intimate partner violence in Gauteng , South Africa. Global Health Action, 11(3). https://doi.org/10.1080/16549716.2018.1491114 Mann, J., & Takyi, B. K. (2009). Autonomy, dependence or culture: Examining the impact of resources and socio-cultural processes on attitudes towards intimate partner violence in Ghana, Africa. Journal of Family Violence, 24, 323–335. https://doi.org/10.1007/s10896- 009-9232-9 Max, W., Rice, D. P., Finkelstein, E., Bardwell, R. A., & Leadbetter, S. (2004). The economic toll of intimate partner violence against women in the United States. Violence and Victims, 19(3), 259–272. Maxwell, C. D., Garner, J. H., & Fagan, J. A. (2001). The effects of arrest on intimate partner violence: New evidence from the spouse assault replication program. National Institute of Justice (Vol. JR000241). Washington, DC. Mburia-Mwalili, A., Clements-Nolle, K., Lee, W., Shadley, M., & Yang, W. (2010). Intimate partner violence and depression in a population-based sample of women: Can social support help? Journal of Interpersonal Violence, 25(12), 2258–2278. https://doi.org/10.1177/0886260509354879 McCloskey, L. A., Williams, C., & Larsen, U. (2005). Gender inequality and intimate partner violence among women in Moshi, Tanzania. International Family Planning Perspectives, 31(03), 124–130. https://doi.org/10.1363/3112405 222 University of Ghana http://ugspace.ug.edu.gh McDonald, M. (1994). Gender, drink, and drugs. ((Ed), Ed.). Oxford, UK: Berg. Mcevoy, P., & Richards, D. (2006). A critical realist rationale for using a combination of quantitative and qualitative methods. Journal of Research in Nursing, 11(1), 66–78. https://doi.org/10.1177/1744987106060192 McEvoy, P., & Richards, D. (2007). Gatekeeping access to community mental health teams: A qualitative study. International Journal of Nursing Studies, 44, 387–395. https://doi.org/10.1016/j.ijnurstu.2006.05.012 Meyers, J. L., Brown, Q., Grant, B. F., & Hasin, D. (2017). Religiosity, race/ethnicity, and alcohol use behaviors in the United States. Psychological Medicine, 47(1), 103–114. https://doi.org/10.1017/S0033291716001975 Michau, L., Horn, J., Bank, A., Dutt, M., & Zimmerman, C. (2015). Prevention of violence against women and girls: Lessons from practice. The Lancet, 385(9978), 1672–1684. https://doi.org/10.1016/S0140-6736(14)61797-9 Mitchell, J., Wight, M., Van Heerden, A., & Rochat, T. J. (2016). Intimate partner violence, HIV, and mental health: A triple epidemic of global proportions. International Review of Psychiatry, 28(5), 452–463. https://doi.org/10.1080/09540261.2016.1217829 Mitchell, R. E., & Hodson, C. A. (1983). Coping with domestic violence: Social support and psychology health among battered women, 240. Mizinga, F. M. (2000). Marriage and bridewealth in a matrilineal society: The case of the Tonga of Southern Zambia: 1900-1996. African Economic History, 28, 53–87. Molm, L. D. (2003). Theoretical comparisons of forms of exchange. Sociological Theory, 21(1). Moore, B. C., Easton, C. J., & McMahon, T. J. (2011). Drug abuse and intimate partner violence: A comparative study of opioid-dependent fathers. American Orthopsychiatric Association, 81(2), 218–227. https://doi.org/10.1111/j.1939-0025.2011.01091.x Morojele, N. K., Kachieng’a, M. A., Mokoko, E., Nkoko, M. A., Parry, C. D. H., Nkowane, A. M., … Saxena, S. (2006). Alcohol use and sexual behaviour among risky drinkers and bar and shebeen patrons in Gauteng province, South Africa. Social Science and Medicine, 62(1), 217–227. https://doi.org/10.1016/j.socscimed.2005.05.031 Mozhdeh, N. L. A., Ghazinour, M., Nojomi, M., & Richter, J. (2012). The buffering effect of social support between domestic violence and self-esteem in pregnant women in Tehran, Iran. Journal of Family Violence, 27(3), 225–231. https://doi.org/10.1007/s10896-012- 9420-x Mulawa, M., Kajula, L. J., Yamanis, T. J., Balvanz, P., Kilonzo, M. N., & Maman, S. (2018). Perpetration and victimization of intimate partner violence among young men and women in Dar es Salaam, Tanzania. Journal of Interpersonal Violence, 33(16), 2486 –2511. https://doi.org/10.1177/0886260515625910 Mulawa, M., Reyes, H. L. M. N., Foshee, V. A., Halpern, C. T., Martin, S. L., Kajula, L. J., & Maman, S. (2018). Associations between peer network gender norms and the perpetration 223 University of Ghana http://ugspace.ug.edu.gh of intimate partner violence among urban Tanzanian men: A multilevel analysis. Prevention Science, 19(4), 427–436. https://doi.org/10.1007/s11121-017-0835-8 Murphy, C. M., & O’leary, K. D. (1989). Psychological aggression predicts physical aggression in early marriage. Journal of Consulting and Clinical Psychology, 57(5). Namy, S., Carlson, C., O’Hara, K., Nakuti, J., Bukuluki, P., Lwanyaaga, J., … Michau, L. (2017). Towards a feminist understanding of intersecting violence against women and children in the family. Social Science and Medicine, 184, 40–48. https://doi.org/10.1016/j.socscimed.2017.04.042 Naved, R. T., & Akhtar, N. (2008). Spousal violence against women and suicidal ideation in Bangladesh. Women’s Health Issues, 18(6), 442–452. https://doi.org/10.1016/j.whi.2008.07.003 Neil, A. O., & Scovelle, A. J. (2018). Intimate partner violence perpetration and cardiovascular risk: A systematic review. Preventive Medicine Reports, 10(September 2017), 15–19. https://doi.org/10.1016/j.pmedr.2018.01.006 Ngubane, H. (1987). The consequences for women of marriage payments in a society with partrlineal descent. In D. Parkin & D. Nyamwaya (Eds.), Transformations of African Marriage (pp. 173–182). Manchester, UK: Manchester University Press. Nukunya, G. K. (1969). Kinship and Mariage among the Anlo Ewe. London: Athlone Press. Nukunya, G. K. (1992). Tradition and Change: The case of the family. Accra, Ghana: Ghana Universities Press. Nukunya, G. K. (2003). Tradition and change in Ghana: An introduction to sociology. Accra, Ghana: Ghana Universities Press. Nwabunike, C., & Tenkorang, E. Y. (2017). Domestic and marital violence among three ethnic groups in Nigeria. Journal of Interpersonal Violence, 32(18), 2751–2776. https://doi.org/10.1177/0886260515596147 Odimegwu, C. O. (2001). Couple formation and domestic violence among the Tiv of Benue State, Nigeria. In Paper presented at the International Colloquium Gender, Population in Africa organized by UAPS, INED, ENSEA, IFORD, Abidjan, Ivory Coast. Abidjan, Ivory Coast. Oduro, A. D., Deere, C. D., & Catanzarite, Z. B. (2015). Women’s wealth and intimate partner violence: Insights from Ecuador and Ghana. Feminist Economics, 21(2), 1–29. https://doi.org/10.1080/13545701.2014.997774 Ofei-Aboagye, R. O. (1994a). Altering the strands of the fabric: A preliminary look at domestic violence in Ghana. Journal of Women in Culture and Society, 19(41), 924–938. Ofei-Aboagye, R. O. (1994b). Domestic violence in Ghana: An initial step. Gender & Law, 1(25). Oheneba-Sakyi, Y., & Takyi, B. K. (2006). African families at the turn of the 21st Century. 