Attitudes toward Homosexuals: Assessing the Structure of Prejudicial Attitudes and the Moderating Effects of Religious Commitment and Morality By Angela Anarfi Gyasi-Gyamerah This thesis is submitted to the University of Ghana, Legon, in partial fulfilment of the requirements for the award of PhD Psychology degree December 2014 University of Ghana http://ugspace.ug.edu.gh i Declaration I, Angela Anarfi Gyasi-Gyamerah, author of this thesis, hereby declare that except for references to other people‘s works which have been duly acknowledged, the work presented here was done by me as a student of the Department of Psychology, University of Ghana, Legon, under the supervision of Professor S. A. Danquah, Professor C. Charles Mate-Kole, Professor J. Y. Opoku, and Professor Charity S. Akotia. This work has never been submitted in whole or in part for any degree elsewhere. ………………………………… ...................………… Angela Anarfi Gyasi-Gyamerah Date Student ………………………………… ...................………… Professor S. A. Danquah Date Supervisor ………………………………… ...................………… Professor C. Charles Mate-Kole Date Supervisor ………………………………… ...................………… Professor J. Y. Opoku Date Supervisor ………………………………… ……………………… Professor Charity S. Akotia Date Supervisor University of Ghana http://ugspace.ug.edu.gh ii Dedication To my Sweethearts – Kwasi Gyasi and Nana Kwaku. You make it all worthwhile. This is for you! And to my parents – Prof. John K. Anarfi and Mrs. Agnes Anarfi. But for the sacrifices you made to ensure that I got all the care and support needed to train me up the way I should go, I won‘t be where I am today. Words cannot convey how much I love you all. Thank you for being there for me. University of Ghana http://ugspace.ug.edu.gh iii Acknowledgements Although this thesis was authored by me, I received support from many people through the process and would like to thank all of them. To my supervisors, Prof. Danquah, Prof. Mate-Kole, Prof. Opoku, and Prof. Akotia, I say a heartfelt thank you for your constructive feedback on the many drafts I sent to you all throughout the process. Your insights and invaluable advice have contributed to the quality of my work and for this I am very grateful. Prof. Danquah and Prof. Opoku, but for the two of you, I would not have been in academia and therefore would not have gotten this far. Thank you so much for believing in me and for ensuring that I got all the support and encouragement needed in my career and for the completion of this thesis. I am most indebted to you both. God richly bless you all! Prof. Akotia, you have not only been a supervisor to me, but a mother and a friend. No matter how busy you were, you always found time to listen to my whining when the journey got tough and provided encouragement that always got me back on my feet and believing that I will accomplish this. Maggie, colleague extraordinaire, I say thank you for always checking up on me and my progress and prodding me along every now and then till I reached the finish line. You are a true friend. Inusah and Osafo, I could not have made it without your support as colleagues but it is your friendship that I cherish the most. Thank you both! Chris, Donald, Kakra, Bill, Evans, Elias, Hannah, Julia, and DK, I owe you all a wealth of thanks for being there for me. Sister Circle members, sisters of my heart, thanks for your support, prayers, and encouragement. We are only about a year old but your impact on my life and therefore on my work cannot be time bound. Daddy, you are who I aspire to be in my life and in my career. Thank you for showing me the way and for ensuring that I never settle for less than God‘s best. Mom, thank you for being my number one cheerleader and always urging me on. I could not University of Ghana http://ugspace.ug.edu.gh iv have asked for better parents than you two. Kwasi Gyasi, my loving husband and best friend, thank you for your immeasurable support and willingness to shoulder a lot of the responsibility in managing our home in spite of your own schooling and work schedule especially when I needed to meet my deadline. Nana Kwaku, you came when I had just started this work, and although I agonized about leaving you so I could finish up, you quickly became the reason why it was worthwhile to accomplish this. Thanks to you and your Dad for being such a blessing to me. Afia, dearest daughter, what would I have done without you? This process would have been even more difficult without your help. God richly bless you. I would also like to thank all the students who participated in my study, especially those who shared their sexual orientation and details of their personal lives with me. This thesis would not have been possible without all of you. And last but not least, to God Almighty, who saw it fit to cast my lot in the same age and place as all the people mentioned here, be glory, honour, and praise for ever more. Amen!!! University of Ghana http://ugspace.ug.edu.gh v Abstract Homosexuality has been a topic for debate for quite some time in Ghana and a lot of negative sentiments have been expressed publicly about it. This study therefore examined the structure of prejudicial attitudes toward homosexuals and the extent to which these attitudes are moderated by religious commitment and morality. It also examined the role of attributions in determining attitudes toward homosexuals and how homosexuals in turn view societal attitudes toward themselves. To achieve these, three studies (Ns = 190, 183, 159) were conducted using mixed methods to gather data from students of the University of Ghana. Overall, participants were found to have prejudicial attitudes that are solely driven by stereotypic beliefs and moderated by religious commitment and moral values. Both Study 1 and Study 2 found that attitudes toward homosexuals are negative regardless of personal characteristics. Unlike was the case in Study 1, in Study 2, the attitude components were associated with each other and their effects on the evaluation of homosexuals were moderated by religious commitment and moral values. Additionally, evaluation of homosexuals was relatively more favourable in the post-test condition although still negative and the most favourable behavioural intentions were towards the vignette 3 person. In Study 3, it emerged from the FGDs that religion permeates all aspects of Ghanaians‘ lives although their religious commitment and morality is low albeit with justification. The negative attitudes found in Studies 1 and 2 were replicated here also as was the favourable behavioural intentions toward the vignette 3 person. The IDIs with homosexual students generally confirmed these findings. Limitations of the study and their implications for future research are discussed. University of Ghana http://ugspace.ug.edu.gh vi Table of Contents Content Page Declaration ............................................................................................................................. i Supervisor .............................................................................................................................. i Dedication ............................................................................................................................. ii Acknowledgements ............................................................................................................. iii Abstract ................................................................................................................................. v Table of Contents ................................................................................................................. vi List of Tables........................................................................................................................ ix List of Figures ...................................................................................................................... xi Chapter One ......................................................................................................................... xi Introduction ........................................................................................................................... 1 Overview of the Study ....................................................................................................... 1 Statement of the Problem ................................................................................................ 21 Rationale for the Study .................................................................................................... 23 Aims and Objectives of the Study ................................................................................... 24 Chapter Two ........................................................................................................................ 26 Literature Review ................................................................................................................ 26 Introduction ..................................................................................................................... 26 Theoretical Framework ................................................................................................... 26 Criticisms and Relevance of the Theories ....................................................................... 31 Review of Related Studies............................................................................................... 33 Summary ......................................................................................................................... 46 University of Ghana http://ugspace.ug.edu.gh vii Conceptual Model ........................................................................................................... 48 Hypotheses and Research Questions ............................................................................... 50 Operational Definition of Terms ..................................................................................... 53 Chapter Three ...................................................................................................................... 54 Study One ............................................................................................................................ 54 Introduction ..................................................................................................................... 54 Methodology ................................................................................................................... 54 Results ............................................................................................................................. 62 Discussion ....................................................................................................................... 72 Conclusion and Observed Model for Study One ............................................................. 81 Chapter Four ....................................................................................................................... 84 Study Two ........................................................................................................................... 84 Introduction ..................................................................................................................... 84 Methodology ................................................................................................................... 84 Results ............................................................................................................................. 91 Discussion ..................................................................................................................... 112 Conclusion and Observed Model for Study Two .......................................................... 118 Chapter Five ...................................................................................................................... 120 Study Three ....................................................................................................................... 120 Introduction ................................................................................................................... 120 Methodology ................................................................................................................. 121 Data Organization and Analyses of Interviews ............................................................. 127 Findings ......................................................................................................................... 131 University of Ghana http://ugspace.ug.edu.gh viii Discussion ..................................................................................................................... 200 Conclusion ..................................................................................................................... 211 Chapter Six ........................................................................................................................ 217 General Discussion............................................................................................................ 217 Overview ....................................................................................................................... 217 Summary and Discussion .............................................................................................. 217 Other Recommendations ............................................................................................... 230 Conclusion and Final Observed Model ......................................................................... 231 References ......................................................................................................................... 236 Appendices ........................................................................................................................ 247 Appendix A: Group Evaluation Questionnaire (GEQ) ................................................. 247 Appendix B: Group Evaluation and Behavioural Intention Form (GEBIF) ................. 255 Appendix C: Focus Group Discussion Guide ............................................................... 257 Appendix D: In-Depth Interview Guide ........................................................................ 260 Appendix E: Vignettes .................................................................................................. 261 Appendix F: Informed Consent Forms .......................................................................... 267 Appendix G: Ethical Clearance ..................................................................................... 280 University of Ghana http://ugspace.ug.edu.gh ix List of Tables Table 1 Demographic Characteristics of Respondents: Study 1 ......................................... 55 Table 2 Evaluation of Target Groups: Study 1 ................................................................... 62 Table 3 Bonferroni Multiple Comparison of Evaluation of Target Groups: Study 1 ........ 63 Table 4 Summary of Regression Analyses Examining the Influence of Respondents‘ Personal Characteristics on Evaluation of Homosexuals ........................................ 65 Table 5 Correlations among Attitude Components (Cognitive and Affective) .................. 66 Table 6 Most Frequently Elicited Responses: Study 1 ....................................................... 66 Table 7 Summary of Regression Analyses Examining Predictors of Evaluation of Homosexuals: Study 1 ............................................................................................ 67 Table 8 Summary of Intercorrelations for Evaluation of Homosexuals, Attitude Components, and Moderator Variables: Study 1 .................................................... 69 Table 9 Summary of Regression Analyses Examining the Effect of the Moderator Variables on the relationship between Stereotypic Beliefsa, Symbolic Beliefsb, and Evaluation of Homosexuals .................................................................................... 71 Table 10 Demographic Characteristics of Respondents: Study 2 ...................................... 85 Table 11 Summary of Experimental Design and Content of Vignettes .............................. 86 Table 12 Participants‘ Behavioural Intentions .................................................................... 88 Table 13 Cross-tabulation of Gender and Experimental Groups/Conditions ..................... 90 Table 14 Evaluation of Target Groups: Study 2 ................................................................. 91 Table 15 Bonferroni Multiple Comparison of Evaluation of Target Groups: Study 2 ....... 92 Table 16 Correlations among Attitude Components (Cognitive and Affective) – Study 2 ................................................................................................................................. 94 Table 17 Most Frequently Elicited Responses: Study 2 ..................................................... 95 University of Ghana http://ugspace.ug.edu.gh x Table 18 Summary of Regression Analyses Examining Predictors of Evaluation of Homosexuals: Study 2 ............................................................................................ 95 Table 19 Summary of Intercorrelations for Evaluation of Homosexuals, Attitude Components, and Moderator Variables: Study 2 .................................................... 97 Table 20 Summary of Regression Analyses Examining the Moderating Effect of Religious Commitment and Moral Values on the relationship between Stereotypic Beliefs and Evaluation of Homosexuals .................................................................................... 99 Table 21 Summary of Regression Analyses Examining the Moderating Effect of Religious Commitment and Moral Values on the relationship between Symbolic Beliefs and Evaluation of Homosexuals .................................................................................. 102 Table 22 Summary of Regression Analyses Examining the Moderating Effect of Moral Values on the relationship between Affect and Evaluation of Homosexuals ........ 105 Table 23 Summary Statistics and Paired Samples T-Test on Evaluation of Homosexuals ............................................................................................................................... 107 Table 24 Summary of Independent Samples T-Test on Gender Differences in Evaluation of Homosexuals Pre-test Scores ............................................................................ 109 Table 25 Summary of One-way ANCOVA on Gender Differences in Evaluation of Homosexuals Post-test Scores .............................................................................. 109 Table 26 The Influence of Experimental Condition on Participants‘ Behavioural Intentions ............................................................................................................................... 110 Table 27 Tukey HSD Multiple Comparison of Influence of Experimental Condition on Participants‘ Behavioural Intentions ..................................................................... 110 Table 28 Summary of Groupings ...................................................................................... 124 University of Ghana http://ugspace.ug.edu.gh xi List of Figures Figure 1: Conceptual model showing the expected outcome of this study (Author‘s Construct). .......................................................................................................... 49 Figure 2: Evaluation thermometer showing points corresponding with Study 1 respondents‘ mean evaluation scores for the various target groups. O.C.P. = Orthodox Church Priests; S.C.L. = Spiritual Church Leaders; T.P. = Traditional Priests; CSWs = Commercial Sex Workers; H. = Homosexuals. ...................... 64 Figure 3: Observed model showing findings of Study 1..................................................... 83 Figure 4: Evaluation thermometer showing points corresponding with Study 2 participants‘ mean evaluation scores for the various target groups. O.C.P. = Orthodox Church Priests; S.C.L. = Spiritual Church Leaders; T.P. = Traditional Priests; CSWs = Commercial Sex Workers; H. = Homosexuals. ...................... 93 Figure 5: Moderation graph showing the moderating effect of participants‘ religious commitment on the relationship between their stereotypic beliefs and their evaluation of homosexuals. .............................................................................. 100 Figure 6: Moderation graph showing the moderating effect of participants‘ moral values on the relationship between their stereotypic beliefs and their evaluation of homosexuals. .................................................................................................... 101 Figure 7: Moderation graph showing the moderating effect of participants‘ religious commitment on the relationship between their symbolic beliefs and their evaluation of homosexuals ............................................................................... 103 Figure 8: Moderation graph showing the moderating effect of participants‘ moral values on the relationship between their symbolic beliefs and their evaluation of homosexuals ..................................................................................................... 104 Figure 9: Moderation graph showing the moderating effect of participants‘ moral values on the relationship between their affect and their evaluation of homosexuals. 106 Figure 10: Evaluation thermometer showing points corresponding with Study 2 participants‘ mean evaluation scores in the pre-test and post-test conditions. . 108 University of Ghana http://ugspace.ug.edu.gh xii Figure 11: Observed model showing findings of Study 2................................................. 119 Figure 12: Thematic map showing summary of themes and sub-themes under ‗Religiousity and Religious Commitment‘. ............................................................................ 137 Figure 13: Thematic map showing summary of themes and sub-themes under ‗Morality‘. .......................................................................................................................... 150 Figure 14: Thematic map showing summary of themes and sub-themes under ‗Homosexuality‘. .............................................................................................. 160 Figure 15: Thematic map showing summary of themes and sub-themes under Vignettes 1A and 1B. ........................................................................................................ 166 Figure 16: Thematic map showing summary of themes and sub-themes under Vignettes 2A and 2B. ........................................................................................................ 171 Figure 17: Thematic map showing summary of themes and sub-themes under Vignettes 3A and 3B. ........................................................................................................ 176 Figure 18: Thematic map showing summary of themes and sub-themes under Vignettes 4A and 4B. ........................................................................................................ 181 Figure 19: Thematic map showing summary of themes and sub-themes derived from interviews with male homosexuals. .................................................................. 192 Figure 20: Thematic map showing summary of themes and sub-themes derived from interviews with female homosexuals. ............................................................... 199 Figure 21: Conceptual model showing a proposed sexual orientation continuum pertaining to Ghanaians. .................................................................................................... 214 Figure 22: Observed model showing findings of all parts of this 3-in-1 study. ............... 233 University of Ghana http://ugspace.ug.edu.gh 1 Chapter One Introduction Overview of the Study Homosexuality is the sexual orientation in which one‘s emotional and sexual attraction are directed toward members of his/her own sex (Strong, DeVault, Sayad, & Yarber, 2002). LeVay and Valente (2003) agree with Strong et al.‘s definition but further added that the way homosexuality is defined has changed over time and is currently thought of as people‘s sexual feelings that are predominantly directed toward persons of the same sex as themselves. This current definition presupposes that any given person can exhibit any of the sexual orientations, but might lean more towards or be more comfortable with one of them and therefore exhibit that particular orientation more often. It also means that sexual orientation is not a dichotomy but a continuum with human beings falling on different points on it. The view of sexual orientation as a continuum should however not be seen as an exclusively modern notion as Sigmund Freud (1856 – 1939) stated as far back as 1905, that everyone is born with bisexual tendencies and that expressions of homosexuality can be a normal phase in a person‘s heterosexual development. He however maintained that adult homosexual behaviour is due to ‗arrested‘ psychosexual development (Drescher, 2009). Arriving at the notion about the existence of different sexual orientations did not just happen. The term ‗homosexuality‘ itself did not exist until 1868 when it was coined by Karl Maria Benkert, a Hungarian writer, to describe same-sex attractions and sexual behaviour in humans (Feray & Herzer, 1990; Gulia & Mallick, 2010). According to Katz (1995), the term ‗heterosexuality‘ was coined even later, but before then, the concept of ‗sexuality‘ and the attractions, behaviours, and relationships that are now characterized as ‗heterosexual‘ or ‗homosexual‘, were understood quite differently from today. Currently, University of Ghana http://ugspace.ug.edu.gh 2 the concept of sexuality and the fact that different sexual orientations exist, are well- known. According to Kottak (2004) and Strong et al. (2002), there are four (4) sexual orientations found across the world, including homosexuality, heterosexuality (i.e. emotional and sexual attraction between men and women), bisexuality (i.e. emotional and sexual attraction to both men and women), and asexuality (i.e. indifference toward, or lack of attraction to, either sex). However, due to the efforts by sexual minority individuals to avoid sexual labels and also to generate more inclusive terminologies to describe variations in human sexuality, new terms have emerged. These include such terms as pansexual, humansexual, mehsexual, pomosexual, and bi-curious (Crowley, 2010; Thompson, 2008). Drobac (1999) stated that pansexuality includes heterosexuality, homosexuality, bisexuality, homoerotic and heteroerotic play, sexual aggression, and other sexual behaviour such as masturbation, celibacy, fetishism, and fantasy. Based on this definition, one can deduce that Drobac is also of the view that sexual orientation is a continuum, especially considering her assertion that all human beings are pansexual because each individual has the ability to manifest more than one form of sexuality. Many people cannot be pigeonholed into heterosexual, homosexual, or bisexual categories, especially because for some people, these subcategories fail to describe them adequately, since their biological sex and/or gender itself is in transition (Drobac, 1999). The etiology of homosexuality. Human sexuality, as shown in the above discussion, is a complex concept. Byer, Shainberg, and Galliano (1999) acknowledged this fact, stating that human sexuality is regarded by most scholars as a complex biopsychosocial behaviour, but perhaps what is even more complex is the etiology of these various forms of sexuality. As homosexuality is what is of interest to this study, this part of the discussion is limited to its etiology. According to Byer et al., in Plato‘s Symposium University of Ghana http://ugspace.ug.edu.gh 3 (360 B.C.E.), one of the participants (Aristophanes) gave an account of creation to explain the existence of different sexual orientations. In this account, he stated that human beings originally consisted of double creatures having four arms, four legs, and two sets of genitals so that they were double females, double males, or male-female hybrids. These double creatures were eventually cut in two by an angry god (Zeus) to form the two-legged humans we are today and since sexual desire is basically the desire to reunite with one‘s other ancestral half, those descended from the double females became lesbians, those descended from the double males became gay, and those descended from hybrids became heterosexual men and women. Today, researchers are also interested in determining the etiology of homosexuality in a much more scientific manner. According to Sheldon, Pfeffer, Jayaratne, Feldbaum, and Petty (2007), among the scientific studies examining the etiology of homosexuality, much work have focused on the possible influences of one‘s genetic make-up. Viewing the origins of homosexuality as genetic implies that it is immutable and uncontrollable. If so, then it is not a choice or an alternative lifestyle. Sheldon et al. (2007) noted that the factors contributing to an individual‘s sexual orientation still remain poorly understood by science, a fact that is evidenced by the use of general terminology in this field. While the use of the term ‗sexual orientation‘ presupposes a relatively neutral stance on etiology, the use of the term ‗sexual preference‘ presupposes a certain degree of individual choice. Furthermore, scientific explanations that have been proposed to explain the origins of homosexuality fall into two major categories including biological theories and psychosocial theories. Biological theories proposed anatomical, hormonal, and/or genetic factors (e.g. Bailey & Pillard, 1991; Brown, Fin, Cooke, & Breedlove, 2002; LeVay, 1991) but a lot of such studies have been critiqued as having problems with sampling bias, University of Ghana http://ugspace.ug.edu.gh 4 reliability of measures, generalizability, representativeness, controls, operationalization of variables, and lacking replication (Sheldon et al., 2007). These studies have not been able to pinpoint a single specific gene as responsible for homosexuality and have not been able to identify any genetic region in lesbians as linked to their sexual orientation. The situation is the same for studies that have tried to put forth psychosocial theories (e.g. Bem, 1996; Cass, 1979; Peplau & Cochran, 1990). Such studies have investigated the roles that individual choice and decision making, identity development, societal hierarchies, gender role stereotypes, role expectations, and conformity play in the etiology of homosexuality (Sheldon et al., 2007). According to Haslam and Levy (2006), the scholarly debates on the etiology of homosexuality are between constructionists on one side, and essentialists on the other. The argument of the constructionists is that sexual orientations are socially constituted identities and sexuality is subject to wide cultural and historical variations. Essentialists on the other hand argue that sexual orientations are ―objectively occurring categories that are grounded in biology, difficult to change, and likely to appear, with limited cultural shaping, wherever and whenever people have lived‖ (Haslam & Levy, 2006; p. 472). Obviously, each pair of biological theories and essentialist beliefs, psychosocial theories and constructionist views, is referring to the same ideas, but as Sheldon et al. noted, none as yet has the complete approval of the scientific community. Social history of homosexuality and homosexual experiences in the Western world. Societies and cultures the world over, especially the western societies, have gone through various stages of defining and redefining sexuality and sexual roles. Pickett (2011) in discussing homosexuality gave a comprehensive picture of how this notion of the existence of different sexual orientations came about through a review of the social history of homosexuality. He suggested that the social history of homosexuality could be University of Ghana http://ugspace.ug.edu.gh 5 divided into three eras, notably the Ancient Greece, the Medieval, and the Modern eras. The following discussion of the social history of homosexuality therefore follows these specific eras. The Ancient Greece era and similar cultures. According to Pickett (2011), in the Ancient Greece era, the assumption was that individuals can respond to beauty in either sex erotically and so the gender that one was attracted to was not important. What was of importance were other issues including status concerns, whether one exercised moderation, and whether one took the active or passive role in sexual relationships. The concerns with status dictated that free men could not have sexual relations with other free men, but only with women, male slaves, or male youths. Only free men had full status and every other person was inferior. Pickett adds that within that era, the cultural ideal of a same-sex relationship was between an older man in his 20‘s or 30‘s, known as the erastés, and a boy whose beard had not yet begun to grow, known as the eromenos or paidika. It must be noted that even within this cultural ideal, the eromenos often avoided penetration and rather opted for what is known as intercrural sex which involved the partners facing one another and the erastes placing his penis between the thighs of the eromenos. Hans Licht (1875–1929) explained in his book titled Sexual Life in Ancient Greece, translated and published in 1932, that the word ‗boy‘ should not be understood merely in the modern colloquial sense as: it is never a question of boys (as we mostly use the word), that is, of children of tender age, but always of boys who are sexually matured, that is, who have reached the age of puberty.……corresponding sometimes to what we should call ‗a young man‘….[today] (p. 369). It must be noted however that this sexual relationship between man and boy was not regarded as pedophilia, at least not in the sense that the word is used today. According University of Ghana http://ugspace.ug.edu.gh 6 to Pickett (2011), such relationships between a man and a boy were to be temporary. As such, during the courtship ritual (which usually happened before such relationships are contracted), the erastés had to show that he had nobler interests in the boy and the boy was not to submit easily, but to show discretion by picking a more noble partner if his attention was sought by more than one man. Also, in accordance with the dictate that such relationships were to be temporary, relations were broken when the boy reached adulthood (Dover, 1989). The eromenos, upon reaching adulthood and having the opportunity then to be an equal citizen, was expected to discontinue being in a submissive role. Failing to do this was considered troubling, although some adult male same-sex relationships existed and were noted, but not strongly stigmatized (Pickett, 2011). Considering all these, it is obvious that during the Ancient Greece era, although the passive role was seen as problematic and therefore stigmatized, a man‘s attraction to other men was often taken as a sign of masculinity and therefore not conceived as homosexuality and also not labeled as such. Kottak (2004) described other cultures that have some similarities to the Ancient Greece era, including the Etoro and the Kaluli, both of Papua New Guinea. The Etoro believe that men naturally have a limited supply of semen which can be depleted through frequent heterosexual activity, thereby sapping the man‘s vitality. They also believe that young boys cannot produce their own semen and therefore need to acquire this source of male vitality orally from older men. Based on these beliefs, Etoro boys are, from the age of 10 to adulthood, inseminated by older men. This practice by the Etoro is governed by a code of propriety that discourages same-sex activity between boys of the same age, because the boy who gets his colleague‘s semen is believed to be sapping his life force and thereby stunting his growth. So, a boy seen to be developing more rapidly than his peers is suspected to be ingesting semen from other boys and is therefore branded a witch. The University of Ghana http://ugspace.ug.edu.gh 7 Kaluli on their part believe that semen has a magical quality that promotes knowledge and growth. As such, when a boy reaches the age of 11 or 12, an older man who is not a relative is chosen by his father and the boy forms a sexual relationship that entails anal sex with this man. A peach-fuzz beard that appears on the boys‘ faces after the start of such relationships is cited by the Kaluli as evidence that the relationships promote growth. Another practice that is seen as part of the growing process is the extended period of time spent at hunting lodges by young Kaluli men and older bachelors. The older bachelors use this time to teach the young men the lore of the forest and hunting and also have homosexual intercourse (Kottak, 2004). Evidently, the Etoro and the Kaluli, as well as the Ancient Greeks, believe that same-sex activities are very vital for the growth of young boys into manhood and it is also obvious from their beliefs and the reasons behind these practices that they do not see homosexuality as a sexual orientation that one may ascribe to per se. They may not compartmentalize sexual expression into various forms at all but see their sexual practices as the way things should be because they believe that it is essential for the growth of the males within the society. The medieval era. This era, according to Pickett (2011), was quite different from the Ancient Greece era and by extension, from the Etoro and the Kaluli way of life as well. He noted that the gender of one‘s sexual partner was more important in the medieval era and so, a person who engages in same-sex activities was referred to as a sodomite. This label was however not reserved for same-sex partners only as any person who succumbed to the temptation of engaging in certain non-procreative acts was seen as a sodomite. This was an era where sexuality was defined against the backdrop of a broader theological framework that set the standard of morality. This standard of morality was drawn by the University of Ghana http://ugspace.ug.edu.gh 8 Catholic Church such that homosexual sex, extramarital sex, non-procreative sex within marriage, and masturbation, were forbidden. This meant that even for married couples, any sexual relations other than for the sole purpose of procreation, was forbidden and seen as immoral. As such, those who engaged in heterosexual anal sex were also considered sodomites. However, a person who had same-sex desires but never actually had same-sex sexual relations and a person who was a sodomite but had repented of his sin and vowed never to do it again, were not sodomites. Clearly, it was engaging in the act that put a person in a particular category as the emphasis was on a ‗committing sin versus refraining from sin‘ dichotomy (Pickett, 2011). The modern era. This era developed steadily through the 18th and 19th centuries when, according to Pickett (2011), the discourse about same-sex attraction was no longer dominated by a theological framework but by secular arguments and interpretations. The modern era brought the rise of the notion of ‗homosexuality‘, which placed a person into a specific category even when the person does not act upon his/her inclinations towards same-sex relations. The field of medicine (including psychology) was therefore the most important domain for the discussion of homosexuality; a situation which was occasioned due to doctors being appointed by courts to examine sex crime defendants as professional witnesses. With this rise in the prestige of medicine came the explanation of human sexuality as innate or biologically driven. So, unlike the medieval era where the sodomite is defined by his actions in choosing sin, the homosexual at the start of the modern era is shown to have a physical and mental make-up that makes him/her solely attracted to a person of the same-sex. Nonetheless, the homosexual was usually portrayed by society as somehow defective or pathological. This idea of pathology meant that the ‗homosexual‘ was seen as University of Ghana http://ugspace.ug.edu.gh 9 someone who is diseased and therefore needs treatment. Psychiatrists then worked towards gaining a repeal of criminal penalties for consensual homosexual practices but at the same time proposed interventions to rehabilitate homosexuals (Pickett, 2011). It is not surprising then that the first and second editions of the Diagnostic and Statistical Manual (DSM), published in 1952 and 1968 respectively, listed homosexuality under the category of sociopathic personality disturbance and sexual deviation respectively (Drescher, 2009; Pickett, 2011). For this reason, many gay people and lesbians were hospitalized for treatment, where they received very intense and sometimes very damaging medical counselling services (Robertson, 2004). According to Robertson, heterosexuality was seen as the only normal and natural sexual relation for humans. Having received this stamp of normality, and with the decline of theology‘s power over human sexuality, heterosexual sex lost a lot of its moral inhibitions that existed in the medieval era. The 1960s saw a redefinition of sexual roles and premarital