224 University of Ghana http://ugspace.ug.edu.gh Greenwood Publishing Group. Okenwa-Emegwa, L., Lawoko, S., & Jansson, B. (2016). Attitudes toward physical intimate partner violence against women in Nigeria. SAGE Open, 6(4), 1–10. https://doi.org/10.1177/2158244016667993 Ondicho, T. (2000). Battered women: A socio-legal perspective of their experiences in Nairobi. African Study Monographs, 21(1), 35–44. Owusu-Addo, E., Owusu-Addo, S. B., Antoh, E. F., Sarpong, Y. A., Obeng-Okrah, K., & Annan, G. K. (2018). Ghanaian media coverage of violence against women and girls: Implications for health promotion. Women’s Health, 18(129), 1–11. Oyediran, K. A., & Isiugo-Abanihe, U. C. (2005). Perceptions of Nigerian women on domestic violence: Evidence from 2003 Nigeria Demographic and Health Survey. African Journal of Reproductive Health, 9(2), 38–53. Ozaki, R., & Otis, M. D. (2017). Gender equality, patriarchal cultural norms, and perpetration of intimate partner violence: Comparison of male university students in Asian and European cultural contexts. Violence Against Women, 23(7), 1076 –1099. https://doi.org/10.1177/1077801216654575 Panda, P., & Agarwal, B. (2005). Marital violence, human development and women’s property status in India. World Development, 33(5), 823–850. https://doi.org/10.1016/j.worlddev.2005.01.009 Peralta, R. L., Tuttle, L. A., & Steele, J. L. (2010). At the intersection of interpersonal violence, masculinity, and alcohol use: The experiences of heterosexual male perpetrators of intimate partner violence. Violence Against Women, 16(4), 387–409. https://doi.org/10.1177/1077801210363539 Peterson, C., Kearns, M. C., McIntosh, W. L., Estefan, L. F., Nicolaidis, C., McCollister, K. E., … Florence, C. (2018). Lifetime economic burden of intimate partner violence among U.S. adults. American Journal of Preventive Medicine, 55(4), 433–444. https://doi.org/10.1016/j.amepre.2018.04.049 Phorano, O., Nthomang, K., & Ntseane, D. (2005). Alcohol abuse, gender-based violence and HIV/AIDS in Botswana: Establishing the link based on empirical evidence. Journal of Social Aspects of HIV/AIDS, 2(1), 188–202. https://doi.org/10.1080/17290376.2005.9724842 Pickens, J. (2005). Organizational behavior in health care. Sudbury: Jones and Bartlett Publishers. Pico-Alfonso, M. A. (2005). Psychological intimate partner violence: The major predictor of posttraumatic stress disorder in abused women. Neuroscience and Biobehavioral Reviews, 29, 181–193. https://doi.org/10.1016/j.neubiorev.2004.08.010 Ragavan, M., Iyengar, K., & Wurtz, R. (2015). Perceptions of options available for victims of physical intimate partner violence in Northern India. Violence Against Women, 21(5), 652– 225 University of Ghana http://ugspace.ug.edu.gh 675. https://doi.org/10.1177/1077801215573332 Rani, M., & Bonu, S. (2009). Attitudes toward wife-beating: A cross-country study in Asia. Journal of Interpersonal Violence, 24(8), 1371–1397. Rani, M., Bonu, S., & Diop-Sidibe, N. (2004). An empirical investigation of attitudes towards wife-beating among men and women in seven sub-Saharan African countries. African Journal of Reproductive Health, 8(3), 116. https://doi.org/10.2307/3583398 Reed, J. S. (1972). The enduring south: Subcultural persistence in mass society. Lexington, MA: Lexington Books. Reed, P., & Wu, Y. (2013). Logistic regression for risk factor modelling in stuttering research. Journal of Fluency Disorders, 38(2), 88–101. https://doi.org/10.1016/j.jfludis.2012.09.003 Rees, S., Mohsin, M., Tay, A. K., Soares, E., Tam, N., da Costa, Z., … Silove, D. (2017). Associations between bride price stress and intimate partner violence amongst pregnant women in Timor-Leste. Globalization and Health, 13(1), 66. https://doi.org/10.1186/s12992-017-0291-z Reid, R. J., Bonomi, A. E., Rivara, F. P., Anderson, M. L., Fishman, P. A., Carrell, D. S., & Thompson, R. S. (2008). Intimate partner violence among men. Prevalence, chronicity, and health effects. American Journal of Preventive Medicine, 34(6), 478–485. https://doi.org/10.1016/j.amepre.2008.01.029 Reniers, G., & Tfaily, R. (2008). Polygyny and HIV in Malawi. Demographic Research, 19(53), 1811–1830. https://doi.org/10.4054/DemRes.2008.19.53 Renzetti, C. M., Dewall, C. N., Messer, A., & Pond, R. (2017). By the grace of God: Religiosity, religious perpetration of intimate partner violence. Journal of Family Issues, 38(14), 1974 – 1997. https://doi.org/10.1177/0192513X15576964 Rhodes, K. V., Houry, D., Cerulli, C., Straus, H., Kaslow, N. J., & McNutt, L.-A. (2009). Intimate partner violence and comorbid mental health conditions among urban male patients. Annals of Family Medicine, 7(1), 47–55. https://doi.org/10.1370/afm.936.Department Rigby, S. W., & Johnson, L. F. (2017). The relationship between intimate partner violence and HIV: A model-based evaluation. Infectious Disease Modelling, 2(1), 71–89. https://doi.org/10.1016/j.idm.2017.02.002 Rimonte, N. (1991). A question of culture: Cultural approval of violence against women in the Pacific-Asian community and the cultural defense. Stanford Law Review, 43(6), 1311–1326. Rudwick, S., & Posel, D. (2015). Zulu bridewealth ( ilobolo ) and womanhood in South Africa. Social Dynamics, 3952, 1–18. https://doi.org/10.1080/02533952.2015.1060683 Ryan, R. M., Sheldon, K. M., Kasser, T., & Deci, E. L. (1996). All goals are not created equal: An organismic perspective on the nature of goals and their regulation. Psychology of Action: Linking Cognition and Motivation to Behavior, (January), 7–25. https://doi.org/papers://7EAF3440-77F6-4B89-B12D-846E31E57E4A/Paper/p3085 226 University of Ghana http://ugspace.ug.edu.gh Sampson, R. J., Raudenbush, S. W., & Earls, F. (1997). Neighborhoods and violent crime: A multilevel study of collective efficacy. Science, 277, 918–924. Sanders, C. K. (2015). Economic abuse in the lives of women abused by an intimate partner: A qualitative study. Violence Against Women, 21(1), 3–29. https://doi.org/10.1177/1077801214564167 Santana, M. C., Raj, A., Decker, M. R., La Marche, A., & Silverman, J. G. (2006). Masculine gender roles associated with increased sexual risk and intimate partner violence perpetration among young adult men. Journal of Urban Health, 83(4), 575–585. https://doi.org/10.1007/s11524-006-9061-6 Schneider, H. K. . (1964). A model of African indigenous economy and society. Comparative Studies in Society and History, 7(1), 37–55. Sedziafa, A. P., Tenkorang, E. Y., & Owusu, A. Y. (2016). Kinship and intimate partner violence against married women in Ghana: A qualitative exploration. Journal of Interpersonal Violence, 1–28. https://doi.org/10.1177/0886260515624213 Sedziafa, A. P., Tenkorang, E. Y., Owusu, A. Y., & Sano, Y. (2017). Women’s experiences of intimate partner economic abuse in the Eastern Region of Ghana. Journal of Family Issues, 38(18), 2620–2641. https://doi.org/10.1177/0192513X16686137 Semahegn, A., & Mengistie, B. (2015). Domestic violence against women and associated factors in Ethiopia: Systematic review. Reproductive Health, 12(1). https://doi.org/10.1186/s12978- 015-0072-1 Speizer, I. S. (2010). Intimate partner violence attitudes and experience among women and men in Uganda. Journal of Interpersonal Violence, 25(7), 1224–1241. https://doi.org/10.1177/0886260509340550 Spencer, C. M., Morgan, P., Bridges, J., Washburn-Busk, M., & Stith, S. M. (2017). The relationship between approval of violence and intimate partner violence in college students. Journal of Interpersonal Violence, 1–20. https://doi.org/10.1177/0886260517731315 Stöckl, H., Devries, K., Rotstein, A., Abrahams, N., Campbell, J. C., Watts, C., & García- Moreno, C. (2013). The global prevalence of intimate partner homicide: A systematic review. The Lancet, 382(9895), 859–865. https://doi.org/10.1016/S0140-6736(13)61030-2 Straus, M. A. (1979). Measuring intrafamily conflict and violence: The Conflict Tactics (CT) Scales. Journal of Marriage and the Family, 41(1), 75. https://doi.org/10.2307/351733 Straus, M. A. (2008). Dominance and symmetry in partner violence by male and female university students in 32 nations. Children and Youth Services Review, 30(3), 252–275. https://doi.org/10.1016/j.childyouth.2007.10.004 Straus, M. A., Hamby, S. L., Boney-McCoy, S., & Sugarman, D. B. (1996). The revised Conflict Tactics Scales (CTS2) development and preliminary psychometric data. Journal of Family Issues, 17(3), 283–316. Tait, D. (1961). The Konkomba of Northern Ghana. (Jack Goody, Ed.). London: Oxford 227 University of Ghana http://ugspace.ug.edu.gh University Press. Takyi, B. K. (2001). Marital instability in an African society: Exploring the factors that influence divorce processes in Ghana. Sociological Focus, 34(1), 77–96. https://doi.org/10.1080/00380237.2001.10571184 Takyi, B. K., & Dodoo, F. N.-A. (2005). Gender, lineage, and fertility-related outcomes in Ghana. Journal of Marriage and Family, 67, 251–257. Takyi, B. K., & Gyimah, O. S. (2007). Matrilineal family ties and marital dissolution in Ghana. Journal of Family History, 28(5), 682–705. Takyi, B. K., & Lamptey, E. (2016). Faith and marital violence in sub-Saharan Africa: Exploring the links between religious affiliation and intimate partner violence among women in Ghana. Journal of Interpersonal Violence, 1(28). https://doi.org/10.1177/0886260516676476 Takyi, B. K., & Mann, J. (2006). Intimate partner violence in Ghana, Africa: The perspectives of men regarding wife beating. International Journal of Sociology of the Family, 32(1), 61–78. Tayyab, F., Kamal, N., Akbar, T., & Zakar, R. (2017). Men and women’s perceptions of justifications of wife beating: Evidence from Pakistan Demographic and Health Survey 2012–13. Journal of Family Violence, 32(7), 721–730. https://doi.org/10.1007/s10896-017- 9910-y Tenkorang, E. Y., & Owusu, A. Y. (2018). A life course understanding of domestic and intimate partner violence in Ghana. Child Abuse and Neglect, 79(January), 384–394. https://doi.org/10.1016/j.chiabu.2018.02.027 Tenkorang, E. Y., Owusu, A. Y., Yeboah, E. H., & Bannerman, R. (2013). Factors influencing domestic and marital violence against women in Ghana. Journal of Family Violence, 28(8), 771–781. https://doi.org/10.1007/s10896-013-9543-8 Thornton, A., & Camburn, D. (1987). The influence of the family on premarital sexual attitudes and behavior. Demography, 24(3), 323–340. Tjaden, P., & Thoennes, N. (2000a). Extent, nature, and consequences of intimate partner violence: Findings from the National Violence Against Women Survey. National Institute of Justice & Centers for Disease Control and Prevention. https://doi.org/10.1037/e300342003- 001 Tjaden, P., & Thoennes, N. (2000b). Prevalence and consequences of male-to-female and female-to-male intimate partner violence as measured by national violence against women survey. Violence Against Women, 6(2), 142–161. Tolman, R. M., & Raphael, J. (2000). A review of research on welfare and domestic violence. Journal of Social Issues, 56(4), 655–682. https://doi.org/10.1111/0022-4537.00190 Umubyeyi, A., Mogren, I., Ntaganira, J., & Krantz, G. (2014). Women are considerably more exposed to intimate partner violence than men in Rwanda: Results from a population-based, cross-sectional study. BMC Womens Health, 14(99). https://doi.org/10.1186/1472-6874-14- 228 University of Ghana http://ugspace.ug.edu.gh 99 Usta, J., Makarem, N. N., & Habib, R. R. (2013). Economic abuse in Lebanon: Experiences and perceptions. Violence Against Women, 19(3), 356–357. https://doi.org/10.1177/1077801213486313 Uthman, O. A., Lawoko, S., & Moradi, T. (2009). Factors associated with attitudes towards intimate partner violence against women: A comparative analysis of 17 sub-Saharan countries. BMC International Health and Human Rights, 9(1). https://doi.org/10.1186/1472- 698X-9-14 Uthman, O. A., Lawoko, S., & TaherehMorad. (2010). Sex disparities in attitudes towards intimate partner against women in sub-Saharan Africa: A socio-ecological analysis. BMC Public Health, 10(1), 223. https://doi.org/10.1186/1471-2458-10-223 Vatnar, S. K. B., Friestad, C., & Bjørkly, S. (2018). Differences in intimate partner homicides perpetrated by men and women: Evidence from a Norwegian national 22-year cohort. Psychology, Crime & Law, 0(0), 1–30. https://doi.org/10.1080/1068316X.2018.1438433 Verduin, F., Engelhard, E. A. N., Rutayisire, T., Stronks, K., & Scholte, W. F. (2013). Intimate partner violence in Rwanda: The mental health of victims and perpetrators. Journal of Interpersonal Violence, 28(9), 1839–1858. https://doi.org/10.1177/0886260512469106 Wagman, J. A., Donta, B., Ritter, J., Naik, D. D., Nair, S., Saggurti, N., … Silverman, J. G. (2016). Husband’s alcohol use, intimate partner violence, and family maltreatment of low- income postpartum women in Mumbai, India. Journal of Interpersonal Violence, 1–27. https://doi.org/10.1177/0886260515624235 Walby, S., & Allen, J. (2004). Domestic violence, sexual assault and stalking: Findings from the British crime survey. https://doi.org/10.1037/e634612007-001 Walklate, S. L. (2013). Gender, crime and criminal justice. Routledge. Wang, L. (2016). Factors influencing attitude toward intimate partner violence. Aggression and Violent Behavior, 29, 72–78. https://doi.org/10.1016/j.avb.2016.06.005 Weitzman, A. (2014). Women’s and men’s relative status and intimate partner violence in India. Population and Development Review, 40(1), 55–75. Wendo, C. (2004). African women denounce bride price. Lancet, 363(9410), 716. Were, E., Curran, K., Delany-Moretlwe, S., Nakku-Joloba, E., Mugo, N. R., Kiarie, J., … Baeten, J. M. (2013). A prospective study of frequency and correlates of intimate partner violence among African heterosexual HIV serodiscordant couples. AIDS, 25(16), 2009– 2018. https://doi.org/10.1097/QAD.0b013e32834b005d.A Whitaker, M. P. (2013). Centrality of control-seeking in men’s intimate partner violence perpetration. Prevention Science, 14(5), 513–523. https://doi.org/10.1007/s11121-012- 0332-z White, M. E., & Satyen, L. (2015). Cross-cultural differences in intimate partner violence and 229 University of Ghana http://ugspace.ug.edu.gh depression: A systematic review. Aggression and Violent Behavior, 24, 120–130. https://doi.org/10.1016/j.avb.2015.05.005 Wilson, I. M., Graham, K., & Taft, A. (2017). Living the cycle of drinking and violence: A qualitative study of women’s experience of alcohol-related intimate partner violence. Drug and Alcohol Review, 36, 115–124. https://doi.org/10.1111/dar.12405 Wojcicki, J. M., Straten, A. van der, & Padian, N. (2010). Bridewealth and sexual and reproductive practices among women in Harare, Zimbabwe. AIDS Care, 22(6), 705–710. https://doi.org/10.1080/09540120903349094.Bridewealth Wood, S. N., Glass, N., & Decker, M. R. (2019). An integrative review of safety strategies for women experiencing intimate partner violence in low-and-middle-income countries. Trauma, Violence, and Abuse, 1–15. https://doi.org/10.1177/1524838018823270 World Health Organization, (WHO). (2013). Global and regional estimates of violence against women: prevalence and health effects of intimate partner violence and non-partner sexual violence. 2013, 57. https://doi.org/10.1007/s13398-014-0173-7.2 World Health Organization, (WHO). (2017). Fact sheet violence against women. Retrieved August 15, 2018, from http://www.who.int/news-room/fact-sheets/detail/violence-against- women World Health Organization, (WHO)., London School of Hygiene and Tropical Medicine, & World Health Organization. (2013). Global and regional estimates of violence against women: prevalence and health effects of intimate partner violence and non-partner sexual violence. 