sex gradually became more common and socially acceptable (Pickett, 2011). According to Herek (2007), ―love and sex came to be viewed as intimately related, and heterosexuality was understood by psychiatrists to be their mature, healthy expression‖ (p. 2). Before then, certain gay and lesbian rights groups had already been formed but were virtually secret societies. Gay rights activism and the depathologizing of homosexuality. The very first gay and lesbian rights groups worthy of note are the Mattachine Society and the Daughters of Bilitis respectively (Hay, 1996; Gallo, 2006). Although these two societies performed quite well in terms of organizing gays and lesbians, albeit in a low-key manner, it was not until the Stonewall Inn riots that occurred in New York City‘s Greenwich Village in 1969, that the gay liberation movement started full-force (Carter, 2005). Such activism led to an increased pressure on the fields of medicine and psychology to look at their definition of University of Ghana http://ugspace.ug.edu.gh 10 homosexuality but it was the famous study conducted by Evelyn Hooker (1907 - 1996) that actually got the reforms in the American Psychiatric Association (APA), and the American Psychological Association (APA) started eventually. It is important to note that before Hooker‘s study, Alfred Kinsey (1894 - 1956) and his colleagues had done research in the area of sexuality, and reported that homosexuality was more common in the general population than was generally believed, stating that it was 10% of the population (Drescher, 2009). Hooker (1957) compared 30 gay men with 30 heterosexual men using the TAT, the Make-a-Picture-Story test (MAPS test), and the Rorschach inkblot test with the aim of comparing the psychological functioning of the two groups. She intentionally chose a non-clinical sample of homosexuals with the assistance of the Mattachine Society so as to ensure that they were a match for the heterosexual sample. Based on her findings, she concluded that homosexuality was not a clinical entity and therefore was not inherently pathological. Herek (2007) stated that faced with such empirical evidence from Hooker‘s study and with the changing cultural views of homosexuality (which most likely were occasioned by the increase in gay activism), psychology radically changed its stance. The American Psychiatric Association‘s Board of Directors had voted to remove homosexuality from the DSM in 1973, and soon after, the American Psychological Association passed a resolution that stated in essence that homosexuality is not pathological and urged all mental health professionals to lead the way to ensuring that homosexuals are not stigmatized (Herek, 2007). Consequently, the classification of homosexuality as an illness ended but this did not necessarily mean an end to it being seen as pathology (Drescher, 2009). According to Drescher, the DSM-II and the DSM-III subsequently had in the place of homosexuality, Sexual Orientation Disturbance (SOD) and Ego Dystonic Homosexuality (EDH), University of Ghana http://ugspace.ug.edu.gh 11 diagnoses that regarded homosexuality as an illness if an individual with homosexual inclinations found them distressing and wanted to change. In spite of this, the fact that homosexuality was no longer seen as a mental illness, according to Herek (2007), crumbled one of heterosexism‘s supporting pillars, thereby influencing societal attitudes and providing a basis for reversing many of the antigay policies and laws that were enacted in the 20th century. This notwithstanding, the western world cannot be said to be totally accepting of homosexuality, although there is increasing evidence that changes have occurred. According to Altemeyer (2001), people with certain characteristics have been known to be prejudiced against homosexuals. These include racial and ethnic bigots and right-wing authoritarians. Such people still exist, so researchers have tried to find out what accounts for attitudes toward homosexuals becoming increasingly accepting and tolerant (e.g. Altemeyer, 2001; Loftus, 2001). A major reason for this change in attitudes in the Western world is increased contact with homosexual persons but the effect of essentialist beliefs about the etiology of homosexuality is also a factor (Altemeyer, 2001; Haslam & Levy, 2006). Altemeyer (2001) stated that another reason for the change in attitudes towards homosexuals is the fact that levels of right-wing authoritarianism have dropped. This drop, according to Altemeyer, is attributed to four things, the first being that people are less religious than they used to be. Parents of today attend church less than their parents did and therefore are not stressing family religion as much to their children. Secondly, due to the influence of the mass media, many people have learnt that sexual orientation is affected by genetic factors and so homosexuality seems less unnatural to them. Thirdly, HIV/AIDS (Human Immunodeficiency Virus and Acquired Immune Deficiency Syndrome) is no longer seen as a gay plague or an illness that is caused by homosexuality and affects only homosexuals, but as an avoidable and more controllable illness that can affect people of all University of Ghana http://ugspace.ug.edu.gh 12 sexual orientations. Lastly, certain religious groups have re-examined their condemnation of homosexuality [e.g. United Church of Christ; Presbyterian Church (USA)] and authourities such as the American government have revised its policy toward homosexuals in the military. The perception that many straight and prestigious people attach no stigma to homosexuality has therefore spread, making people think that their beliefs about homosexuals may be wrong (Altemeyer, 2001). Notwithstanding this, homosexuals in some parts of the world and especially in Africa still experience stigma mainly because their orientation is considered un-African. Homosexual experiences across Africa and Ghana. Morgan and Wieringa (2005) noted in their book titled Tommy Boys, Lesbian Men and Ancestral Wives that generally same-sex practices have been difficult to research in Africa due to the fact that many African leaders deem it an un-African practice imported from the west and therefore a taboo. Thus, with regards to the issue of homosexuality, the situation in other African countries is similar to that of Ghana. Taking Uganda as a case in point, after the British Prime Minister David Cameron‘s comments in 2011, the Ugandan presidential adviser, John Nagenda, accused Mr. Cameron of showing an ex-colonial mentality and of treating Ugandans like children (BBC.co.uk, 31 October, 2011). Søgaard (2013) has however disagreed with the assertion that homosexuality is un-African. In his thesis aimed at challenging the perception about homosexuality being un-African, he concluded that homosexuality as a concept is an European import but same-sex intimacy is as African as the colour of the skin. In many African countries, homosexuals face the possibility of death among other atrocities meted out to them by members of the society because of their sexual orientation. Morgan and Wieringa (2005) focused on African women in same-sex relationships and emphasized that for the women they interviewed, secrecy equals survival. The women University of Ghana http://ugspace.ug.edu.gh 13 stated that they ran the risk of losing whatever support they got from their families, being evicted from their homes, and experiencing all sorts of hate crimes including rape. Despite this risk and the fact that homosexuality is deemed as against the culture in most African countries, same-sex experiences not only existed but seemed to occur at very early stages for most homosexuals. For example, among the Kenyan women interviewed by Morgan and Wieringa, most had their first same-sex experience in primary or secondary school and the same is true for women they interviewed from other Southern African countries, including Tanzania, Uganda, Namibia, South Africa, and Swaziland. This is also true of homosexual experiences in Ghana. Kramer (2003) stated that the male homosexual participants in her study conducted in Ghana generally noted that they had their first homosexual experiences in secondary school, especially the single-sex secondary schools. These participants were students/graduates of the University of Ghana (UG) and University of Cape Coast (UCC) and for some of them, while in primary school, they knew they had homosexual tendencies, but did not follow up on their feelings until secondary school. However, such experiences do not necessarily occur only between persons who are both students of a particular secondary school. According to Kramer, some of her interviewees stated that they seek out new sexual relations from among boys in secondary school. This statement can be said to support Attipoe‘s (2004) assertion that boys and young men are being recruited into homosexual practices in Ghana. Banks (2012) in his article on the life of the legendary Ɔkɔmfo Kwabena noted a practice of sexual initiation or that can also be referred to as ‗recruitment‘. Ɔkɔmfo Kwabena was a fetish-priest and leader of a community of queer men in a town in the Central Region of Ghana and their engagement in nteteɛ was to give young men access into their community. Nteteɛ is a Fante (a language of the Akan people of Ghana) term which means ‗training‘. University of Ghana http://ugspace.ug.edu.gh 14 It should be noted that Banks (2012) did not interpret this practice as such, but his description of the practice as it is lived out by this community, who refer to themselves as Saso people, can be said to be open to varied interpretations; a fact he himself acknowledges. Relating the experiences of one Saso man whose first same-sex erotic experience was with Ɔkɔmfo Kwabena, he noted that the man had: described an occasion when Ɔkɔmfo Kwabena had explained to him that some of his friends who are kojobesia engage in same-sex erotic practices and inquired if he would be interested in learning about it….he agreed because of his curiosity [and] upon acceptance, Ɔkɔmfo Kwabena initiated him into same-sex erotic practices, and for a while, was his exclusive Saso partner (p. 14). Banks however added that even though some interpreters may view this as a form of ‗homosexual recruitment‘, nteteɛ is best understood as a practice of community formation through which the Saso people expand and perpetuate their community while offering the initiates the opportunity to learn about and explore their own same-sex erotic desires. Another reason why, according to Banks, this practice should not be understood as recruitment, is the fact that if a man has no interest in same-sex erotic practices in the first place, the initiation will not be effective. Members of the Saso community said, as noted by Banks, that any Ghanaian man can be positively responsive to nteteɛ if it is undertaken correctly and adapted to suit individual circumstances. Terminologies indigenous to the gay subculture. Aside the term nteteɛ, there are other terms that are specific to the gay subculture in Ghana. According to Kramer (2003) one of her interviewees explained to her that there are two types of gay males in Ghana, the Classics and the Locals. The Classics are cultured, travelled, well-educated, wealthy, and classy Ghanaian gay males while the Locals are uneducated, poor, trashy, and stereotypically gay in their dressing. Due to this tendency to dress stereotypically University of Ghana http://ugspace.ug.edu.gh 15 feminine, the Classics refer to the Locals as princesses. Banks (2012) also noted some terms, all in Fante, that are used by the Saso community he studied. As the Saso people like to distinguish between men who like to penetrate and those who like to be penetrated during anal intercourse, they use the terms ɛsor and ase respectively to refer to them. Those identified as ɛsor are usually expected to dress and act masculine while those identified as ase are usually allowed to perform varying degrees of femininity. The term kojobesia is also used to denote someone‘s identity as one who plays ase although it is a general term used to describe men who are effeminate. Also, the leader of the Saso community is referred to as Nana Hemaa (Queen mother), a term denoting the motherhood status of the leader and his attributes as someone upon whom the members of the Saso community depend for emotional, financial, and intellectual support. Ɔkɔmfo Kwabena was the Nana Hemaa during his lifetime. It seems that the use of special terminologies is part of the gay and lesbian subculture in various parts of Africa. For example, of those interviewed in Uganda by Morgan and Wieringa (2005), some of them identified as men and preferred to be called Tommy Boys instead of lesbians. In Ghana, lesbians are not easily noticeable as the gays because as Kramer (2003) found out, girls/women are affectionate in public whether they are lesbians or not, making it difficult to tell who is lesbian while boys/men are easily noticed if they display the kind of affection displayed by women towards each other. The inevitability of marriage. According to Kramer (2003), some of her interviewees‘ admitted that a lot of married men are gay and for one of them, married men were his only choice of sex partners because he felt being with a married man came with minimal health risks as the only other sexual partner the man has is his wife. It seems that marriage in the traditional sense, that is, between a man and a woman, is not only seen as inevitable amongst sexual minority individuals in Africa but as a necessity because of the University of Ghana http://ugspace.ug.edu.gh 16 protective cover it offers. Two of Kramer‘s female interviewees, a couple themselves, stated that they had their separate heterosexual relationships, and that these relationships served to provide the cover they needed to keep their homosexuality a secret. Some of her male interviewees referred to themselves as bisexuals because they would have to marry women in order to have children although they would have wished to marry men instead. They added that if society permitted same-sex marriage, then they would not have to marry women but only get them to bear their children since it is the only way they can have biological children. Morgan and Wieringa‘s (2005) interviewees also shared similar sentiments. One Kenyan interviewee noted that she will have to get married to a man as the Kenyan culture equates having a lot of children to being wealthy and so pressurize women to get married when they are of age to do so. In spite of this, some of them still held out hope of getting married to their same-sex partners. One interviewee in Uganda is noted to have said that though they cannot do it in public, she and her girlfriend plan to get married in future, perhaps secretly, in hopes that the government will eventually accept it. Obviously, culture does have an impact on the experiences of homosexuals in certain parts of Africa. Religion, morality, culture and homosexuality. The issue of homosexuality has been a topic of passionate discussion in Ghana for quite some time and even more especially after the nomination of a human rights lawyer as minister for Gender, Children, and Social Protection by the president of Ghana in January 2014. This was because she had made some statements in the previous years that people within the Ghanaian community saw as evidence that she was sympathetic to homosexuals. In the ensuing discussion on the nation‘s airwaves, it became increasingly obvious that the majority of Ghanaians are not ready to accept homosexuality and the main reasons often given for this stance is that it is foreign to our culture, morally wrong and against the nation‘s major University of Ghana http://ugspace.ug.edu.gh 17 religions (i.e. Christianity, Islam, and Traditional Religion). According to the 2012 Win- Gallup International‘s Global Index of Religiousity and Atheism, Ghana is the first among the top 10 religious populations of the world with 96% of Ghanaians stating that they are religious. Atiemo (2013) noted that religion in Ghana is not only important because of the number of citizens that claim religious adherence but also because of its intense manifestation in the daily lives of the people and the extent of the impact of its ideas. Indeed, there is no doubt that religion does have a great impact on the daily lives of Ghanaians because religious beliefs in Ghana are not a matter of purely personal concerns but are always part of the issues debated on in public in that all public matters, such as politics, the economy, health and education, and of course homosexuality, are subjects of religious discourse. Atiemo added that religion so permeates every aspect of Ghanaian life that even at workplaces and government organized programmes, prayers are said before the start of work or the start of the programme. On the various public universities‘ campuses in Ghana, religious activities take students‘ time so much so that gardens and open spaces such as parks turn into prayer grounds in the night and at dawn (Atiemo, 2013). According to Gyekye (2003), religion refers to ―the awareness of the existence of some ultimate, supreme being who is the origin and sustainer of this universe and the establishment of constant ties with this being‖ (p. 3). Mish, Gilman, McHenry, Pease, Jr., Bollard, Collier et al. (1990), defined it as ―the service and worship of God or the supernatural‖ and as ―a personal set or institutionalized system of religious attitudes, beliefs, and practices‖ (p. 995). Belonging to a particular religion requires a certain level of commitment in order to establish constant ties with the Supreme Being or whoever is the center of the particular religion that one ascribes to (Gyekye, 2003). Worthington (1988) defined religious University of Ghana http://ugspace.ug.edu.gh 18 commitment as the degree to which a person adheres to his or her religious values, beliefs, and practices and uses them in daily living. These definitions and the extent to which religion permeates all aspects of Ghanaian social life as noted above, make it clear that religion is not just what people do in their places of worship, be it a Church, Mosque, or Shrine, but it is a way of life for people, in that it determines how they live and what they would consider right or wrong. Having certain beliefs informs the kind of attitude a person will have regarding certain issues, and this will in turn affect how the person behaves and even what the person thinks is the right behaviour for every other human being. Also, most religions have strict moral codes, and Christianity, Islam and Traditional African Religion are no exceptions. As Drescher (2009) noted, official pronouncements on the meanings of same-sex behaviors were once primarily the province of religions and many of them deemed homosexuality as ‗bad‘. Morality, according to Gyekye (2003), is ―a set of social rules and norms intended to guide the conduct of people in a society‖ (p. 55). He added that these rules and norms are based on people‘s beliefs about right and wrong conduct and good and bad character. This is in line with Mbiti‘s (1989) view that ―the essence of African morality is that it is ‗societary‘ than ‗spiritual‘; it is a morality of ‗conduct‘ rather than a morality of ‗being‘. This is what one might call ‗dynamic ethics‘ rather than ‗static ethics‘ for it defines what a person does, rather than what he is‖ (p. 209). According to Gyekye (2003), African moral values are derived from the experiences of the people in trying to live together in harmony and so these values have a humanistic basis rather than a religious one. As such, what is morally good is that which brings about human well-being. Gyekye also noted that procreation is a very important African cultural value and is viewed as the whole purpose of marriage. In all the uproar about the issue of homosexuality within the public domain, one major point has been that homosexuality is a pattern of behaviour that is not good and University of Ghana http://ugspace.ug.edu.gh 19 does not promote human and social well-being, because it does not allow for procreation. In Ghana currently, and in many other countries all over the world, homosexuals cannot marry, and even if they could do so by law, the problem of procreation still remains. In effect, homosexuality is morally wrong. This is a fact that, as mentioned already, is sanctioned by a lot of religions, and though African morality is not necessarily based on religion, it does play a role in the moral lives of individuals (Gyekye, 2003). Culture, according to Tylor (1958, p. 1), is ―that complex whole which includes knowledge, belief, arts, morals, law, custom, and any other capabilities and habits acquired by man as a member of society‖. Geertz (1973, p. 44) also defines it as ―ideas based on cultural learning and symbols...that people gradually internalize [and use] to define their world, express their feelings, and make their judgements...[and also serves to help] guide their behaviour and perceptions throughout their lives‖. Looking at these definitions, one can safely conclude then that religion and morality are concepts that are encapsulated by culture. The culture of a particular people determines how they live their lives, the religion(s) they are likely to ascribe to and what they are likely to deem as morally right or wrong. So perhaps it makes sense that the majority of Ghanaians, within the realm of public discourse, say that homosexuality is against our culture because unlike the Kaluli and the Etoro cited above, homosexual behaviour is not institutionalized in Ghana as it is in their society. Prejudicial attitudes. As stated earlier, based on the content of public discourse on the issue of homosexuality, Ghanaians obviously have a negative attitude towards it, and as a result, are prejudiced against homosexual persons. According to Kessler and Mummendey (2008, p. 292), prejudice can be defined as ―a derogatory attitude or antipathy towards particular social groups or their members‖. Attitude, in itself, is defined as an overall evaluation of a stimulus object (Haddock & Maio, 2008) and any stimulus University of Ghana http://ugspace.ug.edu.gh 20 that can be evaluated along a dimension of favourability can be conceptualized as a stimulus object or an attitude object. With regard to the issue of antigay attitudes, Herek (2000) stated that the scientific analysis of its psychology could only be facilitated by the use of the term sexual prejudice. This term, according to Herek, is broadly conceived as all negative attitudes based on sexual orientation, whether the target is homosexual, bisexual, or heterosexual, but he uses it to denote heterosexuals‘ negative attitudes toward homosexual behaviour, people with homosexual or bisexual orientation and Lesbian, Gay, or Bisexual (LGB) communities. Like other types of prejudice, sexual prejudice has three principal features including the fact that it is an attitude (i.e. an evaluative judgement), it is directed at a social group and its members, and it is negative (i.e. involves hostility or dislike). In Herek‘s (2000) view, conceptualizing heterosexuals‘ negative attitudes toward LGB individuals as sexual prejudice is better than conceptualizing it as homophobia because it is a much more descriptive term. According to Adams, Wright, and Lohr (1996), homophobia refers to an affective response, including fear, anxiety, anger, and discomfort, that a person experiences when interacting with LGB individuals, which may or may not involve a cognitive component. This cognitive component is referred to as homonegativity and is defined by Hudson and Ricketts (1980) as negative attitudes, beliefs, or actions toward homosexuality. Aside the affective responses mentioned above as markers of homophobia, another affective response that heterosexuals are likely to experience towards gay persons in particular is disgust. For a lot of people, hearing the term homosexuality brings to mind a graphic image of two men having anal sex which they find disgusting and by association, they find a homosexual person disgusting too (e.g. Terrizzi Jr., Shook, & Ventis, 2010). According to Olatunji (2008), several domains of disgust, including core disgust, animal- University of Ghana http://ugspace.ug.edu.gh 21 reminder disgust, and contamination disgust, may relate to negative attitudes toward homosexuals. Core disgust pertains to bodily products such as semen which are commonly associated with gay activity, animal-reminder disgust pertains to the forcible breach of the exterior envelope of the human body which is associated with anal sex, and contamination disgust pertains to concerns about diseases and infection (Haidt, Rozin, McCauley, & Imada, 1997; Haidt, McCauley, & Rozin, 1994). Olatunji also noted that disgust appraisals may be more easily associated with gay sexual activity, but the consideration of lesbian sexual practices as ‗unnatural‘ may also motivate the same appraisals. Statement of the Problem Homosexuality has been a topical issue in Ghana for quite some time now but currently, although negative sentiments still prevail, they are overshadowed by concerns about the Ghanaian economy and other social problems. The truth however is that the issue of homosexuality is quite a complex one. It is more complex and therefore transcends the debate as to what is culturally acceptable and what is not. It is even not simply a matter of what is morally right or wrong and for that matter it is also not simply a question of what is sanctioned or not sanctioned by religion. Although the fact that homosexuality as a human rights issue is unacceptable to most Ghanaians, it is a fact that we cannot run away from. Considering this fact, the only question that can be raised about homosexuality is that of its criminality or otherwise. The laws of Ghana are not clear on the issue though. The Criminal Code Act (1960) states that: (1) Whoever has unnatural carnal knowledge — University of Ghana http://ugspace.ug.edu.gh 22 (a) of any person of the age of sixteen years or over without his consent shall be guilty of a first degree felony and shall be liable on conviction to imprisonment for a term of not less than five years and not more than twenty-five years; or (b) of any person of sixteen years or over with his consent is guilty of a misdemeanour; or (c) of any animal is guilty of a misdemeanour. (2) Unnatural carnal knowledge is sexual intercourse with a person in an unnatural manner or with an animal (ACT 646, §104). Often, it is the phrase ―unnatural carnal knowledge‖ that is used by laypeople and even some lawyers as the standpoint for arguing that homosexuality is illegal but there is no consensus on what the phrase actually means. The second part of the law, as shown above, tries to define it, but the attempt probes the question, ―what exactly is ‗unnatural manner‘‖? , deepening the confusion. So then, what we have are more questions: Is homosexuality an abnormality or not? Is it genetic or is it acquired? Can one be cured of it? Is homosexuality a sin? Are homosexuals going to hell? Should what happens in the bedroom between two consenting adults be their business and theirs only? The list of questions is truly endless but the answers are still not clear. One thing is clear though. While the scientific community and all those who have an interest in the issue try to find answers to these questions, the issue of attitudes remains pressing as negative attitudes taken to the extremes can mean the lynching to death of a human being alleged to be homosexual. For example, during the first and second weeks of March 2012, it was in the news that an angry mob of youth in James Town, a suburb of Accra, had assaulted a group of people within the community believed to be homosexuals. The attack is said to have happened during a marriage ceremony between two lesbians living in the community. The guests at the ceremony who University of Ghana http://ugspace.ug.edu.gh 23 were unable to run away were beaten up and some were arrested by the mob and handed over to the police who had to let them go because there was no legal reason to hold them (GhanaToGhana.com, March 19, 2012). The nature of such attitudes therefore need to be identified so that armed with that knowledge, these extremes can be managed until a time when Ghana, and for that matter Ghanaians, will decide on exactly how best to deal with the issue of homosexuality. It is against this background that this study was conducted. Rationale for the Study From the public discourse about homosexuality, it is obvious that most Ghanaians have a negative attitude towards homosexuality, arguing that it is immoral, against religious laws and our culture. There are however some other behaviours that can be said to be equally immoral and against religious laws and the Ghanaian culture, such as induced abortion, bribery/corruption, commercial sex work, incest, premarital/extramarital affairs, to name a few. Most religions preach against these, and expect their followers to desist from them. Also, none of them have ever caused this level of public outcry in opposition. Granted, some years back, there was a lot of discussion on the issue of commercial sex work, but not to the extent of pronouncements by government officials or religious leaders being highly passionate about it. In the news, there are daily reports of rape or murder, incest, teenage pregnancy, bribery/corruption scandals, etc. however, these only make the headlines, and are read in passing on radio and on television. This fact brings up the question, asking why these ‗other vices‘ do not attract the same amount of criticism from Ghanaians. So long as there are reports of such happenings in the news, then it can be assumed that a lot of Ghanaians are themselves engaged in these behaviours too, but do people complain? If they do, then the complaints are usually not loud enough for everybody to hear. So the question again is ―Why‖? Why University of Ghana http://ugspace.ug.edu.gh 24 are Ghanaians seemingly able to tolerate these other ‗vices‘ but not homosexuality? That is the question this study sought to answer. The rationale for conducting this study therefore stemmed from the fact that although anecdotal evidence makes it safe to assume that there is a negative attitude towards homosexuals in Ghana, there is very limited scientific evidence of this. As is obvious from the review of related literature however (see Chapter Two), the issue of homosexuality has piqued the interest of various professionals who are researching into it here in Ghana with varied focus. This study is relevant in that it does not only subject the assumption of a generally negative attitude to the process of science as previous studies have done (e.g. Anarfi & Gyasi-Gyamerah, 2014; Owusu et al., 2013). It also adds to the literature by going further to assess the underlying structure of these attitudes and how they are moderated by religious commitment and morality. Additionally, the fact that this study creates the opportunity for the whole issue of attitudes toward homosexuals to be viewed from the perspective of the Ghanaian homosexual persons themselves, shows its relevance. Aims and Objectives of the Study The main aim of this study was to determine the underlying structure of prejudicial attitudes toward homosexuals, and the specific objectives are to: 1. Determine the relative importance of stereotypes, symbolic beliefs, and affective associates in predicting attitudes toward homosexuals. 2. Examine the role of behavioural intentions in attitudes toward homosexuals. 3. Determine the extent to which religious commitment and morality will moderate attitudes toward homosexuals. 4. Ascertain the role of attributions in determining attitudes toward homosexuals. University of Ghana http://ugspace.ug.edu.gh 25 5. Examine how homosexuals view themselves and how they feel about the views of society. University of Ghana http://ugspace.ug.edu.gh 26 Chapter Two Literature Review Introduction This chapter reviews the literature that is pertinent to this study and is structured based on the funnel method proposed by Hofstee (2006). It begins with a discussion of the main theories upon which the study is based, its criticisms and relevance to the study, followed by a review of related studies, and ends with a summary of the review and the conceptual model that was tested in conducting this study. Theoretical Framework Multicomponent model of attitude (Zanna & Rempel, 1988). This model conceptualizes attitudes as summary evaluations that have affective, cognitive and behavioural components that, to some extent, are independent of each other. This means that it is not necessary for an individual to be consistent across the different sources and that s/he may have more than one attitude towards a stimulus object depending on which component the evaluative judgment is based on. The affective component. The affective component refers to feelings or emotions associated with an attitude object (Haddock & Maio, 2008). These feelings or emotions, known as affective responses, influence attitudes in a number of ways, the primary way being due to the affective reactions that are aroused in an individual after s/he is exposed to the attitude object. Negative affective responses are likely to make an individual have a negative attitude towards attitude objects. Research evidence suggests that feelings can become associated with attitude objects in several ways to produce a positive or negative attitude. Some researchers have used classical conditioning paradigms and the mere exposure phenomenon to achieve this University of Ghana http://ugspace.ug.edu.gh 27 association (e.g. Krosnick, Betz, Jussim, & Lynn, 1992; Murphy & Zajonc, 1993). Findings from these studies have shown that associating negative information with an attitude object makes one develop a negative attitude towards it and that being exposed to or having contact with an attitude object can make one develop a positive attitude towards it. The cognitive component. The cognitive component of attitudes refers to the beliefs, thoughts and attributes we associate with a particular object. As such, a person‘s attitude might usually be based primarily upon a consideration of the positive and negative attributes of the attitude object (Haddock & Maio, 2008). Such beliefs about the attributes of an attitude object, especially about a particular social group, are known as stereotypes. Numerous studies have shown that possessing negative stereotypes about a group of people is associated with having a prejudicial attitude towards the group (e.g. Esses, Haddock & Zanna, 1993; Kawakami, Dion & Dovidio, 1998). Stereotypic beliefs or cognitions, are a key part of one approach to attitudes, specifically the expectancy-value approach (Fishbein & Ajzen, 1975). This approach describes an attitude towards an object as the total sum of the product of one‘s expectancies multiplied by the value of each expentancy. According to Haddock and Maio (2008) expectancies are beliefs or subjective probabilities that the object possesses a certain attribute, which may range in strength from 0 to 1 while values, or evaluations, are ratings of the attributes, normally from -3 to +3. An attitude object will be evaluated positively if it is seen as leading to or being associated with positive things and avoiding negative things, and only salient beliefs (i.e. beliefs that a person considers most relevant) count towards the overall attitudes. Using this model, a person‘s attitude towards a particular stimulus object can be calculated. An example is given by Haddock and Maio (2008, p. 116) as follows: University of Ghana http://ugspace.ug.edu.gh 28 [a] person might think that golf is (1) a Valuable form of exercise, (2) a good way to see friends and (3) frustrating. Each of these beliefs will have both an expectancy and a value. For example, exercise might have a high expectancy (.9) and positive evaluation (+3); seeing friends might be perceived as having a lower expected outcome (.7) that is somewhat positive (+2); while frustration is (thankfully!) somewhat infrequent (.3) but very negative (-3). The individual‘s overall attitude towards golf is computed by summing the belief-evaluation products [e.g., 2.7 + 1.4 – 0.9 = 3.2]. Haddock et al. (1993) stated that a second type of belief that is relevant to the cognitive component of prejudicial attitudes is symbolic beliefs. These are beliefs that social groups either violate or promote the attainment of cherished values, customs, and traditions of a particular society. Symbolic beliefs can also be calculated the same way as shown above for stereotypic beliefs. The behavioural component. This refers to an individual‘s behavioural intentions or past behaviours associated with an attitude object. One way in which behaviours are linked to attitudes is that people sometimes infer their attitudes on the basis of their previous actions. This idea was developed by Bem (1972) and is explained through his self-perception theory, which states that when unsure of our attitudes, we infer them just as someone observing us would, by looking at our behaviour and the circumstances under which it occurs (Myers, 2002). Bem‘s argument was that this lack of access is especially likely when the person‘s attitude is particularly weak or ambiguous, a reasoning that has been confirmed by results from many studies (e.g. Chaiken & Baldwin, 1981). Another way in which behaviours may also influence strongly held attitudes is shown by Festinger‘s (1957) cognitive dissonance theory. This refers to an individual‘s internal state that is as a result of him/her noticing an inconsistency between two or more University of Ghana http://ugspace.ug.edu.gh 29 of his/her attitudes or between his/her attitudes and his/her behaviour (Baron et al., 2009). Behaviours also influence attitudes much more directly in that engaging in actions that have evaluative implications or connotations (such as nodding or shaking one‘s head and flexing or extending one‘s hand) influence the favourability of attitudes (Briñol & Petty, 2003; Cacioppo, Priester, & Berntson, 1993). Social learning theory (Bandura, 1997). This theory states that we learn social behaviour by observing and imitating others and by being rewarded and punished for our behaviour (Myers, 2002, p. 387). Bandura initially developed the theory to explain how aggression is learned in children but the theory can be applied to how various other behaviours and attitudes develop. According to Baron et al. (2009), our views are mostly acquired in situations in which we interact with others or observe their behaviour, a phenomenon they referred to as social learning. Social learning is the process through which we acquire new information, new forms of behaviour, or attitudes from others. So, whether a behaviour or attitude is positive or negative, we acquire them through observing others, especially significant others (e.g., parents). Of interest to this study are two variants of this theory that explain how prejudicial attitudes and religious attitudes are formed including the social learning view of prejudice and religious socialization. Social learning view of prejudice (Baron et al., 2009). This is the view that prejudice is acquired through direct and vicarious experiences in the same way as other attitudes are acquired. According to Baron et al. (2009), children acquire negative attitudes toward various social groups because they hear such views expressed by significant others and because they are directly rewarded (with love, praise and approval) for adopting these views. Also, people‘s own direct experience with members of other groups shape their attitudes. University of Ghana http://ugspace.ug.edu.gh 30 Religious socialization (Spilka, Hood, Hunsberger, & Gorsuch, 2003). Spilka et al. (2003) stated that Bandura‘s social learning theory can be considered as the theoretical underpinning of the socialization process which has an important implication for religiousness. According to them, religiousness is typically strongly influenced by one‘s immediate environment, especially one‘s parents, through both modeling and reinforcement processes. Apart from parents, many external influences have the potential to affect people‘s religiousness including peers, schools, religious institutions, books, the mass media, and so on. These external influences can affect people directly through explicit teachings or family practices or indirectly through influencing their school, marital, and career choices, or through cultural assumptions, subtle modeling, or lack of exposure to alternative positions. According to Spilka et al. (2003), many studies have linked various aspects of religiousness with increased discriminatory attitudes, adding that people who try to live their religion will also display discriminatory attitudes openly in dealing with the world around them. Attribution theory of controllability (Weiner, 1979, 1985; Weiner et al., 1988). This theory states that stigmatized behaviours that are believed to have biological origins will be evaluated positively while those that are believed to be due to individual choice will be negatively evaluated. In other words, individuals who are believed to have caused their stigma will be evaluated negatively than those who are stigmatized through circumstances beyond their control or as a result