2013. Geneva. https://doi.org/10.1007/s13398-014-0173-7.2 Yarbrough, M. W. (2017). Very long engagements: The persistent authority of bridewealth in a post-apartheid South African community. Law and Social Inquiry, 1–31. Yount, K. M., Pham, H. T., Minh, T. H., Krause, K. H., Schuler, S. R., Anh, H. T., … Kramer, M. R. (2014). Violence in childhood, attitudes about partner violence, and partner violence perpetration among men in Vietnam. Annals of Epidemiology, 24(5), 333–339. https://doi.org/10.1016/j.annepidem.2014.02.004 Zablotska, I. B., Gray, R. H., Koenig, M. A., Serwadda, D., Nalugoda, F., Kigozi, G., … Wawer, M. (2009). Alcohol use, intimate partner violence, sexual coercion and HIV among women aged 15–24 in Rakai, Uganda. AIDS and Behavior, 13(2), 225–233. Zapor, H., Wolford-Clevenger, C., & Johnson, D. M. (2018). The association between social support and stages of change in survivors of intimate partner violence. Journal of Interpersonal Violence, 33(7), 1051–1070. https://doi.org/10.1177/0886260515614282 Zembe, Y. Z., Townsend, L., Thorson, A., Silberschmidt, M., & Ekstrom, A. M. (2015). Intimate partner violence, relationship power inequity and the role of sexual and social risk factors in the production of violence among young women who have multiple sexual partners in a peri-urban setting in South Africa. PLoS ONE, 10(11), 1–18. https://doi.org/10.1371/journal.pone.0139430 230 University of Ghana http://ugspace.ug.edu.gh APPENDICES Appendix A: Bridewealth Payment and Normative Constraints on Women’s Lives in Ghana Study: Sample Vignette (Questionnaire) Name of village: ________________ Time started: ______ Interviewer Name and Code: _______________________________________ [ ] Version 16 Vignette 1 Part A I am going to tell you a story about a man and a woman. I am going to start by telling you the beginning of the story. I will tell it to you twice. Please listen carefully, because after I tell you the beginning of the story I am going to ask you some questions about it. The man and the woman have been living together for three years. The man has paid some of the bridewealth. They have one child. The man works in a governmental institution. The woman has a big store selling cloth in the market that she started with her own money. One day the man found out that the woman had been giving away most of her earnings from her shop to an old female friend in the next village without telling him. Questions Vignette 1 Part A I want you to think about how different people in the community will react to the woman. I’m going to ask you about what the man, the man’s family, other men in the community, the woman’s family, and other women in the community will say about the woman in the story. 1.1. Think about the man, who is an average traditional man. How right or wrong will he say the woman was? To answer this question we would like you to use these beans. There are 10 beans in all. If the man thinks the woman was very right, he will choose 10 beans. If the man thinks the woman was very wrong, he will choose 1 bean. If the man thinks the woman was somewhere in between, he will choose some number of beans in between 1 and 10. The more right the man thinks the 231 University of Ghana http://ugspace.ug.edu.gh woman was, the more beans he will choose. The more wrong the man thinks the woman was, the fewer beans he will choose. How many beans will the man choose? Number of beans: [ ] 1.2. Think about the man’s family which is an average traditional family. How right or wrong will they say the woman was? How many beans will the man’s family choose? Number of beans: [ ] 1.3. Think about the men in the community. How right or wrong will they say the woman was? How many beans will the men in the community choose? Number of beans: [ ] 1.4. Think about the woman’s family which is an average traditional family. How right or wrong will they say the woman was? How many beans will the woman’s family choose? Number of beans: [ ] 1.5. Think about the women in the community. How right or wrong will they say the woman was? How many beans will the women in the community choose? Number of beans: [ ] 1.6. How about you? How right or wrong will you say the woman was? How many beans will you choose? Number of beans: [ ] Other Questions Now I’m going to ask you a few other questions about the stories. 232 University of Ghana http://ugspace.ug.edu.gh 4.1. In the first story, what did the woman in the story do? Did she use contraception or give away her earnings? Contraception [ ] Earnings [ ] 4.2. In the second story, did the man pay chop money or not? Yes [ ] No [ ] 4.3. In the third story, was the man HIV positive or negative? Positive [ ] Negative [ ] 4.4. Since they got together, has the man had sex with any other women? Yes [ ] No [ ] 4.5. In all the stories, did the man pay none, some, or all of the bridewealth? None [ ] Some [ ] All [ ] Demographic Questions Now I’m going to ask you some questions about yourself. 5.1. How old were you on your last birthday? Years [ ] 5.2. What is your highest level of education? 1. No education [ ] 2. Primary [ ] 3. Middle/JHS [ ] 4. Secondary/SHS [ ] 5. Higher [ ] 5.3. What is your religion? 1. Catholic [ ] 2. Anglican [ ] 3. Methodist [ ] 4. Presbyterian [ ] 5. Pentecostal/Charismatic [ ] 233 University of Ghana http://ugspace.ug.edu.gh 6. Other Christian [ ] 7. Moslem [ ] 8. Traditional/Spiritualist [ ] 9. No religion [ ] 5.4. What is your occupation? 5.5. How many children have you ever had? Number [ ] 5.6. How many of your own children are currently living? Number [ ] 5.7. What is your current marital status? 1. Never married [ ] END INTERVIEW 2. Married [ ] 3. Living together [ ] 4. Divorced [ ] END INTERVIEW 5. Separated [ ] END INTERVIEW 6. Widowed [ ] END INTERVIEW 5.8. How long have you been in the relationship? Months [ ] Years [ ] 5.9. For women: Does your husband/partner have other wives or does he live with other women as if they are married? For men: Do you have other wives or do you live with other women as if you are married? 1= Yes [ ] 2= No [ ] 8= Don’t Know [ ] 5.10. For women: If yes, how many partners does he have including you? For men: If yes, how many partners do you have? Number [ ] 5.11a. Was bridewealth negotiated in your current union? (If a man has more than one wife, ask for each wife) (If no to all of these, skip to Question 10) Spouse 1 1= Yes [ ] 2= No [ ] Spouse 2 1= Yes [ ] 2= No [ ] Spouse 3 1= Yes [ ] 2= No [ ] 5.11b. What is the status of bridewealth in your current union? 234 University of Ghana http://ugspace.ug.edu.gh (If a man has more than one wife, ask for each wife) Spouse 1 Spouse 2 Spouse 1. Paid in full [ ] Paid in full [ ] Paid in full [ ] 2. Partly paid [ ] Partly paid [ ] Partly paid [ ] 3. Not paid at all [ ] Not paid at all [ ] Not paid at all [ ] 4. Other [ ] Other [ ] Other [ ] 5.12. Did you ever observe violence between your parents/guardians? 1= Yes [ ] 2= No [ ] 5.13. What is your spouse’s/partner’s occupation? 5.14. What is your husband’s/wife’s/partner’s highest level of education. 1. No education [ ] 2. Primary [ ] 3. Middle/JHS [ ] 4. Secondary/SHS [ ] 5. Higher [ ] 5.15. Is your husband/wife… 1. Much older than you? [ ] 2. A little older than you? [ ] 3. About the same age as you? [ ] 4. A little younger than you? [ ] 5. Much younger than you? [ ] 5.16. For women: Have you ever been beaten by your partner/husband? For men: Have you ever beaten your wife/partner? 1= Yes [ ] 2= No [ ] Time ended: ___ INTERVIEWER COMMENTS ……………………………………………………………………………………………………… ……………………………………………………………………………………………………… 235 University of Ghana http://ugspace.ug.edu.gh Appendix B: List of communities and the respective sample sizes Name of Village Number Percentage 1. Anum 82 14.16 2. Boso 43 7.43 3. Tosen 23 3.97 4. Anyasu 19 3.28 5. Mpakadan 20 3.45 6. Nnudu 19 3.28 7. Aboasa 28 4.84 8. Apeguso 18 3.11 9. Akwamufie 52 8.98 10. Old Senchi 1 0.17 11. Mampong 95 16.41 12. Apirede 40 6.91 13. Awukugua 34 5.87 14. Dawu 39 6.74 15. Abiriw 24 4.15 16. Obosomase 1 0.17 17. Mamfe 41 7.08 Total 579 100 Source: 2016 Bridewealth Study (RIPS). No male interviews were conducted in Akropong 236 University of Ghana http://ugspace.ug.edu.gh Appendix C: Selected characteristics of participants for the key informant interviews Respondents Locality Ethnicity Lineage Gender R1 Abiriw Guan Patrilineal Male R2 Akwamufie Akan Matrilineal Male R3 Anum Guan Matrilineal Male R4 Anyasu Akan Matrilineal Male R5 Anyasu Akan Matrilineal Male R6 Mampong Akan Patrilineal Male R7 Mampong Akan Patrilineal Male R8 Tosen Guan Matrilineal Male R9 Abiriw Guan Patrilineal Female R10 Anum Guan Matrilineal Female R11 Anum Guan Matrilineal Female R12 Anyasu Akan Matrilineal Female R13 Anyasu Akan Matrilineal Female R14 Mampong Akan Patrilineal Female R15 Akwamufie Akan Matrilineal Female R16 Tosen Guan Matrilineal Female Source: 2018 Bridewealth Study (RIPS) 237 University of Ghana http://ugspace.ug.edu.gh Appendix D: Interview guide for qualitative data collection We have had a series of discussions with men and women in your community about marriage and bridewealth associated rights, and we would like to clarify a few things with you. 1.Please tell me about the significance of bridewealth payment in this community. a. What rights does payment give to a man or woman? b. What are the responsibilities associated with payment? c. What about the sanctions for not fulfilling your bridewealth related obligations? 