of misfortune. According to Haider- Markel and Joslyn (2008), Weiner‘s theory of controllability helps explain the underlying mechanisms involved in attributions of various stigmatized groups, including homosexuals. University of Ghana http://ugspace.ug.edu.gh 31 Therefore, applying the theory to homosexuality in particular, it means that believing that homosexuality has biological origins would suggest that sexual orientation is not controllable. Conversely, believing that it is a personal choice, acquired, or learned, suggests that homosexuality can be controlled. Such attributions influence attitudes toward homosexuals with those believing it is biological in origin having positive attitudes and those believing it is an individual choice having negative attitudes. Criticisms and Relevance of the Theories The theories discussed above are central to the aims of this study in that they deal directly with what is basically of interest for the study which is the assessment of the structure of prejudicial attitudes towards homosexuals and how these are moderated by religious commitment and morality. Naturally, these theories are not without their criticism. Smith and Nosek (2011), for example, stated that the measurement of attitudes have been conceptualized in different forms over the years. Although Thurstone (1887 – 1955) asserted that attitudes cannot be wholly described by any single numerical index, years after, they were actually simplified into singular summary evaluations. Added to this has been the differentiation between affective and cognitive components of attitudes (e.g., Eagly & Chaiken, 1993) and also the distinguishing between implicit and explicit components of attitudes, not to mention the multicomponent model of attitudes discussed above. All these go to show how complex the attitude construct is. As has been established earlier, Ghanaians seem to have negative attitudes towards homosexuals and so it is gaining an understanding of the structure of these attitudes and identifying other factors that can influence them that were of importance to this study. It is for this reason that the multicomponent model of attitudes was deemed the right theory base for this study University of Ghana http://ugspace.ug.edu.gh 32 since it gives an explanation of this structure, describes its three components, and how each of them can be assessed so as to understand the nature of attitudes toward a particular object, in this case, toward homosexuals. The social learning theory was criticized by Akers, Krohn, Lanza-Kaduce, and Radosevich (1979) as lacking empirical testing in applied natural settings. Akers et al. therefore decided to test the theory by conducting a study on social learning and adolescent drinking and drug behaviour. Based on their findings, the theory was supported. This fact was subsequently corroborated by other studies (e.g. Akers, 1998; Skinner & Fream, 1997), leaving no doubt that the theory can be tested in natural settings. Since this study aimed to determine whether religious commitment and morality moderate attitudes toward homosexuals, the two variants of the social learning theory noted previously were seen as a right theory base for this study because religious commitment and morality can be learnt by individuals through observing and imitating others, especially significant others such as parents and other relatives. As such, although the theory does not inform the design of this study directly, it was necessary to adopt it as a framework for this study and incorporate it into the explanation of its findings. According to Manusov and Spitzberg (2008), there is increasing evidence that attributional thought processes may be moderated by culture to some extent but it is rare to find a scholar of merit claiming that the theory, including all its variants, is fundamentally flawed. As discussed previously, attributions are a central part of how we form attitudes and what the valence of that attitude would be. So, when it comes to attitudes toward stigmatized behaviours such as homosexuality, this theory points out how believing that a behaviour is controllable produces negative attitudes towards the target person while believing that it is uncontrollable produces positive attitudes towards the target person. With this understanding and based on Manusov and Spitzberg (2008) assertion, the University of Ghana http://ugspace.ug.edu.gh 33 attributional theory of controllability was also deemed right as a theory base that informed aspects of the methodology of this study and helped with explaining the findings. Review of Related Studies Demographic variables and attitudes toward homosexuals. As the following review of related studies will show, research abounds on attitudes toward homosexuality as a sexual orientation and on attitudes toward homosexuals themselves, especially in the western world. A lot of these studies focus on determining the factors related to these attitudes. These factors include sexism, contact with homosexuals and other demographic variables including religion, religious affiliation, religious attendance, educational level, gender, and ethnicity, among others. Herek and Capitanio (1996) for example, conducted a two-wave national AIDS telephone survey a year apart in which their probability sample of English-speaking adults stated their attitudes toward gay men in the first wave and toward gay men and lesbians in the second wave. Their findings at Wave 1 showed that heterosexuals who reported interpersonal contact with gays had more positive attitudes toward gay men than those without contact. This positive attitude was depicted in the reports of more relationships, closer relationships, and receiving direct disclosure about another‘s homosexuality. Findings at Wave 2 indicated a general replication of the previous findings for both gay men and lesbians with cross-wave analyses suggesting a reciprocal relationship between contact and attitudes. Similarly, a study by Sakalh and Ugurlu (2002) which investigated the effects of social interaction with homosexuals on attitudes toward homosexuality among heterosexual Turkish university students found that, generally, exposure to and contact with homosexuals resulted in more positive attitudes toward homosexuality. University of Ghana http://ugspace.ug.edu.gh 34 Newman (2007) examined attitudes toward lesbians by comparing data from a survey conducted in 1985 among students living in a college residence hall with data from a survey conducted in 2001 among students living in the same hall. She found that the sample of college students in 2001 showed more positive attitudes about lesbians than the 1985 sample, suggesting a trend of increasing acceptance of lesbians over the 16-year period. She also found that the 2001 sample reported more contact with lesbians and assessed their parents‘ attitudes as more positive than the 1985 sample. Other studies have also made use of college students as participants, targeting the general population of students or students in specific disciplines. For example, Schulte and Battle (2004) had as their study sample college students from several campuses across the United States. Their focus was to examine whether ethnic differences in attitudes toward gays and lesbians are a function of religious attendance. Comparing African Americans and European Americans, they found that ethnic differences were present in the absence of religious attitudes but these differences were absent in the presence of religious attitudes when predicting attitudes toward lesbians in particular. For attitudes toward gay men, ethnic differences did not exist but religious attitudes were always statistically significant. They therefore concluded that attitudes toward lesbians and gay men are not necessarily a function of ethnicity but possibly a function of religious attendance in general and of the ―Black Church‖ specifically. Another study by Sakalli (2002) examined the relationship between sexism and attitudes toward homosexuality among Turkish college students. It was found that more negative attitudes toward homosexuality was predicted by sexist attitudes, not knowing any homosexuals, and being more traditional and conservative. Hostile sexism more than benevolent sexism was highly correlated with anti-homosexual attitudes. Also, male respondents proved to be more sexist and to hold more negative attitudes toward University of Ghana http://ugspace.ug.edu.gh 35 homosexuality compared to their female counterparts. Although it did not use university students as participants, a study by Ben-Ari (2001) set in the university environment assessed the views of 235 academics in the helping professions (social work, psychology and education) about homosexuals. These academics were faculty members from the five main universities in Israel and the instrument used was the Index of Homophobia (IHP). Results showed that all the participants were low-grade homophobic. Males were found to be more homophobic than their female counterparts except for those in psychology but even for them, secular faculty members were found to be significantly more homophobic than their religious counterparts. Lim and Johnson (2001) assessed the levels of homophobia among social work students from two major universities in the Republic of South Korea. Their results indicated that compared to U.S. samples, the students had high levels of homophobia but that in-class discussion of homosexuality was found to be significantly associated with lower levels of homophobia. Additionally, in a study by Lim (2008) to investigate individuals‘ attitudes toward homosexuals in Singapore and to determine whether gender differences exist in these attitudes, she found that the respondents generally had negative attitudes towards homosexuals. Majority of them also reported that they would feel upset if their child, brother, or sister was homosexual. With regards to the gender differences, generally women reported that they were more comfortable in working closely with male homosexuals while men reported that they were more comfortable in working closely with female homosexuals. Both genders were however in agreement on the fact that they would feel uncomfortable if they discovered that their doctor was homosexual. Religiousity, religious commitment, and attitudes toward homosexuals. On the subject of religion and how it relates to the issue of attitudes toward homosexuality and homosexuals themselves, a study by Whitley (2009) gives a good insight. The study used University of Ghana http://ugspace.ug.edu.gh 36 meta-analysis to examine the relationships between attitudes toward lesbians/gay men and seven forms of religiousity including fundamentalism, frequency of religious service attendance, endorsement of Christian orthodoxy, self-ratings of religiousity, and intrinsic, extrinsic, and quest orientations. He found that with the exception of quest and extrinsic orientation, all the other forms of religiousity had at least small negative relationships with attitudes toward lesbians and gay men. While the extrinsic orientation had no relationship to these attitudes at all, higher quest orientation showed a positive relationship with attitudes toward lesbians and gay men. Also, a study by Arndt and de Bruin (2006) on how gender, race, and religion are related to attitudes toward lesbians and gay men had undergraduate students attending a university in Gauteng as its participants. Findings indicated that both male and female students were relatively more accepting of lesbians than gay men and that a higher level of negative attitudes exists when students are deeply religious, especially with respect to gay males. According to them, race did not have a statistically significant effect on attitudes towards lesbians and gay men. Mohr and Sedlacek (2000) on their part conducted a study aimed at gathering descriptive data regarding college students‘ perceived barriers to friendship with lesbian and gay (LG) individuals. They were interested in three potential barriers including disapproval of LG sexual orientations, perceived lack of commonalities with lesbians and gay men, and discomfort with the thought of befriending a lesbian or gay man. Mohr and Sedlacek‘s findings indicated that gender, diversity orientation, shyness, and religious commitment were all related to perceived barriers. As such, female participants were less likely to perceive barriers and to report not wanting to have LG friends compared to their male counterparts and participants who placed less value on racial/religious diversity were those who did not want an LG friend. Social discomfort was also a barrier to LG friendships for participants who were most shy. Additionally, the most religious University of Ghana http://ugspace.ug.edu.gh 37 participants were those for whom disapproval of LG sexual orientations was a barrier to friendships with lesbians and gay men. Attitudes towards homosexuals inferred from secondary data. Some studies have also provided very good insight into the issue of attitudes towards homosexuality and attitudes toward homosexuals using secondary data with a more nationally representative sample. For example, Ohlander, Batalova, and Treas‘ (2005) study looked at the established association between schooling and support for same-sex relations using General Social Survey (GSS) data (1988-1994) and considering three alternative explanations for this association. Their findings indicate that the association between education and attitudes is not a spurious one but that it is due partly to education‘s association with support for civil liberties and partly to schooling‘s correlation with cognitive sophistication. Loftus (2001) also used GSS data (1973-1998) to examine the changing American attitudes toward homosexuality. She found that, from 1973 to 1976, Americans‘ attitudes regarding the morality of homosexuality became slightly more liberal but became increasingly conservative through to 1990 and since then have become more liberal. Over the same period, the willingness of Americans to restrict the civil liberties of homosexuals declined steadily except for a brief increase in negative attitudes in the late 1980s. According to Loftus, about one-half of the change in attitudes toward homosexuality among Americans can be linked to changes in the demographics of the population, specifically increasing levels of education, and changing cultural ideological beliefs. Components of attitude, attributions, and homosexual experiences. Haddock et al. (1993 conducted two studies with the aim of assessing the relative importance of cognitive and affective information in predicting attitudes toward homosexuals and assessing the extent to which right-wing authoritarianism (RWA) moderates the relative University of Ghana http://ugspace.ug.edu.gh 38 importance of cognition and affect in predicting such attitudes. The first study focused mainly on determining the relative importance of stereotypic beliefs, symbolic beliefs, and affect in predicting attitudes toward homosexuals. One hundred and forty-five (145) introductory psychology students of the University of Waterloo served as participants in groups of 3 or 4 at a time and until they were fully debriefed as to the goals of the study at the end of it, they believed that the study was meant to measure group evaluations. The groups they evaluated included homosexuals, English Canadians, French Canadians, Native Indians, and Pakistanis. Findings indicated that attitudes toward homosexuals were negative and that the most negative attitudes were expressed by high authoritarians. Though not high enough to suggest that they were eliciting completely overlapping information, positive stereotypes were found to be correlated with positive symbolic beliefs and positive affective associates. All three variables were also positively and significantly correlated with participants‘ attitudes towards homosexuals. Additionally, a content analysis of participants‘ open-ended responses showed that with regards to stereotypic beliefs, the most frequently elicited responses were that homosexuals are effeminate, friendly, and normal. For symbolic beliefs, the responses were that homosexuals promote freedom, block the attainment of the traditional family, and promote peace while with regards to the emotions most frequently elicited by homosexuals, feelings of disgust, discomfort, and confusion topped the list. For the second study, the focus was on finding out if results from the first study will be replicated and determining how past behaviours relate to intergroup attitudes, how authoritarianism relates to contact with homosexuals and how interpretations of the term homosexual would affect perceptions. The procedure used in this second study was similar to that followed in the first but there were some exceptions, the most significant of University of Ghana http://ugspace.ug.edu.gh 39 them probably being the fact that homosexuals were the only target group assessed in this study. Generally, findings from the second study were similar to that of the first study. Haddock et al.‘s (1993) study is undoubtedly a classic one in that it does not make use of attitude scales like most studies on attitudes toward homosexuality or homosexuals but looks at people‘s overall evaluation of other group members and also at the various components of attitudes. Another study that also tried to analyze the components of attitudes toward homosexuals to some extent is one by Negy and Eiseman (2005). They compared African American (70 persons) and White (non-Hispanic, 143 persons) College students‘ affective and attitudinal reactions to LGB individuals. These students, who were freshmen or sophomores, were all from general psychology classes and volunteered to participate in the study. The findings initially suggested that African Americans might have slightly higher negative affective and attitudinal reactions to LGBs than Whites. Upon further analyses however, it was found that when frequency of church attendance, religious commitment, and socio-economic status (SES) are taken into consideration, these differences diminished. In addition, gender and religiousity variables were predictive of homophobia. With regards to gender, both African American and White men showed significantly higher homophobia and homonegativity than their female counterparts. For the religiousity variables, with African Americans it was found that the higher their frequency of church attendance, the more negative their attitudes toward LGBs were and with the Whites, it was found that the higher their religious commitment, the more negative their attitudes were. Also, the more socialized and immersed into their culture and community African Americans were, the higher their homophobia and homonegativity. Negy and Eiseman‘s (2005) study is different from others in that they did University of Ghana http://ugspace.ug.edu.gh 40 not state any formal hypotheses, their reason being the contradictory nature of empirical literature on African Americans‘ and Whites‘ reactions to LGBs. Attributions. As noted earlier, attitudes toward homosexuals can be determined by attributions and this is evidenced by Haider-Markel and Joslyn‘s (2008) study on beliefs about the origins of homosexuality and support for gay rights. They used data from a Pew Research Center for the People and Press survey (October 15 – 19, 2003) and from a Gallup national survey (May, 2006) of adults and employed the use of Weiner‘s (1979, 1985) attribution theory of controllability to examine people‘s beliefs about the origins of homosexuality. From the Pew data, they found that out of those who gave responses, about 47% believed that homosexuality was a controllable state and from their analyses of the Gallup data, they found that the results were highly consistent with results from their analysis of the Pew data. The perception of homosexuality as controllable increased negative affect while the perception of homosexuality as uncontrollable increased positive affect towards gays and lesbians. Haider-Markel and Joslyn (2008) therefore concluded that attributions are powerful predictors of people‘s affective reactions toward gays and various gay civil rights policies. They however acknowledge that their data do not prove that respondents believe homosexuality can be controlled but that they inferred this from theoretical expectations and prior research. As such, their study basically only demonstrated the plausibility and parsimony of Weiner‘s attribution theory. Furthermore, they used secondary data instead of conducting their own research to better manipulate variables of interest to them, and thereby gain more insight into people‘s beliefs about the etiology of homosexuality and how these relate to their feelings toward homosexuals and support for gay rights. Boysen and Vogel (2007) on their part looked at the attribution theory in relation to attitudes toward homosexuality and went further to examine how biased University of Ghana http://ugspace.ug.edu.gh 41 assimilation and attitude polarization can affect learning about the biological explanations of homosexuality and therefore affect attitudes. Their study sample consisted of 210 participants who were undergraduates at a large Midwestern university in the United States and volunteered to be part of the study. Boysen and Vogel (2007) first made their participants to complete some demographic questions and an initial measure of their attitudes and later completed measures of biased assimilation and attitude polarization. In-between these, they read a selection from an introduction to psychology textbook that discussed the evidence for biological influences on homosexual behaviour (i.e. brain anatomy, genetics, and parental hormones) and were tested on it. Their findings indicated that participants with initial positive attitudes toward homosexuality tended to see biological evidence as reason for acceptance of homosexuality while participants with initial negative attitudes tended to see it as reason for no acceptance. With regards to biased assimilation, it was found that participants with initial positive attitudes perceived biological explanations as more persuasive evidence for the legitimacy and acceptability of homosexuality while those with initial negative attitudes perceived these explanations as more persuasive evidence against the legitimacy and acceptability of homosexuality. Also, both groups of participants reported more certainty in their original attitudes after studying the reading material. Self-identification and disclosure. Although this current study is mainly on attitudes toward homosexuals, its third part, Study 3, also looked at the experiences of both male and female homosexual persons and what they perceive to be the attitudes of the Ghanaian general public towards them. Other studies have done the same in the past, a good example being Maguen, Floyd, Bakerman, and Armistead‘s (2002) study. Maguen et al. investigated coming-out milestones, disclosure and self-esteem among gay, lesbian, University of Ghana http://ugspace.ug.edu.gh 42 and bisexual youths from the southeastern United States. One hundred and seventeen (117) respondents completed questions relating to their demographics and sexual orientation and they also completed the 10-item Rosenberg (1965) self-esteem scale. With regards to persons to whom respondents were likely to disclose their sexual identity, Maguen et al.‘s findings indicated that friends were the most cited (95%), followed by mothers (84%), siblings (79%), and teachers (72%). Respondents were less likely to have disclosed to their fathers (67%) and the least likely to have disclosed to a doctor or other relative. However, it was those respondents who were out to their fathers who showed higher self-esteem scores. Crowley (2010) conducted an exploratory study examining topic threads in Lesbian/Bisexual Groups on MySpace. Her study addressed two questions, the first being how young biwomen self-identify in online Topic discussions with other women in the Gay, Lesbian, and Bi (GLB) Group category on MySpace and the second being what attitudes and beliefs about bisexual identities emerge from online discussions involving young biwomen in those GLB Groups. Her analyses showed that the most relevant discussions emerged from 3 Groups. In one of these groups known as ―GIRLS WHO KISS GIRLS‖, a discussion was generated on the question of labeling and involved 80 responses from women aged between 18 and 40. Crowley found that 12 of these women preferred no label and they had varied reasons for choosing not to label themselves. For some it was because though they experienced attractions to women, they did not want to exclude future relationships with men and for others it was because they felt that people should not limit themselves by labeling themselves just for the sake of being identified. Some of the young women also avoided labels because they felt that the term ‗bisexual‘ carried with it too much negative baggage. University of Ghana http://ugspace.ug.edu.gh 43 Related literature specific to Ghana. In Ghana, literature on homosexuality, especially relating to this current study, does not abound but the few that exist give interesting insights that are worth reviewing. For example, a study conducted by Sabin, Adu-Sakordie, Agyarko-Poku, Ashigbie, Hollenbeck-Pringle, DeSilva et al. (2013) on attitudes and behaviours among older MSMs in Ghana used men aged 30 years and above. The main goal of the study was to collect and analyze data to improve the outreach and effectiveness of local programmes targeted at providing important HIV prevention and treatment information to older MSM. Sabin et al. collected data from a total number of 44 participants who were in two groups, those aged between 30 and 39 and those aged 40 and above (22 in each group), through In-depth Interviews (IDIs) and Focus Group Discussions (FGDs). Among other things, their findings showed that 39% of participants were married, 40% of them were single, and 9 of the participants (20%) were cohabiting with someone. According to Sabin et al., one of the IDIs participants in the 30-39 age group reported that though he is gay and therefore had no feeling for women, he had taken a girlfriend to cover up his deeds. Perhaps a more related study is one by Owusu, Anarfi, and Tenkorang (2013) on attitudes and views on same-sex sexual behaviour in Ghana. Their aim was to answer three main research questions including whether Ghanaians are more open about and discuss their sexual behaviours, what the patterns of sexual behaviours are among Ghanaians and what sexual behaviours are considered mainstream, which are not, and what the implications are. To this end, they organized 81 qualitative interviews which entailed 33 individual IDIs, 24 FGDs, and 24 individual lifeline stories that allowed respondents to talk about their individual experiences over a lifetime. Respondents were aged between 10 and 80 years and were selected such that all socio-economic and demographic groups (in-and-out of school youth, people with disabilities, the aged, University of Ghana http://ugspace.ug.edu.gh 44 community opinion leaders, commercial sex workers, and a woman who has sex with a woman [WSW]) were covered. Although they aimed to interview MSMs and persons who have sex with both males and females (PSWBMFs), they were unable to do so as no one self-identified as such. In discussing their findings, Owusu et al. (2013) stated that their key finding is the hostility and prejudice of respondents directed toward persons who express non- heteronormative sexual behaviours. They emphasize the fact that all their respondents did not approve of such sexual behaviours and expressed very negative sentiments about it. The most prevalent view among the respondents was that MSM, WSW, and PSWBMFs‘ behaviours constitute a ―spiritual‖ problem as the persons involved are demonized and are socially unacceptable because they bring curses and bad omen to the society. They also assert that the high level of anti-MSM and WSW sentiments coupled with the fact that such non-conventional sexual preferences have not yet assumed human rights status here in Ghana, can lead to the abuse of persons with such preferences. This conclusion was drawn based on some of the respondents‘ comments as to what they would do to a hypothetical MSM and WSW, including threats to cane them or evict them from their houses. They further highlight the fact that respondents expressed differing attitudes toward the hypothetical MSM and WSW if they were their relatives or their spouse/romantic heterosexual partner, showing they would be harsher towards the latter than their relatives. One of the limitations of Owusu et al.‘s (2013) study, as acknowledged by the authors, is their inability to recruit and interview individuals who engage in same-sex sexual relations and PSWBMFs as part of their study. Another study that can also be considered as more related to this study is by Anarfi and Gyasi-Gyamerah (2014). Their main aim was to determine Ghanaian university students‘ level of religiousity and how it University of Ghana http://ugspace.ug.edu.gh 45 affects their attitudes toward homosexuality. The study was done in the three oldest public universities in Ghana, which are the University of Ghana (UG), the Kwame Nkrumah University of Science and Technology (KNUST), and the University of Cape Coast (UCC). The data collection was in two parts, self-completed questionnaires and Focus Group Discussions (FGDs). As Participants were selected separately for each part of the study, the first part had a total number of 1,258 as respondents. For the second part, three FGDs were conducted in each university, including one with an all-female group, an all- male group, and a mixed group (equal number of males and females). Each group consisted of at least eight participants from the general student population with all levels represented. According to Anarfi and Gyasi-Gyamerah (2014), students‘ level of religiosity and their religious affiliation significantly affected their attitudes toward homosexuals. Highly religious respondents had more negative attitudes toward homosexuality than moderately religious and non-religious respondents and they also showed negative attitudes toward homosexuality regardless of their religious affiliation. Respondents who were Pentecostal Christians had the most negative attitudes. With regard to gender, no difference was found between males and females in terms of their attitudes toward homosexuals and the attitudes of both genders were generally negative. The only gender difference noted was found in the analyses of the FGDs and pointed to the fact that female participants would be more comfortable interacting with male homosexuals than female homosexuals while male participants stated that they would be more comfortable interacting with female homosexuals than male homosexuals. The reason for this was that participants felt there could be some sexual attraction to them on the part of the homosexuals which they would prefer to avoid. University of Ghana http://ugspace.ug.edu.gh 46 One of the limitations of the Anarfi and Gyasi-Gyamerah (2014) study is the fact that for the FGDs, some of the students might have felt pressured by the presence of others within the various groups to provide responses that did not reflect their true personal feelings towards the issue of homosexuality. They also alluded to the possibility that this same pressure to conform might have prevented any of the participants who is homosexual or has had homosexual experiences in the past, from making mention of it, thereby creating an impression that all the FGDs participants were heterosexual. Summary From the above discussion, it is evident that research indeed abounds in the area of attitudes toward homosexuality and its related issues. Attitudes toward homosexuals are generally negative but especially so in parts of the world other than the West. However contact in the form of social interaction with homosexuals generally result in more positive attitudes toward homosexuals. Certain demographic variables including religion, religious affiliation, religious attendance, educational level, gender, and having children or not, do affect the valence of attitudes. For example, males generally have more negative attitudes toward homosexuals than females, one of the reasons being that males are sexist. There are however instances where this is not the case as seen from Ben-Ari‘s (2001) study where for academics in the discipline of psychology, no gender differences were found. The study also showed that though there were no differences in gender for academics in psychology, those of them who were religious had significantly less homophobia than their secular colleagues, which is quite unusual. Not only is educational level a determinant of attitudes toward homosexuals but whether one has learnt anything about homosexuality while in school also affects attitudes. As the literature suggests, increasing levels of education is one of the factors informing the University of Ghana http://ugspace.ug.edu.gh 47 liberalization of Americans‘ attitudes toward homosexuality. Generally, there is a trend of increasing acceptance of LGB individuals in the American society. This does not seem to be the case in Ghana as studies from Ghana show that Ghanaians have very negative sentiments about homosexuals and even believe that it is a spiritual problem. Such beliefs can cause them to treat LGB individuals harshly especially where that individual is not a relative. Whether homosexuality is controllable or not and how this affects attitudes is the focus of some of the studies reviewed here and as is obvious from their results, believing it is controllable is linked with negative attitudes. It is however clear that people‘s extreme preexisting attitudes toward homosexuality produces more biased assimilation and attitude polarization. An interesting aspect of the review is that many of the studies used university (college) students as participants, especially psychology students but only one analyzed the components of attitudes toward homosexuals. From this analysis it was found that the cognitive, affective, and behavioural components of attitudes are all relevant to the understanding of attitudes toward homosexuals. Evidently, there are also a lot of studies that deal with personal experiences of LGB individuals. From the review, it is obvious that LGB individuals themselves are aware of the fact that people inclined towards religion usually have negative attitudes toward them and they acknowledge that they personally experience conflicts between their own quest for religion and their sexual orientation. Such conflict causes them to hide their sexual orientation and not come out or do so when they are older although they invariably realize their orientation during their early teens. This does not however mean that LGB individuals are generally unreligious. As seen above, when LGB individuals decide to disclose their orientation, the literature shows that they would most likely do so to a friend or their mothers but are least likely to do so to their fathers. Interestingly however, those who disclose to their fathers University of Ghana http://ugspace.ug.edu.gh 48 are usually those with higher self-esteem. A lot of the disclosure is done online in chat groups where LGB individuals are likely to meet other sexual minority individuals. In such online discussions, some women have shown their preference not to label their sexuality but allow the circumstances and their emotional attachment to the person they are attracted to at a particular time, whether male or female, to determine which side of the sexual orientation continuum they would place themselves. As the review above shows, one of the reasons for young women not labeling their sexuality is that they want to keep their future options open and not exclude future relationships with men. Conceptual Model Considering the above discussion of the theories underpinning this study and the review of related literature, attitudes toward homosexuals can be thought of as a whole concept that has underlying structures (i.e. the stereotypic beliefs, symbolic beliefs, and affect). These structures can affect the nature of attitudes in that they can determine whether the attitudes will be positive or negative. These structures in turn can be affected by other factors including personal characteristics (i.e. gender, age, academic level, and religious affiliation), religious commitment, and morality (moral values and personal morals) such that their combined effect can have the ability to determine the nature of attitudes toward homosexuals. With this understanding, the following conceptual model (Figure 1) was developed. University of Ghana http://ugspace.ug.edu.gh 49 Figure 1. Conceptual model showing the expected outcome of this study (Author‘s Construct). Stereotypic Beliefs Symbolic Beliefs Affect Attitudes toward Homosexuals (Evaluation of Homosexuals) Moral Values Personal Morals Religious Commitment Personal Characteristics University of Ghana http://ugspace.ug.edu.gh 50 As evident in Figure 1, the variables of interest to this study are shown in rectangles of various sizes that are solely meant to accommodate the variable names. The interaction effects between variables are depicted by arrows with no tips, the combined effect of the interactions is depicted by a solid tipped arrow, and direct effects are depicted by normal double-tipped arrows. The model demonstrates that this study was conducted with the expectation that stereotypic beliefs, symbolic beliefs, affect, morality and religious commitment would have direct effects on attitudes toward homosexuals with personal characteristics controlled. There was also the expectation that this study would show that moral values, personal morals and religious commitment would have moderating effects on the effects of affect, stereotypic beliefs, and symbolic beliefs on attitudes toward homosexuals. This conceptual model was tested to determine whether it would be supported or not. Hypotheses and Research Questions Based on the review of literature and the objectives of the study, the following are the hypotheses tested under each of the three studies within this study: Study one. For the first part of this study, the hypotheses tested include the following: 1. Respondents will evaluate homosexuals more unfavourably than the other target groups (see operational definitions). 2. Respondents‘ personal characteristics (age, gender, academic level, religion) will influence their evaluation of homosexuals. 3. There will be a positive correlation among stereotypic beliefs, symbolic beliefs, and affect. University of Ghana http://ugspace.ug.edu.gh 51 4. Stereotypic beliefs, symbolic beliefs, and affect will predict respondents‘ evaluation of homosexuals. 5. Respondents‘ religious commitment, moral values, and personal morals will moderate the relationship between their attitude components and their evaluation of homosexuals. Study two. The hypotheses tested under Study 2 include the following: 1. Participants will evaluate homosexuals more unfavourably than the other target groups (see operational definitions). 2. There will be a positive correlation between stereotypic beliefs, symbolic beliefs, and affect. 3. Stereotypic beliefs, symbolic beliefs, and affect will predict respondents‘ evaluation of homosexuals. 4. Participants‘ religious commitment, moral values, and personal morals, will moderate the relationship between their attitude components and their evaluation of homosexuals. 5. Participants will evaluate homosexuals more favourably in the post-test condition than in the pre-test condition. 6. Males will evaluate homosexuals more unfavourably than their female counterparts in both pre-test and post-test conditions. 7. Participants in the biological and biological transgendered groups (see Chapter Four, Table 11) will have more favourable behavioural intentions toward the vignette persons than participants in the choice, choice transgendered, and teenage pregnancy groups (see Chapter Four, Table 11). Study three. This study, which is purely qualitative, is also in two parts (see methodology, Chapter 5): University of Ghana http://ugspace.ug.edu.gh 52 Focus group discussions. The following are the research questions that data gathered from the Focus Group Discussions (FGDs) helped in answering: 1. How do participants define religiousity and religious commitment? 2. What do participants perceive to be the level of religiousity and religious commitment among Ghanaians in general and university students in particular? 3. What do participants perceive their own individual levels of religiousity and religious commitment to be? 4. How do participants define morality and what is their perception of what it means to Ghanaians and especially to university students? 5. What do participants perceive to be the extent to which Ghanaians are willing to engage in acts that can be described as morally wrong (e.g. extramarital affairs, bribery, juju, cheating on tax) and to what extent are they likely to engage in these acts themselves as individuals? 6. How do participants view homosexuality and how do they feel when they see, meet or think about homosexuals? 