2.In this community, and according to the traditions, is partially paid bridewealth acceptable and recognized. What are the rights of a man or woman if bridewealth is partially paid? How different are these from full payment; and bridewealth being negotiated, but none paid yet? 3. What has changed in recent times with bridewealth and what led to the changes if any? Probe for: a. Changes over time – with the ITEMS exchanged; the PRACTICES/CEREMONIES; the AMOUNT/COMPLETENESS before living together and childbearing. b. Women contributing towards bridewealth. Prevalence? Are there any implications? 4. Who enforces the norms about bridewealth payment – or whose opinions matter the most? Eg gender issues – do men and/or women enforce? Families of husband and/or wife? Reasons for differences in perceptions of norms among male and female respondents in our surveys? 5 a. When a man from your community wants another child but his wife does not want to have any more, according to your traditions, whose choice should dominate? Who would various people support and why? Probe: Would bridewealth payment matter in who gets support? 5 b. If a woman wants to have more children but her husband wants to limit or stop, who would different people in the community support and why? Probe: Would bridewealth payment matter in who gets support? 6. Are there any differences between matrilineal and patrilineal groups when it comes to the rights it bestows on couples? What about the sanctions or punishments for not fulfilling your bridewealth related obligations? Probe for: a. Couples rights and responsibilities b. Child ownership c. Control over women: contraceptive usage, domestic violence (all the above with respective to different levels of bridewealth payment) 238 University of Ghana http://ugspace.ug.edu.gh Appendix E: Setting survey weight (STATA command) for 2014 Ghana Demographic an Health Survey, Men’s Sample *gen weight = mv005/1000000* *gen strata = mv023* *gen psu = mv021* *svyset [pweight=weight], psu(mv021) strata(mv022) vce(linearized) singleunit(missing)* Appendix F: Distribution of men by marital status and completeness of bridewealth payment Completeness of bridewealth payment Marital status Fully paid Partially paid Not paid Total Married 324 122 13 459 Living together 0 18 42 60 Total 324 140 55 519 Source: 2016 Bridewealth Study (RIPS) 239 University of Ghana http://ugspace.ug.edu.gh Appendix G: Percentage distribution of men by number of items they approve of wife- beating on Un-weighted Sample Attitudes toward wife-beating Number Percent (%) Do not approve of wife-beating 1,720 87.44 Approve of wife-beating on at least one reason 90 4.58 Approve of wife-beating on at least two reasons 64 3.25 Approve of wife-beating on at least three reasons 46 2.34 Approve of wife-beating on at least four reasons 28 1.42 Approve of wife-beating on all five reasons 19 0.97 Total 1967 100 Weighted Sample Attitudes toward wife-beating Number Percent (%) Do not approve of wife-beating 1711 90.4 Approve of wife-beating on at least one reason 65 3.4 Approve of wife-beating on at least two reasons 44 2.3 Approve of wife-beating on at least three reasons 38 2.0 Approve of wife-beating on at least four reasons 21 1.1 Approve of wife-beating on all five reasons 14 0.7 Total 1893 100 240 University of Ghana http://ugspace.ug.edu.gh Appendix H: Organizing themes and basic themes for reasons for/benefits of bridewealth payment Global theme Organizing themes Basic themes Description Sample quotes Reasons Honours parents, Parents Bridewealth “Hmm! In order that it will honour your family. You for/benefits of woman and man payment honours see. Not that your parents have given birth to you, bridewealth parents but they won’t benefit from you. It honours your payment parents. This woman’s daughter respects herself. Do you understand?” (R11; elderly woman) Husband and Bridewealth “…the purpose is that when a man marries you the wife payment honours woman, he honours you. If all these things are not the couple done, then it means that the man is just treating you anyhow. He treats you anyhow he wants to but if he came for you from your parents, then there are disciplines he needs to ensure”. (R9; elderly woman) … oo marriage is a honourable thing, if a man has a wife it is greatly valued”. (R16; elderly woman) Inheritance Property Bridewealth “…If you marry the woman it helps the man it helps payment gives the woman because you are married and it gives the women and their right to have a stake in the property”. (R8; elderly children right to man) the man’s property Name “…Eeerrrr it is because of the inheritance that is why marriage rites are very important. The reason why it is important is that, without the marriage rites, he is not able to name the child and he went ahead to marry a different woman and they give birth and he perform the other woman’s marriage rites, the first woman’s child is a child born out of 241 University of Ghana http://ugspace.ug.edu.gh wedlock, the child of the one whose marriage rites was not performed is born out of wedlock. Okay, so if he the father has any property the child does not inherit any. Yes, that child is born out of wedlock”. (R6; elderly man) Old age social “…the children are his inheritance. The benefit of security him marrying you is the children”. (R11; elderly woman) Legitimize union Bridewealth “The reason why the bridewealth is paid is that payment maybe you don’t know the man at all and he also legitimizes union says he will marry your daughter. So you have to bring him into the presence of the father. It is a tradition that has been there for a long time. So when you do everything in the presence of the father then the marriage will be complete. That will be a witness that you’ve married the woman.” (R5; elderly man) “…what makes it relevant is the fact that women are a type of being that a lot of men have eyes for and if you see her and you want to live with her, you have to make the entire people know that she is yours. And for everyone to know she is yours then you have to go for her and perform the marriage rites. The marriage rites signifies that truly she is yours and other man who goes near her will be afraid because you have paid her bride price to the father and that makes marriage very relevant” (R8; elderly man). 242 University of Ghana http://ugspace.ug.edu.gh Protect women from Abuse Bridewealth “…You see it is a way by which they use to insulate and payment protects the men from mishandling them in our Akan system. from abuse That is why they say “do not abuse your wife”. You see. All these are done to insulate the woman that you can’t do her evil. Do you understand?”(R2; elderly man). “….the purpose is that when a man marries you the woman, he honours you. If all these things are not done, then it means that the man is just treating you anyhow. He treats you anyhow he wants to but if he came for you from your parents, then there are disciplines he needs to ensure” (R13; elderly woman). Unlawful Bridewealth “…And you cannot just bring the woman that you divorce payment protect don’t like her anymore. You have to see her off. And women from the way it happens you cannot even see her off. unlawful divorce Divorce is a little painful. Let’s say I have gone to marry Yaa and if I have seen this lady so I am going to give her to father, he will say “oh”, even the way to talk, can you come and stand there? (R2; elderly man). Women are valued Bridewealth is “….The reason why the marriage ceremony is paid to indicate significant is because you cannot give a woman out that women are as a gift to a man. I hope you have understood me. important The man must know that she is also someone’s precious being.” (R9; elderly woman) “…so that the man will know that the woman is someone’s royal so I must not hate her”. (R11; elderly woman) 243 University of Ghana http://ugspace.ug.edu.gh Economic Bridewealth “The father gives his daughter the 2 and 6 (26 support/security payment pesewas) to go and work with it. Go and use the 2 provides and 6 (26 pesewas) to work. So he has given you economic ‘dwetire’ [capital]”. (R1; elderly man) support to the wife. 244 University of Ghana http://ugspace.ug.edu.gh Appendix I: Organizing themes and basic themes for implications of non-payment of bridewealth Global theme Organizing themes Basic themes Description Sample quotes Implications of Breeds conflicts Non-payment of “…Yes. If you haven’t performed my marriage rites, non-payment of bridewealth could would I allow you to control me? When you talk I bridewealth breed conflicts would also talk. If you a capable man then you have between partners. to perform my marriage rites”. (R14; elderly woman) “..Then the relationship between you and him was not proper. It depends on both of you. There are women if you have not married them, they will keep disregarding you. There are some men who will say “back off”, the same”. (R15; elderly woman) Woman has been The wife could be “..The relevance is that if you marry a man and he “stolen” considered as has not performed your marriage rites then he is not stolen if your husband he is either your friend or he has bridewealth has stolen you. So your family can take him on but if there is understanding then he can come and not been paid. perform your marriage rites. So that it can be considered as a marriage”. (R16; elderly woman) Physical Female partner “… Yes he can beat her...but he already has kids abuse/beating may be beaten with her so this will also go to the elders who tell when bridewealth them not to repeat that because it has brought about has not been fight”. (R16; elderly woman) “….What about a man who has not performed the exchanged woman’s marriage rites… Yes. He can beat her. Why? Many. He does not care”. (R9; elderly woman) 245 University of Ghana http://ugspace.ug.edu.gh Children do not Cannot name Male partner “Currently as we talk one of my grandchildren has a belong to their the children cannot name his child with a teacher, the teacher is unable to come biological father children when and perform her marriage rites. He has come for the bridewealth has list but has not performed the rites. But they are forcing to name the child but that is not allowed not been paid. unless he has performed her marriage rites”. (R 8; elderly man) “yes if the man doesn’t marry you, you won’t give the children to the man. I: like how? R: they don’t belong to the man. I: but even if he doesn’t marry you the children are still not his. R: I told you that I have some of the children. Their husbands said they won’t marry them so when they gave birth we didn’t permit them to name the children. I: so the children will be for the woman but the name that will be given to the children will be given by the man. R: yes but since he hasn’t married the woman we won’t permit him. So he will call the children by foreign names. He has no authority to name the children. (R 13; elderly woman) No privileges The man does not “…since you have not married, her family can enjoy certain name them, and if you have given birth and the privileges from family has named them are the children yours? You the children when can have so many daughters and when they grow someone will also want to marry them, you the bridewealth has father will be there and you will realize that their not been paid uncle from their mother’s side will be taking the because the bridewealth. It is very painful……What is more children do not painful than this, it is very, very painful”. (R8; belong to him elderly man) Authority Man has no Man has no “…no he doesn’t have authority. He doesn’t have 246 University of Ghana http://ugspace.ug.edu.gh authority authority when much authority. It means that you have stolen the bridewealth has woman and she is with you.” (R5; elderly man) not been paid. “… If he has not performed the rites, then it is not marriage. None has power over the other. There is no law that binds them…” (R6; elderly man) Man has some Male partner has “..22I: Can they also deny their husbands of sex? R: authority some authority Why won’t she allow him? If she will not allow, then since the woman she must leave. If a man lives with you without lives with him paying your bridewealth, he still provides food for you. He is not taking care of an orphan? If the woman will deny her husband then she can as well leave”. (R1; elderly man) No sanctions Cheats No sanctions will “… You know if you live with a woman and you against the woman be meted to the have not married her and she cheats, you cannot ask woman if she for her to pay a fine. It can be that you take care of cheats her but if you do not go and perform the marriage rites and she goes out to have an affair, the fine cannot be taken. The father will say he does not know you because, you have not married her. So it is someone you have married that you can have that law with”. (R2; elderly man) 22 I: Interviewer. R: Respondent 247 University of Ghana http://ugspace.ug.edu.gh Insults No sanctions “…Eeehhee. This is because if I have not married against the you and you insult me that I am a fool and I report woman if she you to the elders, your family would defend you and insults the man say that even the man who says our daughter insulted him is not known to us I hope you understand. Where have you known him? And you say he is your husband, they would ask, when did he marry you? Who gave you the woman that you are reporting yourself that she insulted you that you are a fool? You have even sent yourself. They would take dedua (compensation/fine)” (R6; elderly man) Poor parental Parents are less “It can be that you take care of her but if you do not support likely to support go and perform the marriage rites and she goes out couple when to have an affair, the fine cannot be taken. The bridewealth has father will say he does not know you because, you have not married her. So it is someone you have not been paid married that you can have that law with”. (R2; elderly man) “You haven’t married her so I won’t mind you when you come and report her to me.” (R12; elderly woman) Non-payment does Man has not right “Even when you have gone to pay her bridewealth, not guarantee abuse to abuse his you do not have the right. How much more when you partner have not? Have heard this before?” (R1; elderly man) “None of them have power. Whether you have paid the bridewealth or not, you cannot beat her. The mean put their girlfriends in some corner and slap her because some other guy has called her. Ah, 248 University of Ghana http://ugspace.ug.edu.gh madness! I cannot. For my daughter, if you marry her and even add a car and you slap her, I will also slap you when I come. If you will not marry her anymore, bring her to me. No it shouldn’t happen.” (R2; elderly man) 249 University of Ghana http://ugspace.ug.edu.gh Appendix J: Organizing themes and basic themes for implications of partial bridewealth payment Global theme Organizing themes Basic themes Description Sample quotes Implications of Breeds conflicts Partial payment “You complete everything and then everyone will partial could be pointers see that she is your wife but if you say you are not bridewealth of conflicts entirely ready so you cannot perform the rites