7. To what extent are attributions likely to affect participant‘s definition of homosexuality and their attitudes towards homosexuals? In-depth interviews. The following are the research questions that data gathered from the In-depth Interviews (IDIs) helped in answering: 1. What informs one‘s definition of him/herself as homosexual and what social challenges come with self-identifying as such? 2. How do homosexuals perceive how Ghanaians view homosexuality and how do they feel about it? University of Ghana http://ugspace.ug.edu.gh 53 Operational Definition of Terms For the purposes of this study, the following terms are used such that: 1. Homosexual – refers to both male homosexuals (gays) and female homosexuals (lesbians). 2. Target groups – refers to the attitude objects which are in 5 groups: Orthodox Church Priests, Commercial Sex Workers, Spiritual Church Leaders, Homosexuals, and Traditional Priests. 3. Vignette person – refers to the persons described within the 5 vignettes. 4. Behavioural intentions – refers to how a participant is likely to behave towards a vignette person if he or she were to be acquainted with the person. 5. Presumed heterosexuals – refers to the students who participated in this study and who can only be assumed to be heterosexuals because they were not asked for their sexual orientations. 6. Known homosexuals – refers to individuals who self-identified as homosexual when asked for their sexual orientations. University of Ghana http://ugspace.ug.edu.gh 54 Chapter Three Study One Introduction Study 1 is the first of three studies conducted. All three studies had the University of Ghana (UG) as the main setting. One of the reasons for choosing UG is because it is the oldest and largest of the public universities in Ghana and has a population of about 29,754 with a wide variety of students from across the country, the sub-region and overseas. UG is therefore virtually a microcosm of Ghana. Additionally, UG students get a choice of on-campus accommodation or private hostels around the university campus. This fact provides the second reason for its choice as the setting for this research which is that it offers the advantage of proximity and accessibility. This chapter documents the methods used to collect data for Study One and the findings made. These findings are then discussed in relation to the relevant literature reviewed in Chapter Two and by incorporating other relevant literature to aid explanation of the findings. Five hypotheses were tested using the Statistical Package for the Social Sciences (SPSS, version 16). One hypothesis was confirmed, two were partially confirmed and the remaining two were not confirmed. There were still some unanswered questions upon discussion of the findings. Methodology Population/Sample. The population for this study was undergraduate psychology students (Levels 100 – 400). Two hundred (200) students were selected through convenient sampling from this population. However, only one hundred and ninety (190) students participated in the study. The mean age of respondents was 21.75 years (SD = University of Ghana http://ugspace.ug.edu.gh 55 3.80) and more than half of them were males. Details of their demographics are shown in Table 1. Table 1 Demographic Characteristics of Respondents: Study 1 Instruments. The main instrument used for this study was the Group Evaluation Questionnaire (GEQ) which has 4 sections (see Appendix A). The questionnaire was so named because it was necessary to use deception initially with regards to the true purpose of the study. Section A of the GEQ dealt with the personal characteristics of respondents (gender, age, academic level, religion). The responses on variables in this section were pre-coded except for the age of respondents. The ages provided by respondents were later Variables Frequency (n = 190) Percent (%) Gender Male 104 54.7 Female 86 45.3 Age M = 21.75, SD = 3.80 18 – 25 years 173 91.1 26 – 53 years 17 8.9 Academic Level Level 100 39 20.5 Level 200 68 35.8 Level 300 7 3.7 Level 400 76 40.0 Religion No Religion 1 0.5 Christianity 180 94.7 Islam 7 3.7 Other 2 1.1 University of Ghana http://ugspace.ug.edu.gh 56 coded as 1 for those aged 18 – 25 (younger respondents) and 2 for those aged 26 and above (older respondents). Section B consisted of the Religious Commitment Inventory-10 (RCI-10) developed by Worthington et al. (2003) and an adaptation of the Moral Behaviour Scale (MBS) developed by Rettig and Pasamanick (1959). The RCI-10 is a 10-item inventory with two subscales, the Intrapersonal Religious Commitment subscale (Items 1, 3, 4, 5, 7, & 8) and the Interpersonal Religious Commitment subscale (items 2, 6, 9, & 10). Some of the items on the intrapersonal subscale include ―Religion is especially important to me because it answers many questions about the meaning of life‖ and ―Religious beliefs influence all my dealings in life‖. For the interpersonal subscale, some of the items include ―I enjoy spending time with others of my religious affiliation‖ and ―I enjoy working in the activities of my religious affiliation‖. Its coefficient alphas for internal consistency are .93 for the full scale and .92 and .87 for the intrapersonal and interpersonal subscales respectively. It also has a 3-week test-retest reliability coefficient of .87 for the full scale, .86 and .83 for the intrapersonal and interpersonal subscales respectively. Worthington et al. (2003) noted that the RCI-10 sub-scales are very highly intercorrelated and therefore advocated that in clinical practice and in research, total scores on the full RCI-10 should be used. Scores on the RCI-10 ranged from 1 (Not at all true of me) to 5 (Totally true of me) with total scores for the full scale ranging from 10 to 50. According to Worthington et al., for secular groups, means for the full-scale RCI-10 are between 21 and 26 (SDs = 10–12) so people scoring greater than one standard deviation higher than the mean should be considered highly religious. This study therefore used the full scale scores and Cronbach‘s alpha obtained with the local sample was .86 (N = 12). Respondents who scored from 10 to 35 were considered as having low religious University of Ghana http://ugspace.ug.edu.gh 57 commitment and those who scored from 36 to 50 were considered as having high religious commitment. The Moral Behaviour Scale (MBS) was originally a 50-item scale that presents acts or situations which respondents are to evaluate in terms of ‗rightness‘ or ‗wrongness‘ on a 10-point Likert scale. To ensure that only the acts or situations that are most relevant to the purposes of this study are maintained, 36 items were taken out completely leaving 14 items (i.e. items 1, 2, 3, 4, 6, 10, 16, 19, 20, 26, 27, 28, 33, and 37). Seven new items were then added up to the 14 to make 21 items on the current scale. Some of the 14 items maintained were slightly modified to suit the demands of the study and for better understanding. For example, item 2 on the original scale was modified from ‗Kidnapping and holding a child for ransom‘ to ‗Kidnapping and killing a child for juju‘ which is the first item on the current scale. Another example is item 26 which was changed from ‗Keeping over-change given by a clerk in mistake‘ to ‗Keeping over-change given by a shop attendant‘ and is now the 10th item on the current scale. Some of the 7 new items added to the MBS include ‗Having an abortion‘, ‗Avoiding paying your fare on public transport‘, and ‗Engaging in homosexuality‘. These were also added to suit the demands of the study. The second reason why it was necessary to modify the MBS is that the demands of this study required that respondents‘ morality be tested in two ways, firstly in terms of what they thought to be the accepted norms within the Ghanaian society and secondly in terms of what they each personally thought to be right or wrong and therefore would or would not do. As a result, the modified MBS actually became two scales with the same items, the first named the Moral Values Scale (MVS) and the second named the Personal Morals Scale (PMS). The original 50-item MBS had a reliability coefficient of .93 among college students and with the local sample, a Cronbach‘s alpha of .92 and .91 (N = 12) University of Ghana http://ugspace.ug.edu.gh 58 were obtained for the 21-item MVS and PMS respectively. Scores on the MVS ranged from 1 (Least wrong) to 10 (Most wrong). The numbers in-between 1 and 10 were used for in-between degrees of wrongness with total scores for the full scale ranging from 21 to 210 with a mean value of 168. Using the mean value, scores below the mean were classified as low and those above the mean were classified as high. As a result, a score between 21 to 168 denoted low moral values while a score between 169 to 210 denoted high moral values. For the PMS, scores ranged from 1 (Not at all likely) to 5 (Highly likely) with total scores on the full scale ranging from 21 to 105 with a mean value of 84. Responses were reverse scored and therefore scores between 85 to 105 denoted low personal morals while scores between 21 to 84 denoted high personal morals. Section C of the GEQ consisted of the Evaluation Thermometer Measure for Assessing Attitudes toward Gay Men. This was developed by Haddock and Zanna (1998) and has a test-retest reliability of .77 over a 2-week period and a validity of .91 as established by Haddock (1994). It is a single-item attitude measure that requires respondents to provide a number between 0 and 100 degrees to indicate their overall evaluation of gay men and other target groups. For the purposes of this study, respondents were required to use the thermometer to indicate their evaluation of five target groups (see operational definitions, Chapter 2) and to aid this process, a graphical depiction of a thermometer that is colour-coded from dark red (0o) to light pink (100o) was added to Haddock and Zanna‘s original thermometer (see Appendix A). Cronbach‘s alpha obtained with the local sample was .96 (N = 12). Scores on the evaluation thermometer ranged from 0o (extremely unfavourable) to 100o (extremely favourable) with the adjectives very, quite, fairly, and slightly unfavourable/favourable marked at 10o increments in-between these two extremes. The mid-point was 50o (neither favourable nor unfavourable). As is the nature of the measure, no computations were required to create an attitude score so the University of Ghana http://ugspace.ug.edu.gh 59 number that each respondent provided to indicate their overall evaluation of the target groups served as the attitude score. The final section, Section D, dealt with the cognitive and affective components of the attitude concept and therefore asked questions relating to how people think and feel about typical members of the 5 target groups. Following the procedure used by Haddock et al. (1993), the first question of this section which assessed respondents‘ stereotypic beliefs, asked respondents to list as many characteristics that they will use to describe typical members of each of the target groups and then rate each characteristic on the extent to which they think it is a positive or negative characteristic on a 5-point scale. The question finally asked them to list the percentage of members they believe possess each of the characteristics. The second question for this section, which assessed the respondents‘ symbolic beliefs, asked that they state the values, customs, or traditions that are blocked or facilitated by typical members of the target groups and then rate the extent to which each value, custom, or tradition listed is blocked or facilitated on a 5-point scale. It also asked them to state the percentage of group members they believe possess these characteristics. The final question for the section first asked respondents to list the feelings or emotions they experience when they see, meet, or think about typical members of the target groups and rate each feeling or emotion listed based on the extent to which they think it is a positive or negative emotion on a 5-point scale. They then had to state the percentage of group members who evoke each emotion in them. Scores for stereotypic beliefs and affect ranged from -2 (very negative) to +2 (very positive) while scores for symbolic beliefs ranged from -2 (almost always blocked) to +2 (almost always facilitated). An overall stereotype score was calculated by first multiplying the valence of each characteristic by the percentage of group members believed to possess the University of Ghana http://ugspace.ug.edu.gh 60 characteristic and then dividing the product by the characteristics listed. This same procedure was used to calculate the overall scores for symbolic beliefs and affect. Design and Procedure. The design for this study is a survey. In keeping with ethical principles, the research proposal and the informed consent forms used for data collection were sent to the Institutional Review Board (IRB) of the Noguchi Memorial Institute, University of Ghana, Legon, for approval. Approval was granted for the study for a year and again, for a 6-month extension that was sought after the year expired to complete data collection (see Appendix G). Before the data collection, the GEQ was pretested using 12 students (6 males and 6 females, 3 each from Level 100-400) as respondents. The scales within the GEQ were developed in the west and therefore it was necessary to conduct a pretest to see if students would comprehend what each of the scales required of them and therefore be able to answer them appropriately. The pretest also served to help determine the statistical properties of the GEQ in relation to the local sample (see section on Instruments). With this done, the only problem that was noticed was the fact that students found it difficult to comprehend what the question on symbolic attitudes demanded. The demands of the question were therefore explained to the respondents during the actual data collection. The selection of respondents for the actual data collection was done in various Level 100 to 400 psychology classes. First, the Head of the Department of Psychology and the various lecturers for these courses were notified and with their permission, the last 20 minutes of their lecture hours were used to inform students about the study. They were simply told that their participation was needed in a study aimed at measuring group evaluations. Two hundred (200) students, 50 from each level (25 females and 25 males), were then asked to volunteer to participate in the study by a show of hands. These students were told to remain seated while the rest left the lecture hall. At this point, those University of Ghana http://ugspace.ug.edu.gh 61 who had volunteered were each given a card bearing a serial code and were informed about the research schedule so they each knew when and where to report for the data collection. The plan was for them to report at the Department of Psychology‘s lecture room in groups of 20 (10 females and 10 males) at specific times of the set dates for data collection which spanned 5 working days, bringing along their serial code cards. Unfortunately only about 12 to 14 volunteers turned up each day at the appointed times. Efforts were therefore made to make up the numbers by speaking to students who at the time were either present at the department for tutorials or using the library but had not volunteered previously for anybody who was willing to volunteer then. This explains why the total number of respondents came to 190 with unequal number of females and males and respondents from the various levels as noted earlier. For each group of respondents, before they began completing their questionnaires, they were again informed that the study was intended to measure group evaluations. This deception as to the goal of the study was necessary so as to reduce the respondents‘ likelihood of giving positive or negative information about target groups based solely on a preconceived notion about them. It was to ensure that they would genuinely compare the target groups with each other in their evaluation of them. In line with Haddock et al.‘s (1993, p. 1108) procedure and also to ensure that respondents would think of both positive and negative things while considering the target groups, they were additionally told that “almost everyone has positive or negative things to say about most groups” and that their responses will be kept in the strictest of confidence. They were then given their individual consent forms (see Appendix F), two copies each, allowed time to read it and then sign if they would like to continue participation in the study. They afterwards returned one copy and kept the other. None of them declined participation but they were reminded that they still had the option to decline participation University of Ghana http://ugspace.ug.edu.gh 62 at any point during the session as was stipulated in the consent form. The GEQ was then distributed to them to complete and I was available to clarify the specific instructions pertaining to each section and anything that might not be clear to them. Respondents were allowed to have the pens they were given to use in completing the GEQ as a token of appreciation for their participation. There was no need to debrief them due to the certainty that the session would have no adverse effect on them despite the deception. The whole session took between 45 minutes to an hour to complete. Results Hypothesis 1. This hypothesis stated that respondents are likely to evaluate homosexuals more unfavourably than the other target groups. It was tested using repeated measures ANOVA and Table 2 summarizes the results. Table 2 Evaluation of Target Groups: Study 1 Target Groups M (SD) Df F p Orthodox Church Priests 78.19 (18.15) 4/748 454.70* .000 Commercial Sex Workers 14.57 (10.22) Spiritual Church Leaders 70.29 (13.13) Homosexuals 8.88 (7.22) Traditional Priests 28.14 (12.43) *Greenhouse-Geisser; N = 188 As seen from Table 2, significant differences were found in respondents‘ evaluation of the target groups (F(4,748) = 454.70, p = .000). It was however unclear where these differences lay and this necessitated a post hoc analyses of the mean scores. Table 3 summarizes the results of the post hoc analyses. University of Ghana http://ugspace.ug.edu.gh 63 Table 3 Bonferroni Multiple Comparison of Evaluation of Target Groups: Study 1 Target Groups 1 2 3 4 5 1. 1. Orthodox Church Priests __ 2. 2. Commercial Sex Workers 63.62* __ 3. 3. Spiritual Church Leaders 7.90* 55.72* __ 4. 4. Homosexuals 69.31* 5.69* 61.41* __ 5. 5. Traditional Priests 50.05* 13.56* 42.15* 19.26* __ *p < 0.05 As seen from Table 3, significant mean differences were revealed in the evaluation of all target groups (p < 0.05). Drawing on the mean scores shown in Table 2, homosexuals got the most unfavourable evaluation (M = 8.88, SD = 7.22), followed by Commercial Sex Workers (M = 14.57, SD = 10.22), and Traditional Priests (M = 28.14, SD = 12.43). The target group that received the most favourable evaluation was Orthodox Church Priests (M = 78.19, SD = 18.15), followed by Spiritual Church Leaders (M = 70.29, SD = 13.13). Figure 2 provides a pictorial view of the various target groups‘ position on the evaluation thermometer based on the mean scores of the respondents approximated to the levels of favourability or otherwise closest to them on the evaluation thermometer. University of Ghana http://ugspace.ug.edu.gh 64 POSITIVE 100o Extremely Favourable 90o Very Favourable 80o Quite Favourable 70o Fairly Favourable 60o Slightly Favourable 50o Neither Favourable nor Unfavourable 40o Slightly Unfavourable 30o Fairly Unfavourable 20o Quite Unfavourable 10o Very Unfavourable NEGATIVE 0o Extremely Unfavourable Figure 2. Evaluation thermometer showing points corresponding with Study 1 respondents‘ mean evaluation scores for the various target groups. O.C.P. = Orthodox Church Priests; S.C.L. = Spiritual Church Leaders; T.P. = Traditional Priests; CSWs = Commercial Sex Workers; H. = Homosexuals. It is obvious from Figure 2 that generally, Orthodox Church Priests got the evaluation ‗quite favourable‘, Commercial Sex Workers got ‗quite unfavourable‘, Spiritual Church Leaders got ‗fairly favourable‘, Homosexuals got ‗very unfavourable‘, and Traditional Priests got ‗fairly unfavourable‘ from respondents. Thus, as was hypothesized, respondents evaluated homosexuals more unfavourably than the other target groups. Hypothesis 2. This hypothesis stated that the personal characteristics of respondents are likely to influence their evaluation of homosexuals. Using Multiple Regression to test this hypothesis required recoding certain variables into dummy variables. These variables included gender, which was recoded into 0 for males and 1 for O.C.P. S.C.L. T.P. CSWs H. University of Ghana http://ugspace.ug.edu.gh 65 females, and academic level, of which Levels 200 to 400 were each coded 1 with Level 100 serving as the base variable and therefore coded 0, and religion, which had ‗No Religion‘ serving as the base variable coded 0 and Christianity, Islam, and Other religions each coded 1. The results, as summarized in Table 4, show that the respondents‘ personal characteristics had no influence on their evaluation of homosexuals (F[8,180] = 0.79, p = .61). The hypothesis was thus not supported. Table 4 Summary of Regression Analyses Examining the Influence of Respondents‟ Personal Characteristics on Evaluation of Homosexuals Variables Evaluation of Homosexuals Β R2 ΔR2 Constant 2.08 -.01 .03 Age -0.13 Gender -2.08 Level 400 5.37 Level 300 -1.02 Level 200 0.58 Christianity 8.33 Islam 2.04 Other 15.65 Note. N = 189. Hypothesis 3. The third hypothesis stated that there will be a positive correlation between stereotypic beliefs, symbolic beliefs, and affect. The summary of results of the Pearson product moment correlation performed to test this hypothesis is shown in Table 5 (see the following page). As can be seen from the table, respondents‘ stereotypic beliefs were significantly positively correlated with their symbolic beliefs (r[120] = .37, p = .00) and their affect (r[176] = .16, p = .02). This means that positive stereotypic beliefs are associated with positive symbolic beliefs, and positive affect while at the same time, University of Ghana http://ugspace.ug.edu.gh 66 negative stereotypic beliefs are associated with negative symbolic beliefs, and negative affect. The table also shows that there was no correlation between symbolic beliefs and affect (r[118] = .12, p = .11). This hypothesis was thus only partially confirmed. Table 5 Correlations among Attitude Components (Cognitive and Affective) Variables 1 2 3 1. Stereotypic Beliefs __ 2. Symbolic Beliefs .37** __ 3. Affect .16* .11 __ *p < .05. **p < .01. To complement the results obtained from testing the hypothesis, a content analysis of respondents‘ open-ended responses to the questions asking about their stereotypic beliefs, symbolic beliefs, and affect was done. Having grouped the responses based on their similarity in meaning, the three most frequently stated responses for each of the attitude components were arrived at as shown in Table 6 below. Table 6 Most Frequently Elicited Responses: Study 1 Stereotypic Beliefs Symbolic Beliefs Affect Not Religious Block respect Unhappy Immoral Block morality Disgust Evil Promote materialism Anger As evident in Table 6, with respect to respondents‘ stereotypic beliefs, they most frequently stated that homosexuals are not religious, are immoral, and evil and with respect to their symbolic beliefs, they perceived homosexuals as blocking respect, blocking morality, and promoting materialism. The feelings or emotions most elicited by University of Ghana http://ugspace.ug.edu.gh 67 homosexuals when respondents see, meet, or think about them were feelings of unhappiness, disgust, and anger. Hypothesis 4. This hypothesis stated that stereotypic beliefs, symbolic beliefs, and affect are likely to predict respondents‘ evaluation of homosexuals. Table 7 shows a summary of the results of the hierarchical regression analysis done to test this hypothesis. Respondents‘ personal characteristics (gender, age, academic level, and religion) were controlled. Stereotypic beliefs, symbolic beliefs, and affect were the independent variables whiles evaluation of homosexuals was the dependent variable. Table 7 Summary of Regression Analyses Examining Predictors of Evaluation of Homosexuals: Study 1 Predictors Evaluation of Homosexuals Β R2 ΔR2 Step 1 Gender Age Level Religion -.07 .03 .15 .03 .04 .04 Step 2 Stereotypic Beliefs .11 .16 .12 Symbolic Beliefs .30* Affect .03 *p < .05; N = 113 As shown in the table, only symbolic beliefs significantly predicted respondents‘ evaluation of homosexuals (β = .30, t = 3.038, p = 0.003). Although the contributions of stereotypic beliefs (β = .11, t = 1.104, p = 0.272) and affect (β = .03, t = 0.338, p = 0.736) to the variance in evaluation of homosexuals were not statistically significant at the 0.05 University of Ghana http://ugspace.ug.edu.gh 68 level, in addition to symbolic beliefs, they all explained 16% (R2 = .16) of the variance in evaluation of homosexuals. This would suggest that the present regression model is not a predictor of evaluation of homosexuals. However, based on the order of entry chosen for the present sample and the effect of the control variables, it would appear that symbolic beliefs explained the bulk of the variance in evaluation of homosexuals. Symbolic beliefs therefore emerged as the best predictor of respondents‘ evaluation of homosexuals. This hypothesis was thus only partially supported. Hypothesis 5. Respondents‘ religious commitment, moral values, and personal morals are likely to moderate the relationship between their attitude components and their evaluation of homosexuals. To test this hypothesis, it was necessary to first determine the correlation between all the variables involved as this is a prerequisite for performing moderated regression analysis. These variables include evaluation of homosexuals, the attitude components (i.e. stereotypic beliefs, symbolic beliefs, and affect), and the moderator variables (i.e. religious commitment, moral values, and personal morals). The results from testing Hypothesis 3 have already shown a correlation among stereotypic beliefs, symbolic beliefs, and affect so although they were also included in this analysis it was just as a matter of course. Table 8 shows the results of the correlation test (see the following page). As Table 8 indicates, a significant correlation was found between evaluation of homosexuals and stereotypic beliefs (r[183] = .28, p = .000), symbolic beliefs (r[123] = .34, p = .000), religious commitment (r[189] = -.15, p = .023), and moral values (r[184] = -.13, p = .036). This means respondents‘ stereotypic beliefs, symbolic beliefs, religious commitment, and moral values are all associated with their evaluation of homosexuals. Also, stereotypic beliefs were correlated with personal morals (r[173] = -.18, p = .007). Additionally, religious commitment was significantly correlated with moral values (r[185] = University of Ghana http://ugspace.ug.edu.gh 69 .23, p = .001) and personal morals (r[178] = .25, p = .000), while moral values was significantly correlated with personal morals (r[176] = .34, p = .036). In effect, the suspected moderator variables are associated with each other. Table 8 Summary of Intercorrelations for Evaluation of Homosexuals, Attitude Components, and Moderator Variables: Study 1 Variables 1 2 3 4 5 6 7 1. Evaluation of Homosexuals __ 2. Stereotypic Beliefs .28** __ 3. Symbolic Beliefs .34** .37** __ 4. Affect .05 .16* .11 __ 5. Religious Commitment -.15* -.05 .01 .05 __ 6. Moral Values -.13* .01 .08 .03 .23** __ 7. Personal Morals -.05 -.18** -.04 -.02 .25** .34** __ *p < .05. **p < .01. Based on these results, it was safe to proceed to test for whether religious commitment, moral values, and personal morals do moderate the relationship between stereotypic and symbolic beliefs and evaluation of homosexuals. Since no significant correlation was found between affect and evaluation of homosexuals and affect and the moderator variables, it was excluded from the moderation analyses. Analysis was done separately for stereotypic beliefs and symbolic beliefs and for each one, the personal characteristics of the respondents served as control variables. Therefore for the first analysis, these control variables were entered in the first block with centered scores for stereotypic beliefs, followed by a second block in which all centered moderator variables were entered, and a third block in which all the interaction terms were entered. The same University of Ghana http://ugspace.ug.edu.gh 70 procedure was used for the second analysis with stereotypic beliefs being replaced by symbolic beliefs. Table 9 shows summaries of the results (see the following page). As evident in the upper panel of Table 9, stereotypic beliefs had a significant main effect on evaluation of homosexuals (t(168) = 3.67, p = .000). It explained 11% of the variance in evaluation of homosexuals. The positive β value indicates that where respondents hold negative stereotypic beliefs about homosexuals, their evaluation of them was unfavourable but where they hold positive stereotypic beliefs their evaluation of homosexuals was favourable. However, religious commitment (t(168) = -1.67, p = .10), moral values (t(168) = -1.54, p = .13), and personal morals (t(168) = .71, p = .48) did not significantly predict evaluation of homosexuals. In addition, the results for the interaction effects of religious commitment (t(168) = -.37, p = .72), moral values (t(168) = -1.15, p = .25) and personal morals (t(168) = 1.45, p = .15) on the relationship between stereotypic beliefs and evaluation of homosexuals was not statistically significant. It is also evident in the lower panel of Table 9 that symbolic beliefs had a significant main effect on evaluation of homosexuals (t(115) = 3.57, p = .001). It explained 13% of the variance in evaluation of homosexuals. The positive β value indicates that where respondents believe homosexuals block certain values, customs, or traditions, their evaluation of them was unfavourable and where they believe homosexuals facilitate certain values, customs, or traditions, their evaluation of them was favourable. Religious commitment (t(115) = -.69, p = .50), moral values (t(115) = -.81, p = .42), and personal morals (t(115) = -.58, p = .57) did not significantly predict evaluation of homosexuals. The results for the interaction effects of religious commitment (t(115) = 1.30, p = .20), moral values (t(115) = -.40, p = .69) and personal morals (t(115) = .17, p = .87) on the relationship between symbolic beliefs and evaluation of homosexuals was not statistically significant. Therefore, this fifth hypothesis was not supported. University of Ghana http://ugspace.ug.edu.gh 71 Table 9 Summary of Regression Analyses Examining the Effect of the Moderator Variables on the relationship between Stereotypic Beliefsa, Symbolic Beliefsb, and Evaluation of Homosexuals Predictors Evaluation of Homosexuals Β R2 ΔR2 Step 1 Control variablesc and Stereotypic Beliefsd .27*d .11 Step 2 Religious Commitment -.13 .14 .04 Moral values -.12 Personal morals .06 Step 3 Stereotypic Beliefs × Religious Commitment -.03 .16 .02 Stereotypic Beliefs × Moral Values -.10 Stereotypic Beliefs × Personal Morals .12 Step 1 Control variables and Symbolic Beliefse .32**e .13 Step 2 Religious Commitment -.07 .16 .02 Moral values -.08 Personal morals -.06 Step 3 Symbolic Beliefs × Religious Commitment .13 .17 .01 Symbolic Beliefs × Moral Values -.05 Symbolic Beliefs × Personal Morals .02 aN = 170. bN = 117. cControl variables included age, gender, religion, and academic level. dIndependent variable for first analysis is Stereotypic Beliefs. eIndependent variable for second analysis is Symbolic Beliefs. *p < .01. **p < .05. University of Ghana http://ugspace.ug.edu.gh 72 Discussion The results of Study 1 indicate that attitudes toward homosexuals are negative as they were evaluated more unfavourably than the other target groups. Generally, the rating given homosexuals by respondents on the evaluation thermometer was ―very unfavourable‖ indicating that they have negative attitudes toward homosexuals. This finding is consistent with the literature (e.g. Ben-Ari, 2001; Haddock et al., 1993; Lim, 2008; Lim & Johnson, 2001; Owusu et al., 2013). As stated earlier, this study‘s methodology was very much informed by Haddock et al.‘s (1993) study which also used the Evaluation Thermometer and got their respondents to evaluate homosexuals as part of 5 target groups. Their results showed that attitudes toward homosexuals were negative and were significantly more unfavourable than their attitudes toward the other target groups. In effect, the findings of this study are consistent with theirs although the other target groups used by Haddock et al. were different from that used for this study. As noted previously, their study included English Canadians, French Canadians, Native Indians, and Pakistanis. Obviously, they had included homosexuals among target groups that all share the tag of being minority groups depending on where they might find themselves in Canada. For this study, that was not the case. The target groups were determined based on anecdotal evidence as to which groups of people are respected or not within Ghana. Orthodox Church Priests and Spiritual Church Leaders are very much respected by Ghanaians but there has been numerous instances in recent times where the latter have been in the news for offences that are not expected of them because of their profession and the former have come under suspicion because their counterparts in the West have been in the news for the sexual abuse of young boys. Traditional Priests are also respected but are not viewed in a kind light by most Ghanaians, at least not openly, because the majority of Ghanaians are Christians and therefore are or should be opposed to the traditional religion. University of Ghana http://ugspace.ug.edu.gh 73 Commercial Sex Workers (CSWs) are not respected at all because they are deemed as lacking moral virtue. According to Gyekye (2003), certain moral virtues or character traits are particularly valued in the African moral life, including chastity before marriage. Obviously, by the nature of their ‗profession‘, CSWs cannot be said to be practicing chastity before marriage. The CSWs themselves and their patrons are aware of these cultural values and therefore would not want to be known openly by others except those who share their interests. With this in mind, it was thought that adding these groups to homosexuals to form the target groups would be best since depending on the individual, any of these target groups can be evaluated positively or negatively. However, the CSWs were even evaluated marginally less unfavourably than homosexuals thus confirming the negative attitudes that Ghanaians have as evidenced by public discourse about homosexuality. It was suggested that respondents‘ personal characteristics, including age, gender, academic level, and religion are likely to influence their evaluation of homosexuals. This was however not the case and considering that the results also show that homosexuals were evaluated more unfavourably than the other target groups, it can be surmised that respondents‘ evaluation was done regardless of differences in age, gender, academic level, and religion. Often, studies show some differences in attitudes based on these personal characteristics (e.g., Arndt & de Bruin, 2006; Ohlander et al., 2005; Sakalli, 2002; Whitley, 2009) and therefore this finding is not consistent with some of these findings. However, with regards to gender, it is consistent with some aspects of Ben-Ari‘s (2001) findings because although she found males in her study to be more homophobic than females, academics in the field of psychology were the exception, showing similar levels of homophobia regardless of gender. The question therefore is whether the lack of gender differences in this study is also as a result of the respondents being psychology students. University of Ghana http://ugspace.ug.edu.gh 74 Perhaps that is the reason but the fact that this finding is totally consistent with Anarfi and Gyasi-Gyamerah‘s (2014) finding that no differences existed between males and females in their attitudes toward homosexuals points to a different reason. According to Anarfi and Gyasi-Gyamerah (2014), the lack of gender differences in their study is most likely due to differences in culture but they could not be emphatic due to some gender differences they noted among FGDs participants in their study. Obviously, the fact that the Ghanaian culture is different from that of western countries also explains the lack of gender differences in this study. As stated earlier, one of the reasons that Ghanaians have proffered for their seemingly negative attitudes toward homosexuals and homosexuality is the fact that it is against our culture. Homosexuality is considered a taboo that carries strong stigma culturally. For example, Gyekye (2003) noted that in both the traditional and contemporary Akan society of Ghana, when a man does not get married when he is expected to, he is regarded as a fool, a term that in the Akan sense means that the man is irresponsible, worthless, good-for-nothing, and contemptible. Other ethnic societies in Ghana have similar views. As such, for a man to prefer a homosexual relationship rather than a heterosexual one provides occasion for him to be considered worse than a fool. Considering therefore that both male and female Ghanaians are socialized within the same cultural environment, it is really not surprising that there are no differences in their attitudes. With regards to religion, although Anarfi and Gyasi-Gyamerah (2014) found that attitudes were generally negative regardless of respondents‘ religion, Pentecostal Christians proved to have the most negative attitude. This study did not look at differences in respondents‘ attitudes based on their religious denominational affiliations so perhaps this is one reason why no differences were found aside the fact that majority of the respondents in this study were Christians and therefore likely to share similar beliefs University of Ghana http://ugspace.ug.edu.gh 75 that can have similar impact on their attitudes. Also, majority of respondents had a mean age of 21.75 years and so the fact that they were homogeneous in that sense may be the reason for the lack of differences in their attitudes toward homosexuals. This homogeneity may also explain the lack of differences in attitudes based on academic level. All respondents were undergraduate students of about similar age, majority of who were in Levels 200 and 400 and therefore with quite similar experiences that are likely to affect their attitudes similarly. Had graduate students been part of the sample for this study, perhaps some differences in attitudes commensurate with the effect of education as noted in the literature (e.g. Loftus, 2001; Ohlander et al., 2005) would have been found. Correlation between attitude components. Are the components of attitudes (cognitive and affective) associated with each other in any way? Results from testing the hypothesis that there will be a positive correlation between respondents‘ stereotypic beliefs, symbolic beliefs, and affect, showed that they are associated with each other. Respondents‘ stereotypic beliefs were found to be associated with their symbolic beliefs and also with their affect while their symbolic beliefs and affect were not associated with each other. This is in contradiction to Haddock et al.‘s (1993) study where all three variables were positively correlated with each other. However, they concluded that these correlations were not high enough to suggest that the variables were eliciting information that were completely overlapping. This appears to be the case in the present study. It cannot be said that the stereotypic beliefs and affective information elicited from the respondents were completely overlapping and neither can it be said that the absence of correlation between symbolic beliefs and affect means that they elicited information that did not overlap at all. The most frequently elicited responses for stereotypic beliefs, symbolic beliefs, and affect show a support for this conclusion. Morality or the lack of it, on the part of University of Ghana http://ugspace.ug.edu.gh 76 homosexuals, was stated by respondents for both stereotypic beliefs and symbolic beliefs and may possibly underlie respondents‘ feelings of disgust stated under affect. Considering that morality was the second most elicited response for both stereotypic and symbolic beliefs and disgust was the second most elicited response for affect, this might probably be the point at which there are the partial overlaps between them. Incorporating literature on the relationship between morality and disgust further explains this link beyond mere probability. For example, based on four experiments conducted by Schnall, Haidt, Clore, and Jordan (2008) to determine how disgust influences moral judgements, they found that disgust can increase the severity of the moral judgements that people make depending on their sensitivity to their own bodily sensations. Also, Chapman, Kim, Susskind, and Anderson (2009) have stated based on the findings of their research into the oral origins of moral disgust that immorality elicits the same disgust as disease vectors and bad tastes. Additionally, Olivera La Rosa and Rossellό-Mir (2013) in their review of literature concluded that various studies have shown a bidirectional link between physical disgust-cleanliness and moral cognition although disgust is still a new area of research. As stated earlier, if these partial overlaps indeed exist between stereotypic beliefs, symbolic beliefs and affect because of the link between morality and disgust, then coupled with the fact that the correlations found were positive, the overlaps can be said to be driven solely by respondents‘ stereotypic beliefs. This means that where respondents felt that homosexuals were immoral, they likewise saw them as blocking morality and therefore felt disgusted at them. This is really not surprising because as noted previously, Olatunji (2008) stated that several domains of disgust, including core disgust, animal- reminder disgust, and contamination disgust may relate to negative attitudes toward homosexuals. These negative attitudes are further depicted in respondents‘ description of homosexuals as evil. This description encapsulates other similar responses such as University of Ghana http://ugspace.ug.edu.gh 77 devilish and demonic and though it is the third most elicited response under stereotypic beliefs, it also does not come as a surprise. This is due to its consistence with Owusu et al.