then payment she is not your wife so if she goes for another man they will say you are not married to her and you are behaving like that, then you the man will feel embarrassed. It means you are not married to her and you are behaving like that towards her.” (R16; elderly man) “~~~Traditionally, in the past, you cannot make partial payment. You will perform everything. Unless your in-law likes you and considers you to do that. Even in such cases the woman can use that as an insult anytime. The man can also use that as an insult. You see how the situation is? So you must force the man to perform your marriage rites.” (R9; elderly woman) Man has some Man has some “Once he has come to plead; you know, it is not authority authority since the always that it is all well for a person. So he has family has agreed come to do half of it and he has given you time to on partial payment come and do the rest or “tomorrow I’ll bring the rest.” and you’ve agreed so she is for him. Do you understand what I’m saying?. So whatever he wants is what he will let the woman do. So anytime you need her, you have to seek permission from her husband.” (R11; elderly woman) 250 University of Ghana http://ugspace.ug.edu.gh Woman is not valued Woman is not “Some men decide to make part payment. All these valued by her depend on the woman. For instance, as an educated partner young lady, if a man pays part of your bridewealth and you allow him to impregnate you, then he got you cheaply.” (R1; elderly man) “You have not finished paying so the woman can use that as a basis to divorce you. You see. You are taking her for granted.” (R9; elderly woman) Children do not Children do not “That is what I am saying that if the man does not belong to the man belong to the man do that, then the children belong to the woman.~~~” if he has only (R9; elderly woman) made partial payment Type of union Not The union may not “So over here most of the things are not static but considered as be considered as a the truth is that if you have not finished performing marriage proper one all the rites for the woman then she is not your wife.” (R2; elderly man) “No the tradition doesn’t allow it just like that. If you are going to marry and you don’t do everything, then you haven’t married her.” (R5; elderly man) Considered as When bridewealth “oh! But you came for her from the home. So it is marriage is paid partially, complete because you intentionally came to the the union can be house to plead so that is not like you’ve ignored that considered as you will not come and marry her. You’ve come home marriage to plead so you’ve done some and left the rest. You’ve done some but it wasn’t complete so you will bring it later. I: so will they say the woman is her 251 University of Ghana http://ugspace.ug.edu.gh wife? R: yes.” (R16; elderly woman) No sanctions if the The woman cannot “..but if you say you are not entirely ready so you woman cheats be sanctioned if cannot perform the rites then she is not your wife so she cheats if she goes for another man they will say you are not married to her and you are behaving like that, then you the man will feel embarrassed.” (R16; elderly woman) Poor parental Poor parental “If your wife cheats she will be given a fine (charge) support support in dealing it is her father who will support you to take the with marital “ayifari” (compensation) but since the rites was challenges incomplete he will not. Now where is the honour that is why I said if you are marrying you make sure you make the marriage rites complete”. (R8; elderly man) “..but if you say you are not entirely ready so you cannot perform the rites then she is not your wife so if she goes for another man they will say you are not married to her and you are behaving like that, then you the man will feel embarrassed.” (R16; elderly woman) 252 University of Ghana http://ugspace.ug.edu.gh Appendix K: Organizing themes and basic themes for implications of full bridewealth payment Global theme Organizing themes Basic themes Description Sample quotes Implications of Authority/position Full payment “Oh, you see, the power the man has in a marriage full bridewealth of responsibility places the man is they have handed over the woman in his care, to payment in a position of take care of her so that nothing happens to her. So responsibility that no one will hurt her, she won’t get sick or nothing will happen to her for you to live as the person you came in for or if you can or have some level of development. That is why I said initially that whatever she acquires, she should bring it home but when she is in debt, her husband pays for it. This is a proverb to the man who marries the woman. So you have to take care of the woman, her well-being is the man’s responsibility.” (R2; elderly man) “..the man having authority means that when he wakes up, he will provide the housekeeping money and everything.” (R12; elderly woman) “Yes, the power it gives is that when she is in your hand and something happens there is nothing to fear because you used the right channel to seek her hand and that she leaves with you. So when something happens and the man himself does not think of shooting or stabbing her with a knife, he now has power over you to do whatever that is right. Apart from actions that lead to death which is not part of the powers that he possess over her. But as for cook for me, fetch water for me to bath eerrh wash my stuffs for me are all part of the powers he now 253 University of Ghana http://ugspace.ug.edu.gh possess over you. He has to see to whatever you want to do and do it for you. Whatever you need, you have to ask him. Maybe I need a cloth, you have to ask him. Even if you want to visit your parent, you cannot just go like that you have to ask permission from him that maybe I would be visiting my parent next week or next two weeks or I have learnt my father is not well so I’m asking permission to go and see him. That is the power the man who performed your rites has over you” (R6; elderly man) Man has authority Bridewealth “Um, some of the things he does that we say gives over the woman payment gives him power over the woman is as follows; he has the man performed your marriage rites. You live in his authority over house. He is the one who gives you food to eat. He the woman takes care of you in everything. So he has power over you.” (R9; elderly woman) “Once he has married the woman, then he has power over her because he has performed the rights and brought her to live with him. So she is in his care. If there is any issue with her, he knows she is in his care. He also has the power to take care of her so that there would be no problem and they will be one. The man has power over the woman. If she is doing something which is not pleasing, he has every right to complain.” (R15; elderly woman) “And a woman’s head is her husband. And God is the head of your husband. It means that the authority has been given to you. You have been given permission to marry and the husband has also 254 University of Ghana http://ugspace.ug.edu.gh placed a ring on your finger. Whether it is right you are his wife and both of you are married and he has done everything for you.”