‘s (2013) finding that the most prevalent view among their respondents was that MSM, WSW, and PSWBMFs‘ behaviours constitute a ―spiritual‖ problem because such persons are demonic and therefore bring curses and bad omen to the society. The third most elicited response under symbolic beliefs was that homosexuals promote materialism. There is no clear reason as to why respondents believe so but several of them stated as part of their stereotypic beliefs that homosexuals are rich and dress very fashionably so it is possible that the idea of them promoting materialism is based on such beliefs. Perhaps this is another point at which there is some overlap in the information elicited by stereotypic beliefs and symbolic beliefs. With regards to affect, feeling unhappy and feeling angry were the first and last most elicited response respectively. Given the fact that it has already been established from the results of this study that respondents evaluated homosexuals more unfavourably than the other target groups, it is not unexpected that some of them would report feeling angry when they see, meet, or think about homosexuals. What can be said to be somewhat unexpected is respondents‘ feeling of unhappiness in relation to homosexuals as one would have thought this impossible given their stereotypic and symbolic beliefs about them. It is however not surprising as the homosexuals that the respondents might have seen or might have met may be fellow students and/or relatives with whom they have some affinity. This affinity can cause feelings of unhappiness as they just might wish these fellow students and/or relatives were not exhibiting such behaviours. Other statements that were all collapsed into this point included the feeling of sadness and pity. The attitude components and their prediction of the evaluation of homosexuals. In addition to determining if there is a positive correlation among University of Ghana http://ugspace.ug.edu.gh 78 stereotypic beliefs, symbolic beliefs, and affect, it was also suggested that they are likely to predict respondents‘ evaluation of homosexuals. Knowing that respondents‘ personal characteristics might affect the effect of the predictor variables, they were controlled for by including them in the analysis. Although all the predictor variables together explained only a small percentage of the variance in respondents‘ evaluation of homosexuals, their symbolic beliefs explained the bulk of it. This is because only symbolic beliefs predicted respondents‘ evaluation of homosexuals and this is inconsistent with the findings from Haddock et al.‘s (1993) study. Based on findings of their previous study, Haddock et al. had hypothesized that the evaluative implications of stereotypic beliefs alone would be related to attitudes toward homosexuals but that when the evaluative implications of symbolic beliefs and affect are added on to it, there would be a significant increase in the ability to predict attitudes. Their results showed this to be the case and they therefore concluded that one must measure more than the evaluative implications of stereotypic beliefs in order to predict attitudes toward homosexuals more effectively (Haddock et al., 1993). In the case of this study, symbolic beliefs explained the bulk of the variance in evaluation of homosexuals which is a bit surprising considering the fact that, as stated earlier, it is respondents‘ stereotypic beliefs that were found to be associated with symbolic beliefs and affect. As a result, one would have thought that stereotypic beliefs would explain more of the variance in evaluation of homosexuals. It can be argued that the effect of the control variables may have caused this outcome but since these personal characteristics themselves have been found to have no effect on respondents‘ evaluation of homosexuals, then it is safe to say that they are not to blame. So then there is the need to find answers as to why symbolic beliefs were the best predictors of evaluation of homosexuals. Perhaps the answer lies in the most frequently elicited responses displayed University of Ghana http://ugspace.ug.edu.gh 79 in Table 6. The first most frequently elicited response under symbolic beliefs is the point that respondents felt that homosexuals block respect. It must be noted that there were various comments relating to respect that were all collapsed into one including homosexuals‘ lack of respect for the natural order, for the Ghanaian customs and traditions, for marriage and procreation, and their lack of respect for the self as a human being. Obviously all these statements are linked to one of them, that is the statement regarding Ghanaian customs and traditions which form the basis of the Ghanaian culture. So perhaps it is the view that homosexuals block respect for these customs and traditions, and therefore the Ghanaian culture that is the main fuel for respondents‘ symbolic beliefs being the best predictor of their evaluation of homosexuals. Morals, as is evidenced by the definition of culture by Tylor (1958) noted earlier, are of cultural value and so is religion which is based on belief. It has also been noted that African moral values have a humanistic basis rather than a religious one and therefore what is morally good is that which brings about human well-being (Gyekye, 2003). According to Gyekye, procreation is also of cultural value for Africans since it is viewed as the whole purpose of marriage. Additionally, Geertz (1973) has been noted earlier to define culture as ideas based on cultural learning and symbols that people internalize and use to define their world, express their feelings, and make their judgements such that it serves to help guide their behaviour and perceptions throughout their lives. One can therefore conclude that this explains why symbolic beliefs were the best predictor of respondents‘ evaluation of homosexuals. Effect of the moderator variables and evaluation of homosexuals. It is important to note that in addition to stereotypic beliefs and affect, symbolic beliefs accounted for only a small percentage of the variance in evaluation of homosexuals so the majority was left unexplained. Answers must therefore be found for this. This is University of Ghana http://ugspace.ug.edu.gh 80 especially so since the hypothesis that respondents‘ religious commitment, moral values, and personal morals will moderate the relationship between their attitude components and their evaluation of homosexuals was not supported. It was hoped that the interactions between these suspected moderators and the attitude components would explain a large proportion of the variance in respondents‘ evaluation of homosexuals if not all of it. Rather, even though they were associated with respondents‘ evaluation of homosexuals on their own with the exception of personal morals, when they interacted with the attitude components, they did not predict evaluation of homosexuals at all. However, when these interaction effects were factored in with respondents‘ personal characteristics, their stereotypic beliefs and symbolic beliefs explained the variance in evaluation of homosexuals only marginally more than usual. As such, the problem as to what explains the majority of the variance in respondents‘ evaluation of homosexuals still exists and for this answers have to be sought. All the same, it is worth noting that respondents‘ religious commitment is associated with their evaluation of homosexuals. This is an interesting fact considering that their religion was found to have no association with their evaluation of homosexuals. The negative r value, although low, indicates that where respondents had high religious commitment, they evaluated homosexuals more unfavourably and where they had low religious commitment, they evaluated homosexuals more favourably. This is the case also for the association between respondents‘ moral values and their evaluation of homosexuals which also got a negative r value. Also interesting is the fact that respondents‘ religious commitment was found to be associated with their moral values, the positive r value indicating that where they had high religious commitment, they also had high moral values and vice versa. Additionally, respondents‘ moral values were found to be associated with personal morals, with the positive r value showing that where they had high moral values, University of Ghana http://ugspace.ug.edu.gh 81 they also had high personal morals and vice versa. Respondents‘ personal morals were found to be negatively associated with their stereotypic beliefs, meaning that where respondents had low personal morals, they had positive stereotypic beliefs about homosexuals and where they had high personal morals, they had negative stereotypic beliefs about homosexuals. It is the fact that respondents‘ personal morals did not have any association with their evaluation of homosexuals that is quite unexpected taking into cognizance the fact that it has already been established that respondents listed morality under both stereotypic beliefs and symbolic beliefs as the second most frequently elicited response. Does this mean that respondents‘ own level of morality had nothing to do with how they felt about homosexuals? This does not seem very likely if the positive association between moral values and personal morals are taken into consideration. Perhaps the fact is that respondents draw some aspects of their personal morals from their moral values which in turn are, to a certain extent, based on their religious commitment. If so, then this also ties in with the finding that symbolic beliefs are the best predictor of evaluation of homosexuals because obviously respondents‘ idea of what the Ghanaian society views as right or wrong, which is embodied by their moral values and religiousity and are part of the Ghanaian culture, is what informs their evaluation of homosexuals. Conclusion and Observed Model for Study One As is evident from the findings of this study, respondents showed negative attitudes toward homosexuals based on how they evaluated them and their attitudes had nothing to do with differences in their personal characteristics. Also, with regards to correlation between the respondents‘ attitude components, their stereotypic beliefs were found to be associated with their symbolic beliefs and affect but this association shows University of Ghana http://ugspace.ug.edu.gh 82 only a partial overlapping of information elicited by all three variables. It was deduced based on respondents‘ most elicited responses regarding these variables that the overlaps among them are due to the link between their thoughts about the morality or lack of it on the part of homosexuals and their feeling of disgust towards them. Additionally, it was found that only respondents‘ symbolic beliefs is the best predictor of their evaluation of homosexuals and therefore explain the bulk of the variance in their evaluation. Again, based on deductions made from the respondents‘ most elicited responses, it appears symbolic beliefs is the best predictor of respondents‘ unfavourable evaluation of homosexuals because of their view that homosexuals block respect for certain aspects of the Ghanaian culture. Finally, it was found that although respondents‘ religious commitment and moral values are associated with their evaluation of homosexuals, they did not moderate the relationship between their attitude components and their evaluation of homosexuals. However, combined with respondents‘ personal characteristics, religious commitment and moral values had the effect of marginally increasing the extent to which respondents‘ stereotypic and symbolic beliefs explain the variance in their evaluation of homosexuals. Majority of the variance was however left unexplained. Based on these findings, the following observed model (Figure 3) was developed. The variables are shown in rectangles of various sizes that are solely meant to accommodate the variable names. Correlations or associations are depicted by normal double-tipped arrows while double-tipped arrows having both a normal tip and a solid tip at opposite ends depict a correlation as well as a prediction. University of Ghana http://ugspace.ug.edu.gh 83 Figure 3. Observed model showing findings of Study 1. Symbolic Beliefs Stereotypic Beliefs Affect Attitudes toward Homosexuals (Evaluation of Homosexuals) Moral Values Religious Commitment Personal Morals β = .32 University of Ghana http://ugspace.ug.edu.gh 84 Chapter Four Study Two Introduction Study 2 was conducted as a follow-up to Study 1 to ascertain whether findings of the first study will be replicated and also to determine the role of attributions in participants‘ attitudes toward homosexuals. To that end, this study used a pre-test – post- test experimental design. This chapter therefore includes an outline of the methods used to collect data, the results obtained, and a discussion of the results with reference to the literature reviewed in Chapter Two. Seven hypotheses were tested using the Statistical Package for the Social Sciences (SPSS, version 16) and all of them were confirmed except one. Results of the moderated regression analysis performed to test the fourth hypothesis were also fed into a computer programme designed by Jose (2002) known as ModGraph (Beta version) to generate the moderation graphs. Methodology Population/Sample. The population for this study was the same as for Study 1. Two hundred (200) students were selected through convenient sampling from this population. However, seventeen (17) of them did not participate eventually so the total number of respondents was one hundred and eighty-three (183) students. The mean age of respondents was 21.5 years (SD = 3.10) and majority of them were Level 200 students. Table 10 shows details of their demographics. University of Ghana http://ugspace.ug.edu.gh 85 Table 10 Demographic Characteristics of Respondents: Study 2 Instruments. Eight instruments were used for this study including the Group Evaluation Questionnaire (GEQ), same as for Study 1, the Group Evaluation and Behavioural Intentions Form (GEBIF), and five vignettes with various titles and different contents. The GEQ was scored the same as for the first study (see Chapter 3). The GEBIF is self-titled and like the GEQ, it was so titled to disguise the true nature of the study and reduce demand characteristics. It has two sections (see Appendix B). Section A consists of the Evaluation Thermometer Measure for Assessing Attitudes toward Gay Men which is the same as for Section C of the GEQ and was scored in the same manner. Section B dealt with respondents‘ behavioural intentions toward the vignette persons. Variables Frequency (n = 183) Percent (%) Gender Male 103 56.3 Female 80 43.7 Age M = 21.5, SD = 3.10 18 – 25 years 167 91.3 26 – 53 years 16 8.7 Academic Level Level 100 28 15.3 Level 200 60 32.8 Level 300 46 25.1 Level 400 49 26.8 Religion Christianity 168 91.8 Islam 14 7.7 Other 1 0.5 University of Ghana http://ugspace.ug.edu.gh 86 The vignettes were all self-authored short stories. Vignettes 1, 2, 3, and 4 described the lives of certain fictional homosexual persons. As the stories were informed by the attribution theory of controllability (see Chapter 2), they were designed to depict the persons described as homosexual either due to genetics or due to choice. Those that described the vignette persons as homosexual due to genetics were especially based on information on genetic studies relating to homosexuality found in Strong et al. (2002) and LeVay and Valente (2003). Vignette 5, which served as a control for the effects of the other four vignettes, described the lives of a teenage heterosexual couple. Details of the contents of the vignettes are summarized in Table 11. Table 11 Summary of Experimental Design and Content of Vignettes Experimental Groups/Conditions Condition Pre-test Intervention Post-test 1. Biological GEQ Vignette 1, titled ―Life is Good‖ shows homosexuality as part of person‘s biological make-up using scientific evidence. GEBIF 2. Choice GEQ Vignette 2, titled ―My Life is Mine to Live‖ shows homosexuality as a person‘s chosen way of life. GEBIF 3. Biological transgendered GEQ Vignette 3, titled ―A New Lease on Life‖ shows homosexuality as due to a biological aberration necessitating a person‘s transgendered state. GEBIF 4. Choice Transgendered GEQ Vignette 4, titled ―Life is Comfortable‖ shows homosexuality as due to a person‘s chosen transgendered state. GEBIF 5. Teenage Pregnancy GEQ Vignette 5, titled ―Just Five Minutes of Pleasure‖ shows a teenage couple dealing with an unwanted pregnancy. GEBIF Note. GEQ means Group Evaluation Questionnaire. GEBIF means Group Evaluation and Behavioural Intentions Form. Vignettes 1 to 4 talk about the lives of homosexual persons and Vignette 5 talks about a heterosexual couple. University of Ghana http://ugspace.ug.edu.gh 87 After reading one of the vignettes, the single question for Section B demanded that respondents imagine they know the vignette person(s) personally and then write out how they are likely to behave towards him/her and why. Participants‘ responses were collated and coded from 0 (extremely unfavourable behaviour) to 6 (extremely favourable behaviour), with 3 (neither favourable nor unfavourable behaviour) being the midpoint. The codes in-between 0 and 3 and in-between 3 and 6 referred to either ‗quite‘ or ‗slightly unfavourable behaviour‘ and ‗quite‘ or ‗slightly favourable behaviour‘. For Vignettes 1, 2, 3, and 4, these codes referred to similar behavioural intentions as they all were in reference to homosexuals while for Vignette 5 they referred to different behavioural intentions. Table 12 provides details of the behavioural intentions they referred to (see the following page). University of Ghana http://ugspace.ug.edu.gh 88 Table 12 Participants‟ Behavioural Intentions Codes Behavioural Intentions Vignettes 1 – 4 Vignette 5 0 Totally unaccepting of the person; find it simply unthinkable and disgusting; would have nothing to do with the person. Angry at them; young people should not engage in premarital sex so they should have known better. 1 See it as totally wrong and so the person must change; will be extremely cautious around the person. They are just plain stupid because there are a lot of ways to prevent pregnancy in these modern days. 2 Somewhat unaccepting; feeling pity/sorry for the person; wanting to help but avoiding influence; not sure it‘s wise to be friendly with the person. Scold and pity them but not offer any support or help/comfort. 3 Unsure how to react; ok with it; just fine with it. Indifferent; unsure how to react. 4 Low level acceptance/tolerance with some concerns/caution; will feel a bit awkward though; want to help person get out of it; feel worried about person; mildly curious. Support/Acceptance but with disappointment in them for not being stronger to resist the temptation/ for not using preventive methods. 5 Moderate level acceptance; live and let live; it‘s the person‘s choice and this should be accepted. Let them know they were wrong but since it has already happened, no need for blame game; can happen to anyone of us; important thing is to find solution to it; won‘t want others to change their behaviour towards me if it were me; people make mistakes; parents can help. 6 Totally accepting with full support and readiness to offer whatever help needed by the person to be who he or she is. Response for code ‗5‘ plus support for them in whatever they decide to do. Design and Procedure. The design for this study is a pre-test – post-test experimental design (4 experimental groups and 1 control group) (see Table 11). Since ethical clearance had already been sought and received for the first study and a 6-month University of Ghana http://ugspace.ug.edu.gh 89 extension had also been granted for the continuation of data collection, there was no need to repeat the process, especially since the nature of the study did not really change. Participants were selected through the same procedure as was used for the first study but the various classes were informed that those who had already participated in the first study could not participate again. Those who volunteered to be part of this study were all then asked to meet at the Department of Psychology on a set date and time. When participants‘ arrived on the set date, they were asked to group themselves according to gender (male and female). Afterwards, members of the male group were each asked to pick a number out of a bowl containing 100 paper chips having the numbers 1, 2, 3, 4, or 5 written on them (20 chips per number). These numbers represented the experimental groups 1 to 4 and control group 5 so that based on the numbers picked by participants they were randomly assigned to the various groups. The same procedure was used to randomly assign the females into the various groups. As was the case for the first study, not all of the pre-selected participants came and so efforts were made to make up the numbers using the same procedure used previously. The participants who volunteered in this manner were also randomly assigned to the various groups through the same procedure noted earlier. Since the venue for the study was the same as for the first study, it meant that data collection had to be done over a 5-day period and not simultaneously. Participants were therefore rescheduled to come on another set of dates for the actual data collection in groups of 40 each. Also, as the venue could seat not more than 20 participants at a time, each group had to be split into two which also meant scheduling the participants for two separate times on each set date. Unfortunately, some participants could not be present on the set dates and times, perhaps due to the fact that they had lectures then. It is for this University of Ghana http://ugspace.ug.edu.gh 90 reason that the groups eventually had less than the expected 40 participants with unequal number of males and females as seen from Table 13. Table 13 Cross-tabulation of Gender and Experimental Groups/Conditions Experimental Groups/Conditions Gender Total Male Female 1. Biological 20 15 35 2. Choice 17 19 36 3. Biological transgendered 25 13 38 4. Choice Transgendered 22 14 36 5. Teenage Pregnancy 19 19 38 Before the start of each data collection process, participants were again reminded that the study was intended to measure group evaluations in keeping with the need to use deception as per the goal of the study, and that they still had the option to withdraw from the study at any point during the session. They were then given two consent forms each to read and sign, after which they returned one and kept the other. The pre-test condition was the same for all participants in the experimental groups and the control group and since the pre-test instrument (GEQ) was the same as the one used in the first study, the specific instructions given before participants completed the questionnaires was also same (see Chapter 3). Participants were asked to first indicate their serial codes on the questionnaires before completing them. After their completion, the GEQs were collected and the appropriate vignettes, which served as the intervention, were given to them to read (see Table 12). They had 10 minutes in which to do this and the vignettes were collected when the time was up. The post-test condition, which involved completing the GEBIF, was also the same for all participants in the experimental groups and the control group and came right after the University of Ghana http://ugspace.ug.edu.gh 91 intervention was done. Participants had to indicate their serial codes on their GEBIFs too so they could be matched to their GEQs. The GEBIF had instructions that are straightforward, a fact that was confirmed after it was pretested, and so clarifications were only made when a participant asked for it directly. The whole experimental session lasted for up to 1 hour 30 minutes for each group. Results Hypothesis 1. The first hypothesis for this study stated that participants are likely to evaluate homosexuals more unfavourably than the other target groups. Table 14 summarizes the results obtained from testing it using repeated measures ANOVA. Table 14 Evaluation of Target Groups: Study 2 Target Groups M (SD) df F p Orthodox Church Priests 75.06 (11.57) 4/716 483.72* .000 Commercial Sex Workers 12.61 (9.78) Spiritual Church Leaders 70.72 (15.06) Homosexuals 6.89 (3.28) Traditional Priests 24.00 (5.43) *Greenhouse-Geisser; N = 180 As can be seen from Table 14, significant differences were found in respondents‘ evaluation of the target groups (F(4,716) = 483.72, p = .000) though it was unclear where these differences lay. A post hoc analysis of the mean scores was therefore performed and the results are shown in Table 15 (see the following page). University of Ghana http://ugspace.ug.edu.gh 92 Table 15 Bonferroni Multiple Comparison of Evaluation of Target Groups: Study 2 Target Groups 1 2 3 4 5 6. 1. Orthodox Church Priests __ 7. 2. Commercial Sex Workers 62.44* __ 8. 3. Spiritual Church Leaders 4.33 58.11* __ 9. 4. Homosexuals 68.17* 5.72* 63.83* __ 10. 5. Traditional Priests 51.06* 11.39* 46.72* 17.11* __ *p < 0.05 From Table 15, it is obvious that significant mean differences were revealed in participants‘ evaluation of all target groups (p < 0.05) except their evaluation of Orthodox Church Priests and Spiritual Church Leaders. Considering the mean scores shown in Table 14, homosexuals got the most unfavourable evaluation (M = 6.89, SD = 3.28), followed by Commercial Sex Workers (M = 12.61, SD = 9.78), and Traditional Priests (M = 24.00, SD = 5.43). The target group that received the most favourable evaluation was Orthodox Church Priests (M = 75.06, SD = 11.57), followed by Spiritual Church Leaders (M = 70.72, SD = 15.06). A pictorial view of the various target groups‘ approximate position on the evaluation thermometer based on the mean scores of the participants is shown in Figure 4 (see the following page). University of Ghana http://ugspace.ug.edu.gh 93 POSITIVE 100o Extremely Favourable 90o Very Favourable 80o Quite Favourable 70o Fairly Favourable 60o Slightly Favourable 50o Neither Favourable nor Unfavourable 40o Slightly Unfavourable 30o Fairly Unfavourable 20o Quite Unfavourable 10o Very Unfavourable NEGATIVE 0o Extremely Unfavourable Figure 4. Evaluation thermometer showing points corresponding with Study 2 participants‘ mean evaluation scores for the various target groups. O.C.P. = Orthodox Church Priests; S.C.L. = Spiritual Church Leaders; T.P. = Traditional Priests; CSWs = Commercial Sex Workers; H. = Homosexuals. As evident in Figure 4, generally, Orthodox Church Priests got the evaluation ‗quite favourable‘, Commercial Sex Workers got ‗quite unfavourable‘, Spiritual Church Leaders got ‗fairly favourable‘, Homosexuals got ‗very unfavourable‘, and Traditional Priests got ‗fairly unfavourable‘ from respondents. Thus, as was hypothesized, participants evaluated homosexuals more unfavourably than the other target groups. Hypothesis 2. This hypothesis stated that there will be a positive correlation between stereotypic beliefs, symbolic beliefs, and affect. The summary of results of the Pearson product moment correlation performed to test it can be seen in Table 16 (see the following page). S.C.L. O.C.P. T.P. CSWs H. University of Ghana http://ugspace.ug.edu.gh 94 Table 16 Correlations among Attitude Components (Cognitive and Affective) – Study 2 Variables 1 2 3 1. Stereotypic Beliefs __ 2. Symbolic Beliefs .38** __ 3. Affect .29** .32** __ **p < .01. As can be seen in the table, participants‘ stereotypic beliefs were significantly positively correlated with their symbolic beliefs (r[122] = .38, p = .00) and their affect (r[160] = .29, p = .00) while participants‘ symbolic beliefs were also significantly correlated with their affect (r[120] = .32, p = .00). The implication is that positive stereotypic beliefs are associated with positive symbolic beliefs and positive affect, and positive symbolic beliefs are also associated with positive affect. At the same time, negative stereotypic beliefs are associated with negative symbolic beliefs and negative affect and negative symbolic beliefs are also associated with negative affect. Using the same procedure as was used for Study 1, a content analysis of participants‘ open-ended responses to the questions asking about their stereotypic beliefs, symbolic beliefs, and affect was done to complement the results obtained from testing this hypothesis. Table 17 shows the participants‘ three most frequently stated responses for each of the attitude components (see the following page). As the table shows, with respect to participants‘ stereotypic beliefs, they most frequently stated that homosexuals are not religious, are immoral, and evil and with respect to their symbolic beliefs, they perceived homosexuals as blocking morality, blocking marriage, and promoting respect for human rights. The feelings or emotions most elicited by homosexuals when participants see, meet, or think about them were feelings of anger, unhappiness, and disgust. University of Ghana http://ugspace.ug.edu.gh 95 Table 17 Most Frequently Elicited Responses: Study 2 Stereotypic Beliefs Symbolic Beliefs Affect Not Religious Block Morality Anger Immoral Block marriage Unhappy Evil Promote respect for human rights Disgust Hypothesis 3. The third hypothesis stated that stereotypic beliefs, symbolic beliefs, and affect are all likely to predict participants‘ evaluation of homosexuals. Table 18 summarizes results of the hierarchical regression analysis done to test this hypothesis. Stereotypic beliefs, symbolic beliefs, and affect were the independent variables whiles evaluation of homosexuals was the dependent variable. Participants‘ personal characteristics (gender, age, academic level, and religion) were controlled for. Table 18 Summary of Regression Analyses Examining Predictors of Evaluation of Homosexuals: Study 2 Predictors Evaluation of Homosexuals Β R2 ΔR2 Step 1 Gender -.16 .05 .05 Age .02 Level .16 Religion .00 Step 2 Stereotypic Beliefs .21* .21 .17 Symbolic Beliefs .03 Affect .30* *p < .05; N = 114 University of Ghana http://ugspace.ug.edu.gh 96 As the table shows, both stereotypic beliefs (β = .21, t = 2.129, p = 0.036) and affect (β = .30, t = 3.174, p = 0.002) significantly predicted participants‘ evaluation of homosexuals. Although the contribution of symbolic beliefs (β = .03, t = 0.295, p = 0.769) was not statistically significant at the 0.05 level, in addition to stereotypic beliefs and affect, they all explained 21% (R2 = .21) of the variance in evaluation of homosexuals. This would suggest that the present regression model is not a very strong predictor of evaluation of homosexuals. However, based on the order of entry chosen for the present sample and the effect of the control variables, it would appear that stereotypic beliefs and affect explained the bulk of the variance in evaluation of homosexuals. This makes them the best predictors of participants‘ evaluation of homosexuals. This hypothesis was thus only partially supported. Hypothesis 4. This hypothesis stated that participants‘ religious commitment, moral values, and personal morals are likely to moderate the relationship between their attitude components and their evaluation of homosexuals. Table 19 shows the results of the initial correlation test done on all the variables involved as a prerequisite for conducting the moderated regression analysis (see the following page). As the table shows, a significant correlation was found between evaluation of homosexuals and stereotypic beliefs (r[170] = .27, p = .000), affect (r[166] = .24, p = .002), and moral values (r[174] = -.25, p = .001). This means that participants‘ stereotypic beliefs, affect, and moral values were all associated with their evaluation of homosexuals. With regards to all three attitude components, it has already been established from the results of Hypothesis 3 that a correlation exists among them as is also evident from Table 19. In addition, stereotypic beliefs were correlated with moral values (r[165] = -.16, p = .039). Finally, as the table shows, religious commitment was significantly correlated with moral values (r[168] = .33, p = .000) and personal morals (r[171] = -.25, p = .001), while University of Ghana http://ugspace.ug.edu.gh 97 moral values was significantly correlated with personal morals (r[174] = -.25, p = .001). This means that the suspected moderator variables were also associated with each other. It was therefore safe to proceed to test for whether these suspected moderator variables do moderate the relationship between stereotypic beliefs, symbolic beliefs, affect, and evaluation of homosexuals. Table 19 Summary of Intercorrelations for Evaluation of Homosexuals, Attitude Components, and Moderator Variables: Study 2 Variables 1 2 3 4 5 6 7 1. Evaluation of Homosexuals __ 2. Stereotypic Beliefs .27** __ 3. Symbolic Beliefs -.15 .38** __ 4. Affect .24** .29** .32** __ 5. Religious Commitment -.10 -.10 .06 -.10 __ 6. Moral Values -.25** -.16* -.09 -.12 .33** __ 7. Personal Morals .09 .06 .13 .13 -.25** -.25** __ *p < .05. **p < .01. Although the analyses to test this hypothesis were done in the same manner as for the first study in terms of controlling for the participants‘ demographic variables and entering the variables in the model, there were some differences. Firstly, instead of analyzing each of the attitude components (i.e. stereotypic beliefs, symbolic beliefs, and affect) with all the suspected moderator variables (i.e. religious commitment, moral values, and personal morals) all at once, each attitude component was analyzed with each of the suspected moderator variables separately. This change was made because it was assumed that perhaps the moderating effects of the moderator variables might not have University of Ghana http://ugspace.ug.edu.gh 98 been fully realized in the first study because these variables might have had some control effects on each other. Secondly, instead of excluding some variables from the analyses as was done for Study 1, in the present study, all the variables were included in the analyses. Upon analyses, it was found that personal morals did not moderate the relationship between any of the attitude components and evaluation of homosexuals while religious commitment did not moderate the relationship between affect and evaluation of homosexuals. Table 20 shows summaries of the results from testing the moderating effects of religious commitment and moral values on the relationship between stereotypic beliefs and evaluation of homosexuals (see the following page). As can be seen in the top panel of the table, together with the control variables, stereotypic beliefs had a significant main effect on evaluation of homosexuals (t(159) = 3.36, p = .001). It explained 10% of the variance in evaluation of homosexuals. The positive β value indicates that where participants hold negative stereotypic beliefs about homosexuals, their evaluation of them was unfavourable. Similarly, where participants hold positive stereotypic beliefs about homosexuals, their evaluation of them was favourable. Also, religious commitment (t(159) = -.81, p = .42) did not significantly predict evaluation of homosexuals but results for its interaction effect (t(159) = 3.43, p = .001) on the relationship between stereotypic beliefs and evaluation of homosexuals was statistically significant, explaining 6% of the variance in evaluation of homosexuals. University of Ghana http://ugspace.ug.edu.gh 99 Table 20 Summary of Regression Analyses Examining the Moderating Effect of Religious Commitment and Moral Values on the relationship between Stereotypic Beliefs and Evaluation of Homosexuals Predictors Evaluation of Homosexuals Β R2 ΔR2 Step 1 Control Variables and Stereotypic Beliefsa .26*a .10 Step 2 Religious Commitment -.06 .10 .00 Step 3 Stereotypic Beliefs × Religious Commitment .63* .16 .06 Step 1 Control Variables and Stereotypic Beliefsb .26*b .09 Step 2 Moral values -.22** .14 .05 Step 3 Stereotypic Beliefs × Moral Values -.75* .22 .09 Note. Control variables included age, gender, religion, and academic level but are referred to as such in the table for the sake of simplicity. The β values for these control variables are not included in the table. aN = 161. bN = 163. *p < .01. **p < .05. Figure 5 shows a graphical view of this moderation (see the following page). It is obvious from the figure that, participants who hold more negative stereotypic beliefs about homosexuals (high) and also have high religious commitment evaluated homosexuals more unfavourably than those who hold more positive stereotypic beliefs (low) about homosexuals and also have low religious commitment. University of Ghana http://ugspace.ug.edu.gh 100 Figure 5. Moderation graph showing the moderating effect of participants‘ religious commitment on the relationship between their stereotypic beliefs and their evaluation of homosexuals. The lower panel of Table 20 also shows that together with the control variables, stereotypic beliefs had a significant main effect on evaluation of homosexuals (t(161) = 3.28, p = .001). It explained 9% of the variance in evaluation of homosexuals and the positive β value gives the same indication as the findings shown in the top panel of the table as noted earlier. Also, moral values (t(161) = -2.85, p = .005) had a significant effect on evaluation of homosexuals. The negative β value indicates that where participants have high moral values, their evaluation of homosexuals was unfavourable (low scores), and where participants have low moral values, their evaluation of homosexuals was favourable (high scores). Additionally, results for its interaction effect (t(161) = 4.13, p = .000) on the relationship between stereotypic beliefs and evaluation of homosexuals was statistically significant, explaining 9% of the variance in evaluation of homosexuals. Figure 6 shows a graphical view of this moderation (see the following page). -18 -16 -14 -12 -10 -8 -6 -4 -2 0 low med high E v a lu a ti o n o f H o m o se x u a ls Stereotypic Beliefs Religious Commitment as Moderator Religious Commitment High RC Low RC University of Ghana http://ugspace.ug.edu.gh 101 Figure 6. Moderation graph showing the moderating effect of participants‘ moral values on the relationship between their stereotypic beliefs and their evaluation of homosexuals. It is obvious from Figure 6, participants who hold more negative stereotypic beliefs about homosexuals (high) and also have high moral values, evaluated homosexuals more unfavourably than those who hold more positive stereotypic beliefs (low) about homosexuals and also have low moral values. Participants‘ symbolic beliefs and how these relate to their evaluation of homosexuals in the presence of religious commitment or moral values were also examined. Table 21 shows summaries of the results (see the following page). From the top panel of Table 21, it can be seen that together with the control variables, symbolic beliefs had a significant main effect on evaluation of homosexuals (t(116) = 2.71, p = .008). It explained 11% of the variance in evaluation of homosexuals. The positive β value indicates that where participants hold negative symbolic beliefs about homosexuals, their -30 -25 -20 -15 -10 -5 0 low med high E v a lu a ti o n o f H o m o se x u a ls Stereotypic Beliefs Moral Values as Moderator Moral Values High MV Low MV University of Ghana http://ugspace.ug.edu.gh 102 evaluation of them was unfavourable. Similarly, where participants hold positive symbolic beliefs about homosexuals, their evaluation of homosexuals was favourable. Table 21 Summary of Regression Analyses Examining the Moderating Effect of Religious Commitment and Moral Values on the relationship between Symbolic Beliefs and Evaluation of Homosexuals Predictors Evaluation of Homosexuals Β R2 ΔR2 Step 1 Control variables and Symbolic Beliefsa .25**a .11 Step 2 Religious Commitment -.20** .15 .04 Step 3 Symbolic Beliefs × Religious Commitment .72* .24 .09 Step 1 Control variables and Symbolic Beliefsb .30*b .14 Step 2 Moral values -.19** .18 .04 Step 3 Symbolic Beliefs × Moral Values .47** .21 .04 Note. Control variables included age, gender, religion, and academic level but are referred to as such in the table for the sake of simplicity. The β values for these control variables are not included in the table. aN = 118. bN = 120. *p < .01. **p < .05. Additionally, religious commitment (t(116) = -2.19, p = .030) significantly predicted evaluation of homosexuals. The negative β value indicates that where participants have high religious commitment, their evaluation of homosexuals was unfavourable (low scores), and where participants have low religious commitment, their evaluation of University of Ghana http://ugspace.ug.edu.gh 103 homosexuals was favourable (high scores). Also, results for its interaction effect (t(116) = 3.62, p = .000) on the relationship between symbolic beliefs and evaluation of homosexuals was statistically significant. This explained 9% of the variance in evaluation of homosexuals. Figure 7 below shows a graphical view of this moderation. Figure 7. Moderation graph showing the moderating effect of participants‘ religious commitment on the relationship between their symbolic beliefs and their evaluation of homosexuals As is evident in Figure 7, participants who hold more negative symbolic beliefs about homosexuals (high) and also have high religious commitment evaluated homosexuals more unfavourably than those who hold more positive symbolic beliefs (low) about homosexuals and also have low religious commitment. The lower panel of Table 21 shows that together with the control variables, symbolic beliefs had a significant main effect on evaluation of homosexuals, (t(118) = 3.38, -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 low med high E v a lu a ti o n o f H o m o se x u a ls Symbolic Beliefs Religious Commitment as Moderator Religious Commitment High RC Low RC University of Ghana http://ugspace.ug.edu.gh 104 p = .001). It explained 14% of the variance in evaluation of homosexuals and the negative β value gives the same indication as the findings shown in the top panel of the table as noted earlier. Also, moral values (t(118) = -2.18, p = .031) had a significant effect on evaluation of homosexuals. The negative β value indicates that where participants have high moral values, their evaluation of homosexuals was unfavourable (low scores), and where participants have low moral values, their evaluation of homosexuals was favourable (high scores). Additionally, results for its interaction effect (t(118) = 2.22, p = .028) on the relationship between stereotypic beliefs and evaluation of homosexuals was statistically significant, explaining 4% of the variance in evaluation of homosexuals. Figure 8 below shows a graphical view of this moderation. Figure 8: Moderation graph showing the moderating effect of participants‘ moral values on the relationship between their symbolic beliefs and their evaluation of homosexuals -16 -14 -12 -10 -8 -6 -4 -2 0 2 low med high E v a lu a ti o n o f H o m o se x u a ls Symbolic Beliefs Moral Values as Moderator Moral Values High MV Low MV University of Ghana http://ugspace.ug.edu.gh 105 As evident in Figure 8, participants who hold more negative symbolic beliefs about homosexuals (high) and also have high moral values evaluated homosexuals more unfavourably than those who hold more positive symbolic beliefs (low) about homosexuals and also have low moral values. Finally, the moderating effect of moral values on the relationship between participants‘ affect and their evaluation of homosexuals was also analysed. Table 22 summarizes the results of this analysis. Table 22 Summary of Regression Analyses Examining the Moderating Effect of Moral Values on the relationship between Affect and Evaluation of Homosexuals Predictors Evaluation of Homosexuals Β R2 ΔR2 Step 1 Control Variables and Affect .22** .05 Step 2 Moral values -.23* .10 .05 Step 3 Affect × Moral Values -.80* .16 .06 Note. Control variables included age, gender, religion, and academic level but are referred to as such in the table for the sake of simplicity. The β values for these control variables are not included in the table. *p < .01. **p < .05. As can be seen in Table 22, together with the control variables, affect had a significant main effect on evaluation of homosexuals (t(158) = 2.77, p = .006), explaining 5% of the variance in evaluation of homosexuals. The positive β value indicates that where participants have negative affect for homosexuals, their evaluation of them was unfavourable, and where participants have positive affect for homosexuals, their University of Ghana http://ugspace.ug.edu.gh 106 evaluation of them was favourable. Also, moral values (t(158) = -2.99, p = .003) had a significant effect on evaluation of homosexuals. The negative β value indicates that where participants have high moral values, their evaluation of homosexuals was unfavourable (low scores), and where participants have low moral values, their evaluation of homosexuals was favourable (high scores). Additionally, results for the interaction effect of moral values (t(158) = 3.38, p = .001) on the relationship between stereotypic beliefs and evaluation of homosexuals was statistically significant, explaining 6% of the variance in evaluation of homosexuals. Figure 9 below shows a graphical view of this moderation. It depicts that participants who have more negative affect for homosexuals (high) and also have high moral values evaluated homosexuals more unfavourably than those who have more positive affect (low) for homosexuals and also have low moral values. Figure 9. Moderation graph showing the moderating effect of participants‘ moral values on the relationship between their affect and their evaluation of homosexuals. -25 -20 -15 -10 -5 0 low med high E v a lu a ti o n o f H o m o se x u a ls Affect Moral Values as Moderator Moral Values High MV Low MV University of Ghana http://ugspace.ug.edu.gh 107 In summary, the interaction effects of religious commitment and moral values on the relationship between stereotypic beliefs and evaluation of homosexuals and their interaction effects on the relationship between symbolic beliefs and evaluation of homosexuals were all statistically significant. Also, the interaction effect of moral values on the relationship between affect and evaluation of homosexuals was statistically significant. This means that participants‘ religious commitment and moral values moderated between stereotypic beliefs, symbolic beliefs and affect, on the one hand, and evaluation of homosexuals on the other hand. Therefore, the hypothesis that participants‘ religious commitment, moral values, and personal morals are likely to moderate the relationship between their attitude components and their evaluation of homosexuals was supported to a large extent. Hypothesis 5. The fifth hypothesis stated that participants are likely to evaluate homosexuals more favourably in the post-test condition than in the pretest condition. Table 23 summarizes results of the paired samples t-test performed to test this hypothesis. Table 23 Summary Statistics and Paired Samples T-Test on Evaluation of Homosexuals Evaluation of Homosexuals M (SD) N t df p Pre-test 6.93 (2.32) 179 -4.42 178 .000 Post-test 14.02 (4.12) As can be seen in Table 23, the mean evaluations from the two conditions suggest that participants evaluated homosexuals more favourably in the post-test condition (M = 14.02, SD = 4.12) than in the pre-test condition (M = 6.93, SD = 2.32). This is further confirmed by the fact that the paired-samples t-test revealed a statistically significant difference in the mean evaluations of the two conditions (t(178) = -4.42, p = .000). A pictorial view of the approximate position of the participants‘ mean evaluation on the University of Ghana http://ugspace.ug.edu.gh 108 evaluation thermometer for the pre-test and post-test condition is shown in Figure 10 below. As evident in the figure, although participants evaluated homosexuals more favourably in the post-test condition, their evaluation was only marginally better than that of the pre-test condition. While participants evaluated homosexuals as ‗very unfavourable‘ for the pre-test, they evaluated them as ‗quite unfavorable‘ for the post-test. POSITIVE 100o Extremely Favourable 90o Very Favourable 80o Quite Favourable 70o Fairly Favourable 60o Slightly Favourable 50o Neither Favourable nor Unfavourable 40o Slightly Unfavourable 30o Fairly Unfavourable 20o Quite Unfavourable 10o Very Unfavourable NEGATIVE 0o Extremely Unfavourable Figure 10. Evaluation thermometer showing points corresponding with Study 2 participants‘ mean evaluation scores in the pre-test and post-test conditions. Hypothesis 6. This hypothesis stated that males are likely to evaluate homosexuals more unfavourably than their female counterparts in both pre-test and post- test conditions. Both the Independent samples t-test and One-way Analysis of Covariance (One-way ANCOVA) were used to test this hypothesis as shown in Tables 24 and 25 (see the following page). Post-Test Pre-Test University of Ghana http://ugspace.ug.edu.gh 109 Table 24 Summary of Independent Samples T-Test on Gender Differences in Evaluation of Homosexuals Pre-test Scores Gender M (SD) N t df p Male 5.53 (1.24) 103 78 -1.25 179 .21 Female 8.58 (1.64) p < .05 From Table 24, no significant differences were found between males and females in their pre-test scores on evaluation of homosexuals (t(179) = -1.25, p = .21). Additionally, as Table 25 shows, after holding pre-test scores constant, One-way ANCOVA test results showed no significant gender differences in post-test scores on evaluation of homosexuals (F(1,176) = 1.34, p = .24). The hypothesis was therefore not supported. Table 25 Summary of One-way ANCOVA on Gender Differences in Evaluation of Homosexuals Post-test Scores Source Type III Sum of Squares df Mean Square F p Partial Eta Squared Evaluation of homosexuals (pre-test scores) 14774.11 1 14774.11 36.02 .00* .17 Gender 551.44 1 551.44 1.34 .24 .01 Error 72170.45 176 410.05 Total 122300.00 179 *p < .05 Hypothesis 7. The last hypothesis for this study stated that participants in the biological and biological transgendered groups are likely to have more favourable behavioural intentions toward the vignette persons than participants in the choice, choice University of Ghana http://ugspace.ug.edu.gh 110 transgendered, and teenage pregnancy groups. Table 26 shows results of the one-way ANOVA performed to test this hypothesis. Table 26 The Influence of Experimental Condition on Participants‟ Behavioural Intentions Experimental Condition M (SD) N F df p Biological 2.57 (1.85) 35 8.55 4/178 .000 Choice 2.61 (1.66) 36 Biological transgendered 4.16 (1.37) 38 Choice transgendered 2.75 (1.93) 36 Teenage Pregnancy 4.11(1.66) 38 From Table 26, it is obvious that significant differences were found in participants‘ behavioural intentions (F(4,178) = 8.55, p = .000). To determine the nature of these differences, a Tukey HSD post hoc analyses of the mean scores were performed, the results of which are summarized in Table 27 below. Table 27 Tukey HSD Multiple Comparison of Influence of Experimental Condition on Participants‟ Behavioural Intentions Experimental Conditions 1 2 3 4 5 1. Biological __ 2. Choice .04 __ 3. Biological transgendered 1.59* 1.55* __ 4. Choice transgendered .18 .14 1.41* __ 5. Teenage Pregnancy 1.53* 1.49* .05 1.36* __ *p < .05 As Table 27 shows, significant mean differences were revealed in participants‘ behavioural intentions towards the vignette persons (p < 0.05). No significant differences University of Ghana http://ugspace.ug.edu.gh 111 were found between participants in the biological group and participants in the choice and the choice transgendered groups in terms of their behavioural intentions. Participants in the biological group were however significantly different in their behavioural intentions from those in the biological transgendered group and the control group, that is, the teenage pregnancy group. Significant differences were also found in the behavioural intentions of participants in the biological transgendered group and the choice transgendered group. Participants of the biological transgendered group were however not significantly different in their behavioural intentions from participants in the control group but there was a significant difference in the behavioural intentions of the choice transgendered group and the control group. In effect, participants in the biological, choice, and the choice transgendered groups had similar behavioural intentions towards the vignette persons while participants in the biological transgendered and teenage pregnancy groups also had similar behavioural intentions towards the vignette persons. Drawing on the mean scores shown in Table 26, it is obvious that participants in the biological transgendered group had the most favourable behavioural intentions towards the vignette person (M = 4.16, SD = 1.37), followed by participants in the control group (M = 4.11, SD = 1.66). Participants who showed the most unfavourable behavioural intentions toward the vignette person were those in the biological group (M = 2.57, SD = 1.85), followed by those in the choice group (M = 2.61, SD = 1.66) and the choice transgendered group (M = 2.75, SD = 1.93). Therefore the hypothesis that participants in both the biological and biological transgendered groups are likely to have more favourable behavioural intentions toward the vignette persons than participants in the choice, choice transgendered, and teenage pregnancy groups was not confirmed by the present data. University of Ghana http://ugspace.ug.edu.gh 112 Discussion As was expected, results of Study Two indicated that participants evaluated homosexuals more unfavourably than the other target groups. Plotting this on the evaluation thermometer showed that participants generally gave a rating of ―very unfavourable‖ to homosexuals, indicating that they have negative attitudes toward them, a fact that is consistent with the literature (e.g. Ben-Ari, 2001; Haddock et al., 1993; Lim, 2008; Lim & Johnson, 2001; Owusu et al., 2013). The results also indicated a positive correlation between respondents‘ stereotypic beliefs, symbolic beliefs, and affect which is consistent with Haddock et al.‘s (1993). This implies that positive stereotypic beliefs are associated with positive symbolic beliefs and positive affect, while positive symbolic beliefs are also associated with positive affect. Similarly, negative stereotypic beliefs are associated with negative symbolic beliefs and negative affect while negative symbolic beliefs are also associated with negative affect. Although the correlations among these variables were higher for Haddock et al.‘s (1993) study (.44, .61, .36) than for the current study (.38, .29, .32), they still concluded that the correlations they found were not high enough to suggest that the variables were eliciting information that were completely overlapping. Therefore, it cannot also be said that these variables elicited information that were completely overlapping in this study even though they are positively correlated with each other. With regards to the responses that participants gave in answer to questions regarding their stereotypic beliefs, symbolic beliefs, and affect, the most frequently elicited responses for stereotypic beliefs were that homosexuals are not religious, are immoral, and evil. For symbolic beliefs, participants perceived homosexuals as blocking morality, blocking marriage, and promoting respect for human rights. With respect to participants‘ affect, the feelings or emotions stated as University of Ghana http://ugspace.ug.edu.gh 113 most elicited by homosexuals when they see, meet, or think about them were feelings of anger, unhappiness, and disgust. Considering these responses, it is obvious that morality or the lack of it on the part of homosexuals was stated by participants for both stereotypic beliefs and symbolic beliefs. As already established in Chapter Three, the fact that morality underlies participants‘ feelings of disgust goes beyond mere probability considering that the link between these factors may be the point at which the attitude components partially overlap and considering also that the literature supports this assumption (e.g. Chapman et al., 2009; Olivera La Rosa & Rossellό-Mir, 2013; Schnall et al., 2008). The same assumption can be said to be at play here due especially to the fact that positive correlations were found amongst all the attitude components. This testifies to their interrelatedness and also testifies to the fact that the overlaps can be said to be driven mainly by respondents‘ stereotypic beliefs. Therefore, where participants felt that homosexuals were immoral, they likewise saw them as blocking morality and therefore felt disgusted at them. Such disgust may be strongly related to negative attitudes toward homosexuals (Olatunji, 2008), which attitudes are depicted in participants‘ description of homosexuals as evil and in their feelings of anger towards them (Owusu et al., 2013). It is interesting to note that the third most elicited response under symbolic beliefs for the current study was that homosexuals promote respect for human rights. However, it can still be said that perhaps there is some overlap in the information elicited by stereotypic beliefs and symbolic beliefs here also considering that one of the responses elicited under stereotypic beliefs is that homosexuals are not discriminative. Some participants also said they are friendly, loving, and sociable. With regards to the feeling of unhappiness reported by participants, as established previously in Chapter Two, the homosexuals that students might have seen or might have met may be fellow students University of Ghana http://ugspace.ug.edu.gh 114 and/or relatives with whom they have some affinity. Having this affinity would cause participants to wish that these fellow students and/or relatives were not homosexuals and the reality that they are would result in the feeling of unhappiness. The attitude components‟ prediction of and the moderator variables‟ effect on the evaluation of homosexuals. With regards to the hypothesis that stereotypic beliefs, symbolic beliefs, and affect are likely to predict respondents‘ evaluation of homosexuals, it was found that all the predictor variables together explained twenty-one percent (21%) of the variance in evaluation of homosexuals. The bulk of the variance was explained by stereotypic beliefs and affect. With the addition of the evaluative implications of affect to stereotypic beliefs, it can be said that this finding is consistent with Haddock et al.‘s (1993) finding to a certain extent. Their findings indicated that with the addition of the evaluative implications of symbolic beliefs and affect to stereotypic beliefs, their ability to predict prejudicial attitudes was significantly increased. In the same way, the ability to predict participants‘ prejudicial attitudes toward homosexuals was significantly increased in the current study too even though the 21% variance found here is far below the 40.1% found by Haddock et al. Testing the hypothesis that respondents‘ religious commitment, moral values, and personal morals will moderate the relationship between their attitude components and their evaluation of homosexuals was one way in which it was hoped that an explanation would be found for this large proportion of the variance still yet to be explained in participants‘ evaluation of homosexuals. As seen from the results section above, this hypothesis was confirmed. Although personal morals did not moderate the relationship between any of the participants‘ attitude components and their evaluation of homosexuals, their religious commitment and moral values did. For these two moderator variables, the findings were that where participants held more negative stereotypic beliefs or negative symbolic beliefs University of Ghana http://ugspace.ug.edu.gh 115 about homosexuals and also had high religious commitment or high moral values, they evaluated homosexuals more unfavourably. Also, where participants hold more negative affect for homosexuals and also have high moral values, they evaluated homosexuals more unfavourably. What is even more important is the amount of variance in the participants‘ evaluation of homosexuals that was explained by the interaction effects of religious commitment and moral values. When the interaction effects of religious commitment and moral values were factored in with participants‘ personal characteristics, they explained the variance in their evaluation of homosexuals almost twice as much. It can therefore be concluded that religious commitment and moral values do indeed moderate the relationship between participants‘ stereotypic beliefs, symbolic beliefs, and affect, on one hand and their evaluation of homosexuals on the other hand. Effect of attributions on behavioural intentions and gender differences in attitudes. Answering the above question was the focus of two of the remaining three hypotheses for this study that are discussed subsequently. The first of these predicted that participants are likely to evaluate homosexuals more favourably in the post-test situation than in the pre-test situation and upon testing, this was confirmed. Plotted on the evaluation thermometer, participants‘ mean evaluation of homosexuals was ‗very unfavourable‘ and in the post-test condition their mean evaluation was ‗quite unfavourable‘. As noted previously, this means that participants‘ evaluation of homosexuals was relatively more favourable in the post-test condition than in the pre-test condition. Nevertheless, the post-test evaluation still points to a negative attitude. The second hypothesis dealing with attributions stated that participants in the biological and biological transgendered groups are likely to have more favourable behavioural intentions toward the vignette persons than participants in the choice, choice transgendered, and University of Ghana http://ugspace.ug.edu.gh 116 teenage pregnancy groups. Obviously, this hypothesis also looked at participants‘ behavioural intentions and therefore measures the behavioural component of attitudes. Upon testing the hypothesis, it was found that, contrary to expectations, it was only participants in the biological transgendered and control groups that had favourable behavioural intentions towards the vignette persons while participants in the biological, choice, and choice transgendered groups had unfavourable behavioural intentions toward the vignette persons. Participants in the biological group had the most unfavourable behavioural intentions. As noted previously, Haider-Markel and Joslyn (2008) concluded, based on their study, that the perception of homosexuality as controllable increased negative affect while the perception of homosexuality as uncontrollable increased positive affect towards gays and lesbians and therefore attributions are powerful predictors of people‘s affective reactions toward homosexuals. Therefore, perhaps the reason why participants in the biological transgendered group had the most favourable behavioural intentions while those in the biological group had the most unfavourable behavioural intentions is that while the former saw their vignette person‘s situation as uncontrollable in terms of the fact that the circumstances were beyond his/her control, the latter could not accept that this was so in the case of their vignette person (see Vignettes 1 & 3 in Appendix E). Additionally, Haider-Markel and Joslyn (2008) found that religiousity in terms of increased church attendance, being ―born again‖ and being affiliated to a Protestant denomination, reduced the likelihood of attributing homosexuality to biological origins. Although participants‘ religious affiliations were not ascertained in the current study, their religious commitment was measured. Since it has already been established that when participants have high religious commitment, then they have negative attitudes towards University of Ghana http://ugspace.ug.edu.gh 117 homosexuals if their stereotypic beliefs and symbolic beliefs about homosexuals are also negative, then perhaps this might be another reason for these findings. Boysen and Vogel‘s (2007) findings also offer yet another probable reason. Their findings showed that participants with initial positive attitudes toward homosexuality tended to see biological evidence as reason for acceptance of homosexuality while participants with initial negative attitudes tended to see it as reason not to accept homosexuality. This is the idea of biased assimilation and obviously, the findings of the current study are consistent with this. As was the case with Boysen and Vogel‘s participants, in this study also, participants‘ preexisting attitudes affected their perceptions of biological explanations of homosexuality. Those with initial negative attitudes perceived these explanations as more persuasive evidence against the legitimacy and acceptability of homosexuality and the fact is that all the participants in the current study generally had negative attitudes toward homosexuals. As it was thought that some gender differences might occur among participants between the pre-test and post-test conditions, it was hypothesized that males are likely to evaluate homosexuals more unfavourably than their female counterparts in both pre-test and post-test conditions. Upon testing this hypothesis, no significant differences were found between males and females in their evaluation of homosexuals both in the pre-test and post-test conditions. This is consistent with Anarfi and Gyasi-Gyamerah‘s (2014) finding and as they suggested, it is most likely due to differences in culture. As noted previously, the fact that the Ghanaian culture is different from that of western countries explains the lack of gender differences. University of Ghana http://ugspace.ug.edu.gh 118 Conclusion and Observed Model for Study Two This second study was conducted with the main purpose of determining whether the results of Study 1 will be replicated and also to determine whether attributions will have an effect on behavioural intentions. As has been discussed earlier, findings of this study indicated that participants evaluated homosexuals more unfavourably than the other target groups and that no significant gender differences existed in both the pre-test and post-test conditions. This study also found that a positive correlation existed among all of the participants‘ stereotypic beliefs, symbolic beliefs, and affect. Another finding of this study was that, participants‘ stereotypic beliefs, symbolic beliefs, and affect altogether explained 21% of the variance in evaluation of homosexuals. Additionally, it was found that participants‘ religious commitment and moral values moderated the relationship between their attitude components and their evaluation of homosexuals. It was also found that participants‘ evaluation of homosexuals was relatively more favourable in the post-test condition than in the pre-test condition. Finally, it was found that participants in the biological transgendered and control groups had more favourable behavioural intentions towards the vignette persons contrary to expectation. The question as to why participants in the biological group had the most unfavourable behavioural intentions towards the vignette person, however, still remained unanswered. Based on these findings, the following observed model (Figure 11) was developed. The variables are shown in rectangles of various sizes that are solely meant to accommodate the variable names. The interaction effects between variables are depicted by solid double-tipped arrows having both a diamond tip and a triangular tip at opposite ends. Correlations or associations are depicted by normal double-tipped arrows while solid double-tipped arrows depict a two-way effect. University of Ghana http://ugspace.ug.edu.gh 119 Figure 11. Observed model showing findings of Study 2. Moral Values Affect Attitudes toward Homosexuals (Evaluation of Homosexuals) Symbolic Beliefs Stereotypic Beliefs Religious Commitment Personal Morals Behavioural Intentions University of Ghana http://ugspace.ug.edu.gh 120 Chapter Five Study Three Introduction Study 3 was conducted mainly to determine why results of Study 2 showed participants in the biological group as having the most unfavourable behavioural intentions. It was also to examine how homosexuals view themselves and ascertain whether they would confirm the unfavourable evaluation of themselves by presumed heterosexual students in the previous studies. Unlike the two previous studies, the current study was qualitative. It entailed Focus Group Discussions (FGDs) with presumed heterosexual University of Ghana (UG) students as participants and In-Depth Interviews (IDIs) with known homosexual UG students as participants. This chapter therefore documents the methods used in collecting data, transcribing and analysing them. The findings for both the FGDs and the IDIs are interpreted with supporting relevant extracts from participants‘ transcripts and are finally presented as thematic maps. In all, fourteen (14) themes emerged under the four major topics discussed with the FGDs participants. Additionally, eleven (11) themes emerged from their discussion of the vignettes in terms of the vignette persons‘ circumstances and choices, while nine (9) themes emerged from their discussion of the vignettes imagining they were the vignette persons. For the IDIs, a total of eight (8) themes emerged, four (4) each for the male participants and the female participants. The chapter ends with a discussion of the findings in relation to the literature reviewed in Chapter 2 and by incorporating other relevant literature where necessary. University of Ghana http://ugspace.ug.edu.gh 121 Methodology Population/Sample. The population from which participants were drawn for this study was the same as for Studies 1 and 2. It was conducted in two parts, the first being FGDs with presumed heterosexual students and the second being IDIs with known homosexual students. For the FGDs, one hundred and forty-four (144) participants (72 males & 72 females) were selected through convenient sampling from the population and were interviewed in groups of six (6) each. The selection was done such that each group had at least one student from each of the undergraduate levels (100 – 400). Majority of the participants were Christians, seven (7) of them were Muslims, one (1) of them had no religious affiliation but noted that he believed in God and one (1) identified himself as both a Christian and a Rastafarian. For the IDIs however, only fifteen (15) participants (9 males & 6 females) volunteered to be part of the study although the initial plan was to interview twenty (20) self-identified homosexual students, 10 males and 10 females. Of the nine (9) males interviewed, four of them self-identified as gay or men-who-have-sex-with-men (MSM), three as bisexual, one as ‗humansexual‘, and the last one stated that he did not know his orientation although he admitted that he prefers having sex with men. Those who identified as gay or MSM also alluded to the fact that they were effeminate. According to the urban dictionary, a humansexual person is one who is above all else attracted to humans as human beings and has the ability to make anyone feel attracted to him/her regardless of gender and sexual orientation (www.urbandictionary.com). The urban dictionary also defines effeminate as having characteristics never associated with manliness or men. Additionally, five of the males were Level 400 students (M1, M3, M6, M7, & M8), two of them were Level 300 students (M2 & M4), and the remaining two were a Level 200 student (M5) and a Level 600 student (M8). University of Ghana http://ugspace.ug.edu.gh 122 With regards to the 6 females, one of them stated that she did not like labels, another one stated that she was not sure of her orientation as it depended on who opportunity presented as a sexual partner, and the remaining four (4) identified as lesbians. Of the 4 lesbians, one alluded to being femme while three alluded to being tomboys with one of the latter stressing that she is butch. According to the urban dictionary, a femme is a traditionally feminine woman who is especially attracted to masculine or butch lesbians while a tomboy is a girl who dresses and sometimes behaves the way boys are expected to, including being interested in more masculine things like ―stronger‖ sports, computers, or cars. A butch female does not only behave as a boy sometimes but is an overtly and stereotypically masculine or masculine-acting lesbian (www.urbandictionary.com). Additionally, two of the females were Level 400 students (F4 & F6), another two were Level 200 students (F1 & F2) and of the remaining two, one was a Level 300 student (F3) and the other was a Level 600 student (F5). As the IDIs participants are a special group and therefore do not easily make themselves known, a key informant (male) was used to get access to them. The key informant was a former student of UG and works with the Centre for Popular Education and Human Rights – Ghana (CEPEHRG), a Non-Governmental Organization (NGO) committed to combatting the effects of the AIDS in Ghana and offers support to homosexual persons and other at-risk populations. Interview Guide. Two interview guides were used for this study, the FGDs guide and the IDIs guide. The FGDs guide was self-developed and consisted of three sections with open-ended questions. Section A comprised questions about Ghanaians‘ religious commitment and morality, with an emphasis on the UG students‘ level of religious commitment and morality and that of participants. Section B dealt with participants‘ awareness of homosexuality on campus and their thoughts and feelings (stereotypic University of Ghana http://ugspace.ug.edu.gh 123 beliefs, symbolic beliefs, & affect) towards it. For the last section, Section C, the questions focused on the content of Vignettes 1 – 4, which are the same as used in Study 2. Each vignette was therefore given to participants to read before questions were asked about them. For some of the groups, participants were asked what they think about the vignette person and how they feel about the whole scenario presented by the vignettes and for the others, participants were asked to imagine they are the vignette person and talk about what they will do and why, given the same circumstances (see Appendices C & E). The IDIs guide was also self-developed and comprised open-ended questions in two sections. Section A dealt with participants‘ sexual orientation and the experiences relating to it such as self-disclosure and societal reactions while Section B dealt with participants‘ awareness of other homosexual students on campus and how they are treated by presumed heterosexual students and the general public (see Appendix D). Design and Procedure. This study is qualitative and was conducted with the help of three Research Assistants (RAs), two of whom were teaching assistants at the Department of Psychology and the third person being a graduate student with the same department. As with Study 2, there was no need to seek ethical clearance again for this third study since it had already been sought as part of the first study. Both instruments were pretested and necessary corrections made before actual interviews were done and all the interviews were tape-recorded. To get students to volunteer to participate in the FGDs, the RAs approached them either during tutorials at their various departments or at their halls of residence, informed them of the study, and asked if they would be interested in participating at all. In this manner, first names, levels, room numbers, and telephone numbers were obtained from students with the directive that they would be notified of the time and venue for the discussions. In all, more than 180 students were sought in this University of Ghana http://ugspace.ug.edu.gh 124 manner so as to ensure that the number of participants required would be obtained even in the likely event that some of them would drop out. Participants were informed about the time and venue two days prior to the scheduled discussion date and any six (6) students who agreed to come were again reminded of the interview on the day itself to ensure they would come as promised. Each session involved interviewing 6 participants who fell into three main types of groups, including an all-male group, an all-female group and a mixed group (3 males and 3 females). These groups were further categorized as ‗A‘ groups or ‗B‘ groups depending on the questions asked after participants had read the appropriate vignette for their group. A total of 24 FGDs were conducted, 12 categorized as ‗A‘ and the other 12 categorized as ‗B‘. Table 28 shows details of the groupings. Table 28 Summary of Groupings Groups (6 Participants per group) Group Differences FGDs guide with Vig. 1 FGDs guide with Vig. 2 FGDs guide with Vig. 3 FGDs guide with Vig. 4 Group A: Discussion of issues relating to vignette person. All male All female Mixed All male All female Mixed All male All female Mixed All male All female Mixed Group B: Discussion of vignette person imagining it is one‘s self. All male All female Mixed All male All female Mixed All male All female Mixed All male All female Mixed Total (Groups & Participants) 6 groups, 36 participants 6 groups, 36 participants 6 groups, 36 participants 6 groups, 36 participants Note. ‗Vig.