(R5; elderly man) Does not guarantee Paying physical abuse bridewealth “Herh! Herh! I say don’t beat her at all. It’s not a does not mean matter of if someone gets hurt or not. Don’t touch her! Simple!...You are at fault! If she does not cook the man has for you and you are hungry and you go and drink authority to apeteshie (local gin) and you get drunk. Make sure abuse his wife you find something to eat and go and report her. Don’t say that because you have gone to drink apeteshie and she has not cooked so you are going to beat her. No! It is unacceptable. Simple! “(R2; elderly man) “Marrying her does not include beating. You cannot beat her. If you beat her and she takes you to the police station, they won’t say because she is your wife so you will be forgiven. You will be dealt with according to what the law says. That does not mean that that power is part. The fact that you come and marry her does not mean you should beat her. No. during the time of advice, some people even say “make sure you do not beat my daughter. The day I hear you have beaten her, I will take her back”. That is the meaning of performing the marriage rites and when it comes to the giving of advice, “I don’t want a man who will beat her. If I ever hear that you have beaten my daughter, I will take her back. So I 255 University of Ghana http://ugspace.ug.edu.gh do not want that”. So that power does not exist for you to beat her. But the power is for her to do what you want so that there can be understanding.” (R2; elderly man) Woman has not Bridewealth “You just married her, you haven’t bought her yet. been bought payment does Do you understand? So when you marry the woman not mean does not mean you should trouble and destroy her. woman has Do you understand?”(R6; elderly man) been bought “.. no you haven’t bought her. Even with this our tradition, when you are coming to marry a woman, we tell you. The day you will beat her up, when we call you, heed to our call…” (R11; elderly woman) Physical Husband can If the woman refuses to have sexual relations, he can abuse/beating sometimes even beat her up mercilessly.”(R1; elderly man) physically “…You even beat your child. If your child does abuse his wrong you beat him or her. It doesn’t mean you partner always descend on the woman to beat her. Sometimes when the woman does something which is wrong eerh but even that if the woman behaves in a certain no one will hear. But there is someone who is always descending on his wife to beat her but this is not right…” (R14; elderly woman) if you have married her and she does that (uses family planning) he can beat her” (R16;elderly woman) Authority to The husband “…no! Even if you will punish her, it’s not good that 256 University of Ghana http://ugspace.ug.edu.gh reprimand the has authority to you beat her”. (R5; elderly man) woman reprimand his “…Whether someone’s child misbehaves and your wife child misbehaves, who will you reprimand?... Your own child…. So there is no issue about that”. (R14; elderly woman) Social support Parents Parents are “As you are marrying the woman, the elders will valuable social even tell you that as you are taking her away. Her support the father himself will tell you that as you are taking couple can rely her, if you beat her up, you’ll be in trouble. If she on does anything wrong be patient and talk about it with her. If it doesn’t work, bring her to me.” (R5; elderly man) Friends Friends of the “…If you have issues with your wife, call your friend couple also and talk about it in the room so that no one will support the hear. It still exist.” (R12; elderly woman) couple Queen mother Traditional “We are often confronted with many of these issues leaders like … this will result in fights and after you investigate queen mothers and you will realise there is more to it. When the also address issue comes to the leaders and the people are bold to marital issues give the details and you also cannot say you will live with your wife without having a child, so why don’t you agree to the pregnancy or something like that. We can talk to them and there will be understanding. Sometimes you realise the woman gets pregnant again. She could not prevent it. As for that power, if the man says that the number of children is enough, then the power lies in the hands of the man. “(R15; 257 University of Ghana http://ugspace.ug.edu.gh elderly woman) Chief Chiefs are also “No matter how much she has wronged you, you available to bring the matter to me but do not beat her. Our address some traditional marriages do not allow that” (R8; marital elderly man) challenges Other relative Other “if she does something wrong, go and give it to the significant church to solve it. If the church can’t solve it, bring relatives also it to the family.” (R3; elderly man) support the “That is why when the marriage rites is getting to its couple to deal end, they ask all to give advice. They would advise with their you by talking to you and someone would tell you marital that quarrelling and beatings are not good. They challenges would also advise you the woman that as you are going you are marrying the man and his family. It is not that whenever a family member of the man comes for a visit, you cook and hide the food and claim that you are waiting for your husband. NO! It is not like that. Also, one responsibility of the woman is that she should not starve her husband. Starvation is in two type. Do you know the other one?” (R7; elderly man) 258 University of Ghana http://ugspace.ug.edu.gh Appendix L: Organizing themes and basic themes for sanctions/strategies for dealing with abuse Global theme Organizing themes Basic themes Description Sample quotes Sanctions/strateg Divorce Divorcing an “.. if I marry I and you are always beating me, I ies for dealing abusive husband will leave you. I won’t even marry you at all. I will with abuse can end the series take you to my family. If you refuse to go, the of abuse marriage will end.” (R12; elderly woman) “…For some people if you beat their daughter then the marriage will be dissolved.” (R8; elderly man) Separation Temporary or long “Some can make their daughters stay with them for time separation a long time to punish you very well” (R8; elderly from an abusive man) partner “Because some of such people when you beat them may go to their mother for like one week. Within this one week that she hasn’t returned, she hasn’t cooked for you so before she will come and prepare the food, she will have to kill a fowl before you should eat the food.” (R11; elderly woman) Legal/police case Formal legal “This time, if you beat a woman, the government strategies to will deal with you. That is all” (R1;elderly man) dealing with abuse “In these present times, if you beat up a woman it becomes a police case.” (R16; elderly woman) Deny husband sex Wife may deny the “there are some women that they won’t even sleep abusive partner sex with their husbands for a whole year” (R10; elderly woman) Warning Abusive husband “I got up and nearly beat the man in the church premises. I warned him that if he dare touches my 259 University of Ghana http://ugspace.ug.edu.gh may be warned to daughter, I give him two weeks to return my deter him from daughter. He saw that a man has also arrived. I told repeating such him I was not an old man….Not my daughter.” (R1; behaviour elderly man) Conflict resolution Fines (cash or Abusive partner “If you go and mistreat her, you will run at a lost kind) may be asked to and you’ll come and pay. That is why we are saying compensate the that, when there is a problem, it is for you the man, wife in cash or but when she acquires something, she has to bring it kind home.” (R3; elderly man) R: yes. He also gives the fowl for the woman to cook with it. You understand. If he doesn’t have some, he would have to go and buy it.” (R11; elderly woman) Summoned Abusive partner “Over here when you do that, we will think that you would be are a villager. That is the way it is. That you are a summoned to the villager. It is a villager who beats his wife. And if chiefs or elders you are summoned before us, by the time we release you, you will be weak.” (R2: elderly man) “oh if the woman gives reports that this is what my husband is doing, we can summon him to come to see the elders so that he will compensate the woman.” (R4: elderly man) Apology/sha Abusive partner “ you will apologize to the woman. or do something ming would be asked to to compensate her. In the same way when the apologize or woman also wrongs the man” (R 10; elderly woman maybe disgraced/shamed 260