‘ means Vignette. Although the initial plan was for the process of conducting all 24 FGDs to last for not more than two weeks, the actual process spanned several months. This is because there were difficulties with getting an appropriate venue that was relatively soundproof University of Ghana http://ugspace.ug.edu.gh 125 with no likelihood of being interrupted for the FGDs on UG campus. There was therefore the need to rent an office off-campus at Okponglo, a community close to the UG campus as venue for the FGDs. As a result of this short distance of the venue from campus, potential participants came up with all kinds of excuses when called with information about the scheduled dates for the discussions and while some of them gave their excuses the very first time they were called, others waited till they were called on the morning of the set date. Such situations meant rescheduling the discussion to a time when the required numbers of participants were available. This meant an increase in drop-out rates than was expected but having contacted a lot more students than were actually needed as participants, the drop-out rates did not affect the sample size. On the set dates for the FGDs, two of the RAs met three (3) participants each at designated spots on campus and brought them to the venue in a taxi while the third RA, the graduate student, helped with setting up the venue for the discussions and served as an observer during the discussions. When participants arrived at the designated venue on the scheduled dates for the FGDs, they were again informed about the nature of the study and the fact that it will be recorded. Each participant was then given two copies of the individual consent form 3A (see Appendix F) to sign and return one copy. After this was done, participants were reminded that they have the right to discontinue participation at any point during the discussions if they so wish. However, no participant discontinued participation in any of the discussion sessions. Each tape-recorded session lasted for an hour averagely and participants were given soft drinks and snacks as reciprocity. With regard to the IDIs, dates and timelines could not be set because the interviews depended on the availability of the known homosexuals who volunteered to be part of the study. Interviews were conducted at the same venue as the FGDs and also spanned several months. Compared to the female homosexuals, it was relatively easy to make University of Ghana http://ugspace.ug.edu.gh 126 arrangements to interview the male homosexuals as the key informant is male and was friends with some of them while he was a student of UG. For both groups (males & females), their concern was the fact that the interviews were to be tape-recorded. Their fear was that the interviews might be aired on radio and their identities would be known. The key informant was able to assuage this fear on the part of the male homosexuals however. Despite this, only two of them actually came for the interview while the remaining seven (7) declined when informed by the key informant that they were to be interviewed by a staff/student of UG. The two who came for the interviews did so because they did not read courses in the social science faculty and so felt that the chances of being identified on campus were negligible. To solve this problem, it became necessary to train someone else to interview the male homosexuals and taking their need to remain anonymous into cognizance, it was imperative to use another former student who shares their orientation as the interviewer. This person was a friend and co-worker of the key informant who by virtue of his job already had experience in conducting interviews with homosexual persons for their organization. Upon his agreement to conduct the interviews, he was given a day‘s training so he could familiarize himself with the interview guide and understand the purpose of the study. This training session ended in interviewing him too so he would get a practical experience of how it should be done. After the training, arrangements were made for him to conduct his first interview and another session was held with him where the interview was played to assess how it was done and some minor problems addressed. Subsequently, the remaining interviews with the male homosexuals were conducted by him successfully. All the interviews were conducted one-on-one and tape-recorded with only the interviewer and participant present. Upon arrival at the venue, participants were reminded about the purpose of the study and assured of confidentiality. After this, they were each University of Ghana http://ugspace.ug.edu.gh 127 given two copies of the individual consent form 3B (see Appendix F) to fill and return one copy. Before the actual interview started, they were again reminded that they had the right to discontinue participation at any point during the interview process. None of the participants discontinued participation after the start of the interviews. Reciprocity for them was the same as for the FGDs participants. With the female homosexual students on the other hand, getting volunteers to interview was much more challenging. The key informant had only one female contact on UG campus and getting to meet her and convince her to be interviewed herself and get her to convince fellow students to agree to be interviewed was very difficult. The initial drawback was the fact that the interviews would be recorded but her fears were eventually assuaged when informed that the study had been cleared by the IRB of the Noguchi Memorial Institute of the University of Ghana and copies of the approval letter and consent form were shown to her. It was also agreed that she and any other female student who will volunteer to be interviewed would receive twenty (20) cedis as reciprocity. Based on this understanding she agreed to be interviewed and helped convince the other five (5) female homosexual students to volunteer to be interviewed also. Unlike the male homosexuals, the females expressed no misgivings about being interviewed by a staff/student of UG. The procedure used in conducting the interviews with the female homosexuals was the same as was used with the male homosexuals. Data Organization and Analyses of Interviews Transcription. Tape-recorded FGDs and IDIs were transcribed verbatim and included both verbal and nonverbal (e.g. laughter, false starts, pauses, use of Twi language) so as to ensure that the transcripts retained the information needed in a way which was ‗true‘ to its original nature (Braun & Clarke, 2006). The questions posed to University of Ghana http://ugspace.ug.edu.gh 128 participants and their answers were typed as separate paragraphs with the responses italicised. Transcripts of the FGDs were named based on the group type that was interviewed (see Table 27). So, for example, the transcript of FGDs with a group of participants who are all female, were given Vignette 1 to read and afterwards discussed issues relating to the vignette person was named ‗Female Group 1A‘. Transcripts of the IDIs were named using the participants‘ genders and a number depending on the order in which they were interviewed. The first male interviewee was therefore named ‗M1‘ while the first female interviewee was named ‗F1‘ and so forth. These names were also used as tags for all extracts taken from the transcripts so as to differentiate between extracts. The transcripts and extracts were so named to ensure anonymity and confidentiality especially for the IDIs participants. Analyses. The data corpus made up of all transcripts from the FGDs and the IDIs were analyzed using Thematic Analysis as advocated by Braun and Clarke (2006). According to them, thematic analysis should be seen as a foundational method for qualitative analysis, the first method that researchers should learn, as it provides core skills that will be useful for conducting many other forms of qualitative analysis. They noted also that it is a relatively straightforward form of qualitative analysis which does not require the detailed theoretical and technological knowledge that approaches such as grounded theory and discourse analysis do. For this reason, thematic analysis offers a more accessible form of analysis particularly for those in the beginning stages of a qualitative research career. Braun and Clarke further noted that although thematic analysis is often thought of as something that is simply carried out by someone without the knowledge or skills to perform a supposedly more sophisticated form of qualitative analysis, a rigorous thematic approach can produce an insightful analysis that answers particular research questions. They pointed out that it is more important to choose a University of Ghana http://ugspace.ug.edu.gh 129 method that is appropriate to a study‘s research questions, ―rather than falling victim to ‗methodolatry‘‖ (p. 97), which makes one more committed to method rather than to the topic content or research questions. However, Braun and Clarke (2008) additionally stated that thematic analysis can be conducted within a realist/essentialist paradigm or a constructionist paradigm. The latter was chosen as the most appropriate for this study since, according to them, ―thematic analysis conducted within a constructionist framework cannot and does not seek to focus on motivation or individual psychologies, but instead seeks to theorize the sociocultural contexts, and structural conditions, that enable the individual accounts that are provided‖ (p. 85). The analytical strategy for this study therefore followed the six (6) phases that Braun and Clarke proposed and involved the following: 1. Getting familiar with the data – This first phase involved repeated reading of the transcripts for both the FGDs and the IDIs so as to be ―familiar with the depth and breadth of the content‖ (Braun & Clarke, 2006, p. 87). Since the transcriptions were done with the help of the one graduate student RA, there was the need to read the transcripts repeatedly even though a large part of the data was already familiar from having transcribed them personally. This process was made less laborious by the fact that the transcriptions done by the RA were checked against the original tape-recordings for accuracy. The repeated reading also helped with searching for meanings and patterns in the data, noting them down, and marking ideas for subsequent coding. 2. Generating initial codes – This second phase included the process of manually organizing the data into meaningful groups in a ―theory-driven‖ (Braun & Clarke, 2006, p. 88) manner in terms of the fact that coding was done around the major topics that were discussed in the FGDs and IDIs (see Appendices C & D). All data University of Ghana http://ugspace.ug.edu.gh 130 items were read individually and the codes within them were noted on ‗post-it‘ papers with different coloured pens depending on which discussion topic it was related to. In this process, the data items were coded for as many potential themes as possible. The codes were also matched with data extracts using the same coloured pens to indicate which extracts are linked to the same codes. To ensure that data extracts can be easily traced back to the transcripts from which they were extracted, they were coded inclusively, leaving a little of the surrounding data. 3. Searching for themes – This phase involved sorting out the many different codes identified into potential themes. The first step to achieving this was to group codes together by writing those that could form a potential theme on separate sheets of paper. These codes were then typed using different coloured fonts to separate one group from the other. Each group of codes was read over and over while thinking about the appropriate name for the potential themes and sub-themes they form. The initial themes and sub-themes derived from this process were typed also and the relevant data extracts collated under each of them. 4. Reviewing themes – This involves refining the initial themes derived from the third phase. Firstly, all the themes and the extracts related to them were read again and those that did not have enough data to support them were discarded while some were collapsed into one theme. This helped to ensure that all remaining themes each had data that were linked together meaningfully and were distinct from each other (Braun & Clarke, 2006). An initial thematic map was arrived at after this first process. Secondly, each theme was weighed against the entire data set to ascertain their validity and to determine whether the initial thematic map accurately represented the meanings evident in the data set as a whole. Any University of Ghana http://ugspace.ug.edu.gh 131 additional data within themes that had been overlooked in the earlier coding exercise were coded at this stage and added to the thematic map. 5. Defining and naming themes – This fifth phase involved defining and refining the initial thematic map. Each of the themes and their collated extracts were read again and organized into ―a coherent and internally consistent account with accompanying narrative‖ (p. 92). In this process, each theme was considered in relation to the other themes and each of them was checked for whether it contains subthemes that need to be identified. The names given to the various themes were also considered at this stage and those that needed changing to make them more representative of what aspect of the data set they referred to were changed. At the end of this phase, the final thematic map relating to each major topic discussed with the participants had emerged. 6. Producing the report – according to Braun and Clarke (2006), this final phase involves the final analysis and write-up of a report on the thematic analysis done from phases 1 to 5. The following therefore depict what this phase involves and thereby show what the findings of this study are. Findings Focus Group Discussions. As noted earlier, during the data analysis, coding was done around the major topics that were discussed in the FGDs and therefore the themes that finally emerged were also directly related to the major topics discussed. The topics included: 1. Religiousity and Religious Commitment 2. Morality 3. Homosexuality University of Ghana http://ugspace.ug.edu.gh 132 4. Vignettes For each of these topics, certain themes and sub-themes were derived that were represented in each of the transcripts. The following are the interpretation of these themes and sub-themes. Extracts used to clarify the themes and sub-themes are quoted verbatim as was transcribed with the literal translations of all Twi phrases or sentences added in brackets. The interpretations under each topic are immediately followed by a thematic map that summarizes the themes and sub-themes for the topic. Religiousity and religious commitment. Four themes were derived from analyses of participants‘ transcripts in relation to this topic and these are interpreted one after the other subsequently. They include:  Religion permeates all aspects of life. Participants generally agreed that religion permeates all aspects of the lives of Ghanaians such that they are unable to do anything, including everyday things such as responding to greetings and discussing issues such as football scores and the politics of the day without linking it to religion somehow. Participants also noted that this permeation is evidenced by the fact that national programmes are usually started with prayers from representatives of the major religions within the country. The following extracts depict this: “Ghanaians are notoriously religious because, I mean we seem to attach religion to every single thing we do. Even if the thing is totally devoid of religion we find a way of infusing it somewhere. When we talk of football we try to bring religion in it. Even politics, someone‟s death is caused by something religious so we are very religious.” (Male Group 3A) “I think that Ghanaians are religious because you meet someone and greet them and they say „Nyame adom me ho yɛ‟ [ By God‟s grace I‟m fine] or University of Ghana http://ugspace.ug.edu.gh 133 something…every aspect of our lives, education, family, everything…when you go to people‟s homes you see pictures of whoever they believe in.” (Mixed Group 1B) “During our national celebrations, for instance when we go for our independence celebrations each and every year, you realize that most of the programmes that we do to start, we call and the different sets of religion will come and pour their various libations and to pray…So, on the whole you could see that religion really plays a vital role or has an influence on Ghanaians.” (Male Group 2A)  Religiousity as means for personal gain. Majority of participants agreed that although Ghanaians are religious, most of the time their religiousity is only a means for personal gain. They indicated that most students spend a lot of their time on campus partying but are especially more religious when examinations are approaching. They also indicated that some individuals go to church because they are interested in getting romantic partners as the following extracts depict: “…Me personally going to the school I won‟t say students are religious. Why because if you are saying…if you are talking about religiousness, if you are talking about only going to church on Sunday as being religious, I won‟t take that as being religious in my own this thing because most at times you see students…they will go for parties and all sort and sort and sort but when it gets closer to exams time you see them carrying their Bible. It comes to Sunday they all raise their Bible and they start going to church, behaving as if they are very holy. So me I don‟t see that as being religious.” (Male Group 4B) University of Ghana http://ugspace.ug.edu.gh 134 “For personal gains…a guy has seen a girl in the church…there is one guy in my church like that. Because of me he comes to church almost every day. When I come to school and I don‟t go, he also doesn‟t go to church. Immediately I go home, he will start coming to church. Somebody like that when you are sitting back you think he is committed but in his mind if I don‟t go to church, he too he is not going.” (Female Group 3A)  Religious Commitment means more than attendance. This theme had two sub-themes, the first being that religious commitment is a lifestyle or way of life and the second being that religious commitment is related to religious affiliation. Basically, what the majority of participants were pointing to is the fact that being committed to one‘s religion means more than just attending religious gatherings regularly. It means upholding the tenets of the particular religion such that it becomes part of one‘s lifestyle. According to the participants, compared to Christians, Muslims and practitioners of the traditional religions are those who depict lifestyles that are obviously based on the tenets of their religion. Note some of their observations: “I believe some students are committed but I don‟t believe majority of them are committed because we always go to church and we see them there but then after church when we are relating with them in our halls you realize that the lifestyle and some of the things they say and do is different from or doesn‟t reflect that of a committed Christian or Muslim. At least some are but majority of them are not.” (Female Group 1B) “As for the Christians, I think the issue of hypocrisy is more embedded in Christianity than any other religion, because the Christian is going to church just for the sake of going to church. But then when you go to the Islamic University of Ghana http://ugspace.ug.edu.gh 135 religion and the traditionalists, especially Islam, I respect the religion to some extent because of their devotion. There are certain things they wouldn‟t do. You see a larger chunk of Islam ladies dressed up modestly. Even if they would wear the trousers, the top is like way down to mid-thigh. The Christian woman will wear very tight trousers and a very skimpy top but then she goes to church. I think we take things for granted because of the grace of God. We take advantage of the fact that God is merciful.” (Mixed Group 1B)  Religious Commitment very low. Participants also agreed that religious commitment is generally very low amongst Ghanaians, especially young people, in spite of the fact that they are religious in terms of attendance at religious gatherings. Most of them noted that the low religious commitment amongst young people, including students, is as a result of not really wanting to be part of a particular religion but having to attend their gatherings because they are forced to by parents and because they feel obligated to do so in order to conform to societal expectations. As they reported, “Some of us are forced to go to church. In our homes our parents would tell you that „if you don‟t go to church I‟ll not give you food‟. So you don‟t have the will but still you are forced to do it. Maybe you are not independent so you have to accept that. So the commitment level is very low. We are forced to do it.” (Male Group 2B) “It‟s like a societal obligation. You have to belong to at least one of them. If it‟s in your locality you have to go so that people will know you go to church. When Sunday you don‟t go to church they will be like „as for these people that is how they are‟. So it is a societal obligation that you should belong to one University of Ghana http://ugspace.ug.edu.gh 136 but they don‟t really practice what they listen to at church. There are so many rules that are governing us, don‟t do this, don‟t do that…but we find ourselves doing exactly what we are not supposed to do.” (Female Group 3A) Figure 12 summarizes the themes and sub-themes under this topic (see the following page). The topic is shown in oval, the themes are shown in rectangles, and the sub-themes are shown in round edge rectangles. University of Ghana http://ugspace.ug.edu.gh 137 Figure 12. Thematic map showing summary of themes and sub-themes under ‗Religiousity and Religious Commitment‘. Religiousity and Religious Commitment Religion permeates all aspects of life 1. It‘s a Lifestyle/Way of life 2. It‘s related to Religious Affiliation Religiousity as means for personal gain Religious Commitment means more than attendance Religious commitment is low University of Ghana http://ugspace.ug.edu.gh 138 Morality. With regard to this topic, six themes were derived from analyses of participants‘ transcripts and these are interpreted one after the other subsequently. They include:  Defining Morality. Participants defined morality and based on their definitions, three sub-themes emerged. The first was that morality is defined by personal action. As one participant in Mixed Group 4B stated, it is “…what your conscience tells you that you are doing that is not good that means you are going wayward.” The following extract shows this idea even more clearly: “I also want to say that I think personally when God created all of us, he put in us conscience. Whether you are an atheist, whether you are an Islamic believer or whatever it is, you have a conscience that tells you that this is right, this is wrong. So even if you have no belief like an atheist, you don‟t believe in God, you don‟t believe in Allah, you don‟t believe in anything, you still have your conscience working for you. Something will just tell you that this is wrong and this is right.” (Female Group 3B) The second and third sub-themes were that morality is dictated by society and is dependent on religion respectively. One participant in Female Group 2A captures the focus of these two sub-themes with her statement that morality refers to “the dos and don‟ts of the society, the restrictions, the norms, accepted behaviours by the society or by whatever religion you believe.” The following extracts explain these sub-themes further: “I think I would say it‟s the principles, standards and values or norms that society defines as right. So if you go to a certain community what they say as moral when you go to a different society it might be different so what that University of Ghana http://ugspace.ug.edu.gh 139 particular society defines as standards and norms which are right that is what I think that society would say is moral but if you also go to a different society and they define it differently or maybe different standards and norms and they would say that maybe this is moral values or standards. So I think it‟s something that the society deems it as right” (Male Group 1A). “I‟ll say morality is the secular arm of religion in the sense that, you know, if you are just to pick out the codes of conduct prescribed by each and every religion, you‟ll get morality. Morality tends to be more universal. Everybody appreciates certain ways of living, certain standards, and it‟s repeated in almost all the religions I know of. They all condemn specific stuff, and they mostly endorse certain stuff. So I think it is the link between most of these various religions, devoid of their religious backing, can be termed as morality.” (Male Group 2A)  Exam Malpractice common. A number of moral issues were discussed with participants including the issue of examination malpractice (see Appendix C) and what emerged is that it is very common. Two sub-themes also emerged which point to the fact that participants do not see examination malpractice as immoral but as a form of helping behaviour that is quite normal among students. According to them, “Oh, cheating is help. What is wrong with me helping my friend in an examination hall?” (Female Group 3B) “I don‟t know for the rest of us but I will say that we all cheat. It‟s something we all do. We are limiting it to the examination hall but it‟s not only in the examination hall. For instance, we have something we call dubbing. Those of University of Ghana http://ugspace.ug.edu.gh 140 us who do practicals, we have something we call dubbing, you know, I will go and check something from the net, we all come to the practical room and, you know, we call ourselves together, the book will be there and someone will come and take the whole book and „hey, kyaley [slang meaning „my friend‟] give me five minutes‟, you know. He will go and copy everything and bring it back to you and somebody else will come for it. By the time you realize they are all writing the same thing. It‟s not like somebody else will come to pick it and you didn‟t like it, you know, it‟s like „kyaley, take it, take it, kyaley, take then copy then take give me‟…we call it scanning or photocopy…and you just bring it back. [Mentions name of department] for instance, if somebody has their practicals before you, you can arrange with the person. After his practicals, you go for his books, oh seriously, because we all do the same thing so you go for his practical book, see the work he did, and correct all the mistakes, you know, before you go to the practical room. It‟s cheating, it‟s cheating. Why should you go for somebody‟s marked work, look into it, copy everything and you go and take all the marks and the person is not offended at all? He is alright. He is even happy that you have been able to score all the marks, you know, he is even happy for you. It‟s normal.” (Male Group 4B)  Morality Low. Participants noted that morality is low among Ghanaians generally and especially among young people including students. Considering the four sub-themes that emerged under this theme, participants felt that the low morality was as a result of westernization and modernization as the following extract depicts: University of Ghana http://ugspace.ug.edu.gh 141 “Our morals are low now. It is low in terms of what we do, what we practice and like, right now Ghanaians, I think we are westernizing. Modernization, globalization, technology started plunging us down the moral lane. I‟m not that old but there are some things people wear now that you can‟t wear back then. But the issue is when you wear clothes like that back then you were called a certain name. You were branded. We are modernizing; we are approaching more into the western parts. We copy things blindly without knowing the implication and all those things so I think we are morally down.” (Female Group 2A) Apparently, due to this low level of morality in Ghana, premarital sex is now seen as normal, extramarital affairs have increased and alcohol and hard drugs use have become commonplace. With regard to premarital sex, which was the second sub-theme, participants said that not being a virgin was the new ‗normal‘ and that virginity is only proof of not having had the opportunity to have sex. So, not only do students engage in premarital sex but can go to the extent of having sex in their campus rooms when the opportunity presents itself even when another roommate is around. As participants said, “Premarital sex is now a normal thing. Now they tell you that virginity is not a pride. It is lack of opportunity for you to…that is what they will tell you. Even if you have never had sex before they see you as wo koraa woy3 „john‟ bi saa no [like a fool]. People see it as like, either guys don‟t like you or you are weird or you don‟t like guys. You see at first, being a virgin was a pride. Our grandmothers and mothers...eii, you are a virgin. Now some people are even shy to voice out that they are virgins. Now the normal thing is rather you are not a virgin. The abnormal one is you are a virgin.” (Female Group 3B) University of Ghana http://ugspace.ug.edu.gh 142 “I think premarital sex as at now is…we the youth accept it to be common, to be normal now, to be acceptable within ourselves. So it‟s normal you get a girl, I get a girl and if I see…if you have a date and you are going out with the person without sex „Ah, for all this time? Oh you make lose, you make lose. Even this one last week pɛ [only] I follow am I go do the something. And you‟ll have a situation where there are four people in a room. Maybe two roommates are not around. You are around and the other one is with her boyfriend. They don‟t mind that you are around and they would be having sex in your presence. Four people in a room, bunk bed, let‟s say you are on the top bed and the person is on the down bed. [Interviewer interjects asking if it‟s in the presence of the roommate] Yes, in your presence. They cover themselves with a cloth. If you‟re comfortable and you can stay, fine. If you can‟t stay, just get out.” (Mixed Group 1B) With the issue of extramarital affairs, participants stated that such affairs are not only on the increase but contribute to premarital sex being normal now and although men were known for such affairs in the past, now women also engage in them. According to participants, “Now that too is on the ascendency. Now guys of today , even in our Ghanaian society…okay, in the olden days we thought a man can have more than one wife, or maybe if you are married you can have an extramarital affair, it‟s normal. But now women too have joined. Now married women don‟t care okay. If you do it she will also do it. They don‟t care about it. We know, we see, we hear, and it is true. Especially if you marry and you travel University of Ghana http://ugspace.ug.edu.gh 143 for some months or years, by the time you come na y‟ayɛ wo pasaa serious [you would have been cheated on over and over again].” (Male Group 3B) “Yes we have a lot in Ghana here. You know that even contributes to the premarital affairs because due to extra marital affairs the sugar dada [daddy] involves with the young lady who is not married and they are dating, doing their own things, so they are two things. Wherever there are premarital affairs, there are extramarital affairs. Sometimes the sugar dada will be the one the lady is having premarital affair with but to him it‟s extramarital.” (Mixed Group 3B) With regard to the commonplace nature of alcohol and hard drugs use, participants had these to say: “In the hostels it‟s common. It‟s just like drinking alcohol nowadays. It‟s bad. It‟s that bad. At first they used to go and hide in the bush and do it but now they do it in the open…wee and the cigar…it‟s very common on campus.” (Mixed Group 3A) “Now what is happening is that a lot of the time when they are having drink- ups, they mix it with the drinks. We took some the last time. There was wee inside.” (Male Group 2A)  Engaging in Immoral Acts. This theme emerged from participants‘ discussions on why they would engage in certain immoral acts such as not paying for items bought, keeping overchange, and neglecting to pay bus fares. Two sub-themes also emerged that pointed to the fact that engaging in such immoral acts or not are dependent on personal principles and on University of Ghana http://ugspace.ug.edu.gh 144 circumstances such as distance, not having money, and feeling cheated. On the former, one participant said, “Personally, I would give it back. I don‟t know why. I would give it back no matter how much the money is. I don‟t know, my conscience doesn‟t allow me to take the money.” (Female Group 3B) On the latter, here are what some participants also had to say: “That would depend on my pocket because if I don‟t have money I count it as a blessing. So it depends on the way you look at it because if I don‟t have money and the mate doesn‟t take the money I count it as a blessing. If I don‟t have money and I go and buy something and I‟m willing to pay and you give me overchange, I count it as a blessing. So it depends on the way you look at it” (Male Group 1A). “If I‟m going to somewhere like Jean Nelson, I‟ve bought something from bush canteen, I‟m now at ISH [International Students‟ Hostel], going back will be a problem. And I won‟t take it the next day because if I take it the next day the person will not remember what happened so I‟ll just have to keep it.” (Female group 1B) “For me if you don‟t ask me I won‟t give it to you. My conscience doesn‟t prick me that it is wrong. The reason why my conscience doesn‟t prick me is that sometimes they extort from us on campus. I quite remember there was this woman who sold drinks. The drinks go for 2 cedis and she would sell it for 3 cedis…I give her 2 cedis but then I tell her that I gave her 5 cedis. So I‟ll take the drink and take change again…My conscience didn‟t prick at all. She had extorted from lots of people.” (Mixed Group 1B) University of Ghana http://ugspace.ug.edu.gh 145  Bribery and Corruption commonplace. This theme simply pointed to the fact that bribery and corruption is so commonplace in Ghana that it permeates all aspects of Ghanaians‘ lives as much as religion does. Participants stated that it happens even in homes and churches as the following extracts depict: “It‟s very, very high and it‟s seen in almost every aspect of our lives, not just in the workplace but then even in the home and it‟s gaining root in one way or the other in the society and everywhere. My church for instance after every two years we choose leaders. You know because you are affiliated to somebody even if you are not up to merit to be a leader you will still be chosen because you are associated to the head pastor…Some people will just look at you and because you are handsome, you have a certain build, they feel like you can become a pastor so go and be a pastor, you can be a leader so go and be a leader. Because you are rich they think that you can be the fundraising chairman so go and be the fundraising chairman and it‟s really killing the church.” (Male Group 4A) “It‟s so high. You can‟t rate it because it‟s part of us. It has become part of us. It has become normal. Let‟s say I‟m going to SHS [Senior High School], and my parents know that my IQ level is nothing to write home about…so whiles going they would take money, and it‟s likely she would take the money. So they too, those who are taking it they expect you to bring it…Even with the police and the drivers…there is this particular barrier, when you get there, the policeman, you just put it down and you pass by; he wouldn‟t even talk…If you University of Ghana http://ugspace.ug.edu.gh 146 don‟t put that thing down it means you are not leaving. Even in the churches…So I think corruption is everywhere.” (Female Group 1A)  Homosexuality is unacceptable. Homosexuality was also discussed as a moral issue and this theme with its four sub-themes emerged from analyses of transcripts of participants‘ responses. The first sub- theme pointed to participants‘ view that homosexuality is not a human rights issue. They had these to say: “That thing about gay rights…ok so if you are a kleptomaniac and you go and steal, we all know that this is what is wrong with you so do you still protect their this thing and say because they are this they can just steal? Now in our laws of Ghana, if you are 18 and over you can think for yourself. Now you wake up one day and say that I‟m tired of this world so I‟m going to kill myself and you are caught, you are going to be sent to the police because you don‟t have the right to take, even your own life, you don‟t have that right ok? So, homosexuals don‟t have any right in Ghana so we can‟t protect their rights, simply put.” (Male Group 3B) “Personally, though I‟m educated and maybe people will think with this kind of…the way the world is moving we should respect our rights. I don‟t go by that. Homosexuals don‟t have any right in Ghana. Even left! They don‟t have any right or left in Ghana. We can‟t accept them, yeah, and it‟s something we should all…no matter your level of education or whatever we shouldn‟t even talk about that thing they are talking about. Those who are talking about rights it wouldn‟t work in Ghana here. They don‟t belong to even Africa, even if Africa agrees, not in Ghana.” (Male Group 2A) University of Ghana http://ugspace.ug.edu.gh 147 The second sub-theme suggested that unlike other behaviours that can be considered immoral such as keeping overchange, homosexuality has implications on human life. This is because homosexuality blocks marriage and procreation in the traditional and religious (Christian) sense and prevents society from growing. As participants put it, “Because everywhere…in the bible God says come and multiply. How can you multiply with your fellow male? It‟s basically not correct. Then it means you don‟t want the society to grow.” (Female Group 3B) “We believe that marriage is the legal or customary union between a man and a woman. That is how we define marriage in Ghana, that‟s how we define it. It‟s a legal or customary union between a man and a woman with the idea of procreation so once you are getting married and there is no procreation attached to it then it‟s not marriage. That is why even when we are in a marital home and you have a situation where the woman is barren the man‟s family can force her out or else it‟s not marriage because there is an idea of procreation. If it‟s a union between a man and a woman then surely a man and a man goes against it; then surely a woman and a woman goes against it.” (Male Group 4B) “…Fine, maybe we are all guilty of maybe keeping overchange and all that stuff but that thing it doesn‟t have any serious implications on human life but this is something that if we are to hold on to it and everybody is holding on to it, I think a time would come this 25 million population of ours would be like 5million.” (Male Group 1A) University of Ghana http://ugspace.ug.edu.gh 148 The third and fourth sub-themes in turn pointed to the fact that homosexuality is disgusting and against the Ghanaian culture. With regards to it being disgusting, participants referred mainly to the idea of anal sex between homosexual men, saying: “Disgusting! You just imagine a guy, excuse me to say, inserting his penis into another guy‟s anus. It‟s really disgusting. Just get that image in your head, like where I ease and you are…I think it‟s not something which is pleasant. It‟s not good at all.” (Mixed Group 1B) “Disgusting. I used to imagine what they‟ve been doing. How the act is done. It‟s just disconcerting thinking about what they do, like just to find pleasure. You know, it‟s beyond…it‟s contrary to what we believe in.” (Female Group 2A) With regard to it being against our culture, participants‘ sentiments were that it goes against the essence of being a Ghanaian as Ghanaians uphold the natural laws put in place by God and they also value family. Here is what they said: “Because it‟s against our law, and not just the law it‟s against whatever Ghanaians stand for. I don‟t know for the rest of us here but that is what I believe. It‟s against whatever the Ghanaian stands for. We stand for the fear of God. I believe a lot of Ghanaians fear God even those who don‟t go to church. We stand for the obedience to customs and the natural laws that God has put in place.” (Male Group 4B) “But culturally it‟s not acceptable. We have a culture. Our culture does not entertain such things so why should you adopt someone‟s culture. And culture is for family. If your father or your mother knows that you are doing that, they University of Ghana http://ugspace.ug.edu.gh 149 will try to talk to you; they will try to tell you that traditionally our ancestors and our gods….they will call all of them. And normally those who do it they hide it. Why do they hide it?” (Mixed Group 3A) Figure 13 summarizes the themes and sub-themes under this topic (see the following page). The topic is shown in oval, the themes are shown in rectangles, and the sub-themes are shown in round edge rectangles. University of Ghana http://ugspace.ug.edu.gh 150 Figure 13. Thematic map showing summary of themes and sub-themes under ‗Morality‘. Morality Defining morality 1. Result of westernization and modernization 2. Premarital sex is normal 3. Extramarital Affairs on the increase 4. Alcohol/Hard drugs use commonplace Exam malpractice is common Bribery and corruption is commonplace Homosexuality is unacceptable Morality is low Engaging in immoral acts 1. It‘s not a human rights issue 2. Has implications for human life 3. It‘s disgusting 4. It‘s against our culture 1. Dependent on circumstances 2. Dependent on personal principles 1. Cheating is helping 2. Cheating is normal 1. Defined by personal action 2. Dictated by society 3. Dependent on religion University of Ghana http://ugspace.ug.edu.gh 151 Homosexuality. Four themes emerged with regard to this topic (see Figure 11). Each of these is interpreted subsequently. They include:  Uncertainty about cause. This theme had four sub-themes and they all pointed to the fact that participants did not see homosexuality as having one single etiology. This is most evident in the first sub-theme which dealt with the fact that homosexuality has mixed causes, including biological predisposition, sexual abuse, parents who are too strict and forbid interaction with the opposite sex, high cost involved in getting married, being often in the company of same-sex friends, and wanting to avoid getting pregnant or impregnating someone out of wedlock. The following extracts explain this point further: “I believe God did not create any of us to take interest in the same sex, that one deɛ [that‟s a fact] but there are some people who are born and naturally they just have feelings for the same sex. If it‟s a girl, they rather want to be with a girl than be with a guy. If it‟s a guy, they would rather want to be with a guy than to be with a girl. And then as they grow into society they begin to realize that that thing is not good so you see them hiding…And there are some too who are not born with that interest for the same sex but I find myself in a girl‟s school or yeah, let me use that example. I‟m with girls, it‟s all about girls. When I see a boy I don‟t know how to react. I feel comfortable being with girls so you‟ll find me liking things like that but it doesn‟t happen to everyone. But there are some cases that happen like that.” (Mixed Group 3B) “I think with some people they were introduced to it when they were children, maybe an elderly person, a relative or somebody abused them. So it‟s like they‟ve grown up with that desire.” (Male Group 2B) University of Ghana http://ugspace.ug.edu.gh 152 “Ok I also think that sometimes it‟s due to the parent‟s behavior. Sometimes your parents will be like you shouldn‟t have female friends, only male friends, like studies teacher, just a male everything…all your friends males in order to avoid sex, pregnancy and that stuff.” (Male Group 2A) I think for the ladies most of them are involved to prevent pregnancies, because I know in a particular sports academy most of the ladies are involved in order to prevent pregnancies. So it‟s a way of preventing pregnancy. (Mixed Group 3A) For the men, it is because of high cost of marital stuff, like if you want to get married to someone, the kind of expenses that you will use. So because of that majority of them don‟t want to get married, yeah, and they also think that is the safer side because at the end of the day, the man wouldn‟t get pregnant, and bring about complications and stuff. So all these things, that is why. (Mixed Group 2B) The second and third sub-themes had to do with the fact that homosexuality is caused by the situation where the opposite sex are not available and having had a history of bad heterosexual relationships respectively. As participants put it, “Mostly with the girls it‟s not being homosexuals. They come into the secondary schools and they become bisexual. In school they are with their female partners, outside school they are with their boyfriends. I think the boarding house is the problem. People are limited, they don‟t have their freedom. Just because the boys are not around or maybe they have to go through a lot of stress to get what they want, they just make do with what is available.” (Mixed Group 1B) University of Ghana http://ugspace.ug.edu.gh 153 “And people who have been hurt before or have been constantly rejected. I go into a relationship with a guy, constantly my heart has been broken over and over again. A guy tries to woo women but anytime he tries they reject him but your fellow guy walks to you and the person starts making advances and you feel very loved by the person. Why not? I think sometimes constant rejection by the opposite sex makes people run to the same sex for whatever reason and those who have been very much heartbroken also end up being homosexuals.” (Mixed Group 4B) The last sub-theme on participants‘ thoughts about the etiology of homosexuality had to do with the fact that it is spiritually caused. To some participants, this was the only etiology that made sense because they could not understand why someone should be attracted to and go to the extent of having sexual relations with someone of the same sex as him/herself especially when it would not allow for procreation in the traditional sense. This sentiment was felt strongly in relation to male homosexuals and even for males who have anal sex with females. Apparently, the spiritual attack happens when someone follows through on his curiosity about same-sex sexual relations and when someone is abused sexually (i.e. sodomized). The following extracts depict all these: “It‟s a spiritual thing because I don‟t see why a guy should get attracted to another guy and not a lady because what are you getting attracted to? And in the first place if I get married to a guy, how do I make babies?” (Male Group 4A) “Oh it is more of a spiritual thing. Even if you are a man and you want to have sex through the anus of a woman, it‟s a spiritual something. Why would you do that?” (Male Group 3B) University of Ghana http://ugspace.ug.edu.gh 154 “I think it is a spiritual thing. No one was born that way, it‟s just their mind, the way they think and spiritual things attached. I believe personally that everything starts from the mind. You can sit down one day and say ok, let me go and try and see how it‟s like. Then the spirit will see that oh you also like it so he will also come and visit you and help you to do it.” (Female group 1B) “From what I have really encountered, it is spiritual because there was a young guy who told me that he was in the room and then an elderly person entered the room and forcefully had anal sex with him. And since then, he picked that spirit and also started to sleep with the colleagues and others, and he grew with it. So you realize that there are some of them, it‟s spiritual.” (Mixed Group 2B)  Relating with homosexuals. This second theme had to do with participants‘ interaction with homosexuals, both real and imagined. Only a few participants mentioned that they knew a homosexual person. Most of them rather had people in mind that they suspected to be homosexuals. Nevertheless, they discussed how they would interact with a friend or relative if they should find out that the person is a homosexual. Five sub-themes emerged under this theme, the first being total avoidance of homosexual persons especially because they are likely to be thought of as homosexual themselves by society if seen in the company of a known or suspected homosexual. The following extracts explain this: “I would feel very…what should I even say?…because it is not acceptable so if the people even see me walking with such a person they would also see me as maybe being part of them. I would just shun company with the person or advice the person to stop.” (Female Group 3A) University of Ghana http://ugspace.ug.edu.gh 155 “I will not get closer to you. Let‟s take our community for example and you will get the chance to walk with somebody who is a gay. Let‟s look at what people will say about you. Even your parents will not even allow it, neither your friends. Hey! No one will come to you.” (Male Group 1B) The second sub-theme pointed to the fact that although some participants would not avoid interacting with homosexuals totally, they would be sure to be cautious in their interaction with them. The main reason for such cautious interaction was to avoid being influenced and thereby recruited into practicing homosexuality as is evident from this extract: “I think I can relate with you in a way but I will prevent being with you in secluded areas where you may start showing your interest. I won‟t try to push you away but might try to help in a way. So far as you don‟t show interest in me or bring your things towards me I‟m cool with you as a human being. I wouldn‟t be so close to you to come to your room, eat, hung out and stuff like that.” (Mixed Group 3A) The third sub-theme was social support. Some participants noted that they would provide a homosexual friend or relative whatever the needed support but invariably they alluded to the fact that this support should be aimed at getting the person to stop practicing homosexuality. According to them, “I will try to educate the person to move away from such acts, because if everybody is going to shun the person, then it means the person would continue doing it. So then it means someone must definitely be there to educate the person, try to get the person away from that act.” (Female Group 3A) University of Ghana http://ugspace.ug.edu.gh 156 “Maybe I ought to love the fellow but I have to hate that sin, and I believe that I would let the person know that I really abhor it but I love them. Then I will give him the steps he must take in order to stop, because it‟s not natural. The fellow really needs help…They need help, but in a loving way.” (Mixed Group 2B) Other participants were of the view that the interaction with homosexuals should be in the form of spiritual support since homosexuality is spiritually caused. This was the point of the fourth sub-theme which is depicted in the following extracts: “I would tell the person that it‟s spiritual because it‟s not natural. Even animals don‟t do that. The anus, anus, it‟s unnatural. Animals don‟t do that. So I think this thing, it‟s a spiritual thing, so I think I would have to pray for you just as any other of my friends who is going through something bad, you help the person.” (Male Group 3B) “I would wrap up this child and leave him in a prayer camp; that doesn‟t mean you are rejecting him. The only plausible solution is to wrap him up and place him where I believe…in a church or somewhere…deliverance every day because it‟s spiritual.” (Female Group 2A) The last sub-theme had to do with the fact that relating to homosexuals is dependent on gender, that is, the gender of the homosexual person as well as that of the participant. Male participants generally felt they would be more comfortable interacting with female homosexuals while female participants felt they would be more comfortable relating to male homosexuals. Additionally, some participants felt they would have difficulties relating to a male homosexual because of the feeling of disgust relating to the idea of anal sex. The following extracts make these obvious: University of Ghana http://ugspace.ug.edu.gh 157 “To me if I knew a lady homosexual I could easily relate but for a guy it‟s disgusting. With a guy it‟s disgusting. I don‟t know how I can relate to a male homosexual.” (Mixed group 1A) “I can be close to some extent with a gay but not a lesbian. The reason is that to relate to a lesbian to that extent sometimes you have to be cautious, because before you are aware they are all over you.” (Mixed Group 1B) I‟ll never be friends with a homosexual, a gay, because you cannot know what will happen. I can be cool with them but not intimate. It‟s just like a guy being in the same room with a lady. If you know this guy is a homosexual, don‟t be in the same room with him; anything can happen. (Mixed Group 2B)  Identifying Homosexuals based on appearance. As stated earlier, most participants did not know a homosexual person personally but had suspicions regarding some students on UG campus and acquaintances from their secondary schools or communities. Such suspicions were based mainly on the appearance of the suspected persons in terms of how they dressed, walked, talked and behaved generally which were not deemed appropriate for their gender by societal standards especially with regards to male homosexuals. Apparently the female homosexuals were difficult to identify based solely on appearance. This is what the third theme pointed to as is evident from the following extracts: “Society has expectations as to how a guy should behave and how a lady should behave. If you deviate from it so much then there is a justification for us to classify you as a homosexual. And for me, observing them in [mentions name of hall] from Level 100 up till now, they wear ladies dresses, vests, like it‟s so obvious. For them they have accepted and they know we know. So if you University of Ghana http://ugspace.ug.edu.gh 158 use them as a standard you can measure that guy and say this gentleman is one. They are open, they are friendly, and they…in terms of neatness and stuff, they keep their hair short, nails trimmed. Actually they act like females. The way they walk, the way they dress and stuff. Even the way they talk, their gestures, they want to fit that kind of…yeah” (Mixed Group 3B) “The guys they wear some funny trousers, it‟s tight and they wear fancy colours. They wear pink for you to know them. So those who are also gay when they see one they know. They wear pink, purple and those things and their trousers are very tight. They hang out with girls a lot. For girls they are difficult for you to know, difficult to identify but some of the girls behave like guys. Somebody I know, she behaves like a guy so she is maybe the man.” (Female Group 3A)  Freedom of Choice. This last theme also had no sub-themes. It pointed to the fact that individuals, including homosexuals, should be given the freedom of choice to do whatever they want to do since they are human beings. According to the some participants, allowing female homosexuals the freedom to do what they want might provide a way to bring the female to male ratio in Ghana to 1:1. “They should let them be. It‟s true oo. You see the thing is that right now the women are more than the men. If they accept it and then maybe they allow them to practice lesbianism, let‟s say the population will reduce then it will be one man to one woman, yeah.” (Female Group 4A) “No I don‟t see anything wrong with that because at the end of the day they are still human beings. The fact that let‟s say they have a problem…that‟s if University of Ghana http://ugspace.ug.edu.gh 159 it‟s really a problem, the fact that they do that does not mean they are no longer human beings. At the end of the day they are still humans so whatever we classify humans they should fall under.” (Male Group 4B) Figure 14 summarizes the themes and sub-themes under this topic (see the following page). The topic is shown in oval, the themes are shown in rectangles, and the sub-themes are shown in round edge rectangles. University of Ghana http://ugspace.ug.edu.gh 160 Figure 14. Thematic map showing summary of themes and sub-themes under ‗Homosexuality‘. Homosexuality Uncertainty about cause 1. Total avoidance 2. Cautious interaction 3. Social support 4. Spiritual support 5. Gender dependent interactions Relating with homosexuals Identifying homosexuals based on appearance Freedom of choice 1. Mixed causes 2. Non-availability of opposite sex 3. History of bad heterosexual relationships 4. Spiritual causation University of Ghana http://ugspace.ug.edu.gh 161 Vignettes. Four vignettes were used in the FGDs and each one was discussed in two forms depending on which group of participants were interviewed, firstly in terms of issues relating to the vignette persons and secondly in terms of imagining one is the vignette person. The vignettes are therefore interpreted subsequently according to this format, starting from vignette 1 to vignette 4 with each set of interpretations followed immediately by a thematic map summarizing the themes and sub-themes:  Vignette 1A. Two themes emerged from the analyses of transcripts of participants‘ discussions on Vignette 1A. The first theme pointed to the fact that if homosexuality is due to biological factors as seen in the vignette person‘s situation, then research on treatment should be done while the second pointed to the fact that the vignette person had no justification for being homosexual. With regards to the former, participants thought research can be made into getting cures for homosexuality in the form of medication and hormones as seen from the following extracts: “For me I would say that since researches have been made and this can be proved that those having the sexual genes are…the same research can be made or medicines can be researched into and provided so that those affected can take it and it would control their mental whatever, because if researches have been made, in the same way researches can be made to solve the problem.” (Female Group 1A) “I think I agree with what my brother said because when you look at the scenario you can see that there is a biological basis to his condition. So I think with this one the only thing we can do about it is that the earlier we increase the rate of…because I read androgen here, so it‟s either you increase the University of Ghana http://ugspace.ug.edu.gh 162 androgen content in the person‟s body so that the person can turn into a normal being” (Mixed Group 1A) With regards to the latter, participants thought that the vignette person would have been justified if he/she had at least made efforts to change and chosen to take bribes instead of being homosexual. According to them, “If I were to be her I would have collected the bribe rather than having the same sex partner. We know that both are sins, and we know that there is no big and small sin but so far as society abhors same-sex partners than bribery I think she should at least have collected the bribe. It‟s the lesser evil.” (Mixed Group 1A) “She might have or he or she might have genetic or hormonal whatever which predisposes him or her to be a homosexual. But I think with the nature argument, even if he or she was born homosexual, through reinforcement and nurturing and the little she gains she could have been able to change if she wanted to. Where there is a will there is a way. All persons who can‟t see, they came out as blind, those who are maimed, that is how God created them? No! Someway somehow, during breeding some genes didn‟t form well and this came up. There are twins that came out with their heads together. Do we say that is how God created them? No! People were willing for them to be changed and so they were changed. If this one too was willing to be shaped, I‟m sure he would have been shaped.” (Male Group 1A)  Vignette 1B. Three themes emerged from participants‘ discussion of Vignette 1B. The first theme pointed to the fact that in making moral choices one has to choose the lesser evil as University of Ghana http://ugspace.ug.edu.gh 163 that is preferable, especially in a situation where making a particular choice can be equaled to virtually insulting God. The following extracts depict this: “But for the homosexual thing, the person is a good person but for this one…I have very strong feelings about this one. Somehow I feel if the person was not homosexual and the person accepted bribes, somehow I feel I would be more accepting or accommodative.” (Female Group 1B) “Comparatively, when we compare sins, we have sins that as humans, not God, as humans, the degree that we attach to the way we see the thing…for instance, somebody may steal your phone, but if the person adds killing to it, it becomes weird. Fine, you can take my phone away, but killing me too...fine, you can rape somebody but then after raping and you give the person poison, you see, it‟s outrageous. So I‟m looking at it from the comparative sense, when people compare...this is fornication, fine, God says we shouldn‟t involve in extramarital sex but here is the case where God himself, or maybe a supreme being has created a partner…even though what you are doing is wrong, you go the extra mile, choosing your own sex. I think what you are telling God is that, aside even committing the sin, I‟m even insulting you. You did not create the thing I like. So it is not like there is freewill attached to it, it is weightier than the rest of them.” (Mixed Group 1B) Some participants however felt that if they were the vignette person, they would feel justified to be homosexual because finding out that it is biological from someone in authourity will give them a lightened conscience and a sense of being normal. As one participant put it, University of Ghana http://ugspace.ug.edu.gh 164 “I think I‟ll be happy that my actions and feelings have been justified, ok? Now I have a tangible reason why I am like this. When people ask me I can easily tell them that no, I was born this way. I can feel free, my conscience will be lightened. I can go on and do what I‟m doing. If I am the one, and I have heard that even what I‟m doing it‟s normal, the moment you are doing something and someone justifies it for you that oh what you are doing, people do it, it‟s because of this and that, fine. Then I‟ll do it, because the person is in authority, a doctor, and he‟s saying this, I‟ll do it. I think you‟ll no longer have to worry about feeling like an outcast, and worrying about you being abnormal. Once you‟ve been given a tangible reason why these things happen, it would give you the peace of mind to know that you are one of those special people, and you are really not as abnormal as you thought.” (Mixed Group 1B) The third theme spoke to the fact that the vignette person would likely be vulnerable psychologically given his/her circumstances. Due to societal disapproval of homosexuality, participants felt that the vignette person would face stigmatization because others would not know that his/her sexuality is biologically caused like he/she does. Additionally, he/she would eventually face societal pressure to get married and have children. The following extracts explain further: “But then I will not be walking around with this written on my forehead that this is the reason why I am like this. So I‟m still going to have problems on the outside. I‟m only now comfortable in my home, but I am not going to stay there forever. I‟ll still go outside and I‟ll still get...maybe I‟m now comfortable with myself, the inner conflict with God, but then stigmatization from outside will still be there.” (Male Group 1B) University of Ghana http://ugspace.ug.edu.gh 165 “Life will be good for you at a point in time but then when you grow and the environment you find yourself in...you just imagine, all your friends are getting married and having children and you...other people will start asking why you are not married.” (Female Group 1B) Figure 15 summarizes the themes and sub-themes derived from participants‘ discussions of Vignette 1, both ‗A‘ and ‗B‘ (see the following page). The vignette types are shown in ovals and the themes and sub-themes are shown in rectangles and round edge rectangles respectively. University of Ghana http://ugspace.ug.edu.gh 166 Figure 15. Thematic map showing summary of themes and sub-themes under Vignettes 1A and 1B. Vignette 1A Biological?: Research on treatment needed No justification Moral choices: Lesser evil preferable Vulnerability Vignette 1B Self-justification University of Ghana http://ugspace.ug.edu.gh 167  Vignette 2A. For Vignette 2A, three themes emerged. The first theme spoke to the fact that the vignette person has a strong will that is admirable. The following extract depicts this: “I think the person doesn‟t care no matter what people will say. The person is like this is who I am and this is what I want to do so I don‟t care what you will say. Yeah. The person is having fun. In a way, I share with his will power. I think he‟s doing things according to what his will power tells him to do, not because people will criticize him. He does not care about what people will say about his life. He‟s just being who he thinks he can be.” (Female Group 2A) The second theme had to do with the fact that the vignette person was not justified in being homosexual especially because he obviously had psychological problems that he could have sought help for and also because he was not in a good frame of mind to make any decisions. According to one participant, “Personally I think he‟s got mental issues, like psychological issues. Even right from the start of the whole story, „I‟m so tired of people determining how I should lead my life‟. Clearly you should see he‟s going through a whole lot, and now he‟s kind of fed up. Such a person right from the first statement you can tell he‟s going through some emotional disturbances. So clearly this is someone who is not stable psychologically and so I mean, he wouldn‟t be in the best of positions to make certain decisions. And he‟s trying to justify whatever act he is into with a whole lot of, if I‟m allowed to say, nonsense because they don‟t make sense. So personally he‟s someone who is going through psychological issues and for a person like this…I mean, you can talk about this is what you are into without talking about you being tired, you University of Ghana http://ugspace.ug.edu.gh 168 having so many broken hearts. I mean at the end of the day it‟s a human being that disappoints so if you find yourself in a homosexual act, what is the surety that such a person would not disappoint you? So personally I think he‟s not justifying his actions in any sensible way.” (Male Group 2A) The third theme had to do with the fact that the vignette person felt the urge to justify his actions based on the fact that he is a male homosexual and knows that his sexual preference is not at all acceptable in the Ghanaian society. As the following extract depicts, the vignette person would not have felt the need to complain so much if he were female and therefore a lesbian since lesbianism is rarely mentioned when the issue of homosexuality is discussed in the public domain. Apparently, some men enjoy watching lesbians having sex and therefore would not condemn them as they would gays. “My point is, really, people don‟t normally complain about lesbianism. The truth of the matter is that guys actually love lesbians. They love to see lesbians do their own thing because over here in Ghana you realize the issue that is always on the table of discussion is gay, like men into homosexual act. So the nagging and complaining over here is…that society should allow him to do, should allow whoever to do whatever he wants to do all points to the fact that it‟s probably a man in a homosexual act. Because really women don‟t…people don‟t talk about lesbianism. People don‟t really condemn lesbianism the way they do when it comes to gay, homosexual guys. I know friends who enjoy seeing women make out. It‟s fun to watch. People really don‟t condemn that. My friends will not condemn lesbians, but when you come out to say you are gay they don‟t even want to get close to you.” (Male Group 2A) University of Ghana http://ugspace.ug.edu.gh 169  Vignette 2B. With regards to Vignette 2B, the first theme also pointed to the fact that the vignette person‘s emotional pain was no justification for being homosexual as depicted by these extracts: “I think she is not justified by this broken heart and disappointment. You can‟t use pain to justify your actions. If that were to be the case then if someone slaps me then I‟ll go around and pick a knife, and kill the person. You can‟t justify your action by the pain that you go through.” (Male Group 2B) “Personally, I think there‟s no justification for homosexuality whatever the case may be because you should know what is right and what is wrong. And I don‟t see why you should equate one heart break to a thousand other men, yeah so there‟s no justification for it.” (Mixed Group 2B) The second theme simply had to do with the fact that it is important to conform to societal values for the sake of order and because although individuals own their lives, the fact that they belong to a society means they do not have the freedom to do whatever they like with their lives. The following extracts explain further: “If you are to find yourself in a society, they are not going to set your own rules for yourself. You need to abide by what the society agrees on because if everyone is insisting that I want this, I want that, how would the society be like? If everybody is to set standards for themselves that would make he or she happy, if I say this is what I want to be, it won‟t affect this guy and so that‟s what makes me happy, how would the society be like? There would be confusion.” (Male Group 2B) University of Ghana http://ugspace.ug.edu.gh 170 “Now, when I look at the second paragraph, line 5, “After all, I didn‟t ask to be born, did I?” It looks a little bit contradictory, in the sense that if you say that you are independent, in the first place you never determined, I mean you coming into this world, you never determined it that you should come into this world. Some people determined that you should come into this world, ok, and the fact that, even according to the laws of the state, if you try taking your own life and you don‟t succeed and you are caught, you will be imprisoned. It tells you that you never own yourself. Right from infancy, at the hospital and all that, the government kind of subsidizes a lot of drugs and then all that, to take care of you because it‟s like you‟re an asset to the nation, so you don‟t decide to live anyhow and claim that is your life. As a matter of fact, a man is not an island.” (Mixed Group 2B) Figure 16 summarizes the themes and sub-themes derived from participants‘ discussions of Vignette 2, both ‗A‘ and ‗B‘ (see the following page). The vignette types are shown in ovals and the themes and sub-themes are shown in rectangles and round edge rectangles respectively. University of Ghana http://ugspace.ug.edu.gh 171 Figure 16. Thematic map showing summary of themes and sub-themes under Vignettes 2A and 2B. Vignette 2A Strong will No justification Lesbianism more tolerable Conforming to societal values is important Vignette 2B No justification University of Ghana http://ugspace.ug.edu.gh 172  Vignette 3A. Vignette 3A had two themes emerging from the analyses of participants‘ transcripts with each of the themes having two sub-themes also. The first theme showed participants‘ feelings that parents of the vignette person had no choice but to make her undergo surgery. One of the sub-themes spoke to the fact that the vignette person‘s genetic make-up necessitated having surgery while the other spoke to the fact that the surgery was necessary to avoid societal disapproval. The following extracts explain further: “Let‟s say you have 3 arms and you want to take one off. That one nobody is going to say anything about it because it doesn‟t change…like it‟s changing you for the better so for this one you find out that the female side was more dominant than the male side so it was right to do it.” (Mixed Group 3A) “I really think the parents did the right thing because considering our society you can‟t come back and say „sesei deɛ Kojo yɛ Adjoa oo‟ [Right now Kojo is rather Adjoa]. Even if you are to explain to them the medical condition they still will look at you in a negative way though we are changing. Ghana is in a transitional turn but like we still hold on to our cultural and traditional values. A lot of things haven‟t changed.” (Mixed Group 3A) The second theme pointed to the vignette person‘s vulnerability in terms of having to adjust psychologically to being a female and having trust issues. With the psychological adjustment, participants thought that she would wonder and worry about whether she can be a woman having grown up as male and whether she would be able to have children. As depicted University of Ghana http://ugspace.ug.edu.gh 173 by the following extracts, she would also have the added challenge of forgetting her old identity as male when with family members and friends who knew her before her surgery. “Because she has to battle with the fact that she‟s once been a man, and there are so many question marks about her life. I‟m I going to give birth? Am I going to enjoy being a woman? How is being a woman like? There are so many questions in her life and I think it would even affect her relation with people. And even her family members too. They all think she was adopted so that one too…If they mention her former boy‟s name, if she doesn‟t take care she would say yes. Well, it‟s very sad because she is going to fight it forever; it‟s going to be there forever.” (Female Group 3A) With the trust issues, participants thought that the vignette person should tell her partner about her past if she feels he would not betray her trust and also because her partner would feel betrayed if something should happen later to bring out the truth about her past. As one participant put it, “I think she should tell him because if she keeps it to herself it might pop up one day and she might end up losing all that she has gone through. I think she should tell him and if he can live with that…it‟s been corrected. If you trust the guy you should tell him. If he can live with it he would be cool with it. That doesn‟t mean you should go and tell the family members that my wife was once a man.” (Mixed Group 3A)  Vignette 3B. Vignette 3B also had two themes emerging from the analyses of participants‘ transcripts. The first theme had no sub-theme and pointed to the fact that the vignette person‘s decision to University of Ghana http://ugspace.ug.edu.gh 174 undergo sex-change surgery was right. Avoiding the surgery would have meant depression from having an unidentified gender and not having the new life that surgery had given her as is obvious from these extracts: “Now I‟m a female, and it‟s not like I was completely male either. I had female tissues and then male tissues. And it looked as if the female ones were dominant than the male so I had to be female. I don‟t think it‟s a big deal. It‟s better than being….if I maybe I can‟t do the surgery, I would just be there. I‟m not a boy and I‟m not a girl. The depression alone can kill you.” (Female Group 3B) “I mean she should be happy. That‟s a new life you‟ve been born into. The doctor says you are perfect.” (Mixed Group 3B) The second theme, which had to do with the vignette person‘s vulnerability however had two sub-themes that also had to do with trust issues and psychological adjustment. With the trust issues, participants thought that the vignette person should not tell her partner since it might result in a break-up and subsequently to others finding out about her past because her partner would have to explain why they broke up. According to one participant, “From what I read it‟s only between herself, her parents and maybe the doctors. No one knows. I won‟t tell him. If you tell the person the person might just freak out. If I‟m so sure there wouldn‟t be any after complications or something. In addition maybe everyone knows you are going out with Nathaniel and suddenly you break up, and then Mr. A meets him. „Oh how is your fiancé doing?‟ „Oh we broke up.‟ „What‟s the cause?‟ You‟re pushed to say what happened and now there is a second party and third party coming in and it keeps going so eventually it‟s going to spread University of Ghana http://ugspace.ug.edu.gh 175 everywhere. The moment it gets out to one person, it would spread like wild fire.” (Female Group 3B) With regards to the psychological adjustment, participants felt that it would be more challenging to live in Ghana after the sex-change because one would have to interact with family and friends like a different person while knowing one is the same person they knew. This extract depicts their sentiments: “For me I would never ever come back to Ghana, never. I would just stay in the states. No matter what, if I should come down my guilty conscience is going to have some kind of pressure on me. Even though they have lied that I‟m dead I would still have…there is something you can never take away from me, you can never lie about. If they mention your name you would turn. I can‟t come back to see the same old people and they thinking I‟m a different person whiles I know myself that I‟m the same person. No, I can‟t take it.” (Male Group 3B) Figure 17 summarizes the themes and sub-themes derived from participants‘ discussions of Vignette 3, both ‗A‘ and ‗B‘ (see the following page). The vignette types are shown in ovals and the themes and sub-themes are shown in rectangles and round edge rectangles respectively. University of Ghana http://ugspace.ug.edu.gh 176 Figure 17. Thematic map showing summary of themes and sub-themes under Vignettes 3A and 3B. Vignette 3A Parents had no choice except surgery Vulnerability Vulnerability Vignette 3B Sex-change is right decision 1. Trust issues 2. Psychological adjustment 1. Psychological adjustment 2. Trust issues 1. Genetic make-up of child 2. Avoiding societal disapproval University of Ghana http://ugspace.ug.edu.gh 177  Vignette 4A. Four themes emerged with regards to Vignette 4A. The first two themes had no sub-themes and while one dealt with the fact that the vignette person was not justified in being homosexual, the other one dealt with the fact that the person was vulnerable. With regards to the former, participants‘ sentiments were that the vignette person was just plain crazy to have undergone sex-change since it seemingly made her lesbian by choice. As one participant put it, “I think he is crazy. Because how can you change what you are? No matter how much change you do to the body you are still the person you are. This is madness. I don‟t see why he has to do that. There is no reason in doing it. That philosophy is very stupid because it now seems like she is a lesbian by choice.” (Mixed Group 4A) With the latter, participants‘ sentiments were that it would have been in the vignette person‘s interest to get treatment for his psychological problem of not feeling comfortable with his gender than going for a sex-change that would bring the added challenge of adjusting to life as a different gender. “I think it is a psychological disorder and if the person had actually voiced it out the problem could have been solved.” (Mixed Group 4A) “She can‟t have that mentality that guys have, that thing that boys have, that boys will be boys. She is not going to be able to produce that mindset. So it is only her physical this thing that is changed but her mentality is still going to be the feminine stuff.” (Female Group 4A) University of Ghana http://ugspace.ug.edu.gh 178 The third and fourth themes had to do with participants‘ thoughts on the parents‘ decision and the vignette person‘s decision respectively. They each had two sub-themes which in turn had to do with whether the decision made by the parents and the decision made by the vignette person was good or bad. With regards to the parents‘ decision, some participants thought it was a good decision to forbid the vignette person‘s siblings from interacting with him because he could influence them but others thought it was a bad decision because personally, they would not be able to abandon a sibling whatever the circumstances may be. The following extracts depict these: “I would support the parents because who knows he or she would come around to influence the others for them to do the same thing. So it was a good decision to stop the siblings from seeing him.” (Male Group 4A) “For me personally, I wouldn‟t obey that. No matter what the person has done, the person is still your sibling. I think if you‟ll not need that person as a sibling, you‟ll not be given birth to by the same parents. So it‟s for a reason that that person becomes your sibling. In that difficult time when everybody abandons that person, you are the person who can give that person the consolation, you can give that person the advice, you know, share something with him or her. So in one way or the other I think it‟s a very wrong decision the parents took from my personal opinion. (Mixed Group 4A) With regards to the vignette person‘s decision also, some participants thought it was a good decision because the sex-change meant she was avoiding being a lesbian and others thought it was a bad decision mainly because it was a decision made in haste at a time when she probably was depressed. The following extracts point to these: University of Ghana http://ugspace.ug.edu.gh 179 “In a way, his choice, to me I see it as okay because she did it so that in a way she could avoid certain things. Being attracted to females means that she would have been a lesbian. So in that sense it is okay.” (Mixed Group 4A) “Sometimes in our lives things come up. You can have your downs and your ups so he feeling that way could be one of his down times and you know sometimes when you‟re down, you have so many thoughts coming into your mind like, most of them they end up like death but if you take your time and look through things, you take advice from others, someway somehow, you‟ll feel much better than doing this.” (Female Group 4A)  Vignette 4B. Only two themes emerged here with no sub-themes. The first theme dealt with the fact that the vignette person had no justification for undergoing sex-change and should have been more appreciative to God by accepting who she is. The second theme dealt with the fact that the vignette person‘s sex-change was too drastic because she could have chosen to just wear male clothing or become a lesbian, either of which could have afforded her the opportunity to change later. These extracts explain further: “This person has made God look like he is not God, like He didn‟t know what he was doing from the beginning. These are the kind of people who don‟t appreciate what they have just because they think they are not comfortable. Life is how you make it comfortable. If she had accepted the fact that this is the person I am and I‟ll remain like that and make it suitable for me, she could have made it suitable without having to undergo certain things.” (Mixed Group 4B) University of Ghana http://ugspace.ug.edu.gh 180 “Well, me if I have money for such change I will use it for something more comfortable. I wouldn‟t, she wanted to become a boy, yea, I will wear boys‟ clothes but I don‟t think to the extent of changing my sex. I will like to think of the aftermath of changing my sex. So I don‟t think I will do that. You could wear boy‟s clothes; you could act like a boy. I don‟t think I will have surgery.” (Female Group 4B) “Like I will be a lesbian. I will be a lesbian. Yeah, it‟s better than changing your sex. At least maybe some time to come I will come back to my senses and because that one there is hope. You can go back to your old state but this one unless you reverse the thing, make it a new thing.” (Male Group 4B) Figure 18 summarizes the themes and sub-themes derived from participants‘ discussions of Vignette 4, both ‗A‘ and ‗B‘ (see the following page). The vignette types are shown in ovals, the themes are shown in rectangles, and the sub-themes are shown in round edge rectangles. University of Ghana http://ugspace.ug.edu.gh 181 Figure 18. Thematic map showing summary of themes and sub-themes under Vignettes 4A and 4B. Vignette 4A No justification Thoughts on parents‘ decision Thoughts on person‘s decision Sex-change is drastic Vignette 4B No justification Vulnerability 1. Good decision: Avoiding lesbianism 2. Bad decision: Made in haste 1. Good decision: Siblings would be corrupted 2. Bad decision: If a sibling, would disobey University of Ghana http://ugspace.ug.edu.gh 182 In-depth Interviews. Transcripts of the IDIs were analysed in the same manner as was done for the FGDs with coding done around the topics that were discussed with the interviewees. Although the topics discussed with the male homosexuals and the female homosexuals were the same, the themes and sub-themes derived from analyses of their transcripts were not the same. Each group of themes and sub-themes were however represented in each of their corresponding transcripts and these are interpreted subsequently. The interpretations under each topic are immediately followed by a thematic map that summarizes the themes and sub-themes for the topic. IDIs with male homosexuals. Four themes were derived from analyses of the transcripts of the male homosexual participants‘ interviews. They include:  Cause of homosexuality. This first theme was in relation to participants‘ thoughts about the etiology of their sexual orientation and two sub-themes emerged under this. While the first sub-theme pointed to the fact that some participants thought their homosexuality was biological, the second pointed to the fact that some participants thought it was by choice. For those who thought their homosexuality was biological, they reported that they had had same-sex attraction since childhood as is evident from the following extracts: “Growing up as a child I felt warm towards a boy in my class. He was older than me and he would save me from all kinds of situations like if someone wanted to beat me or…from class four coming to JSS and so I had, instead of dreaming of a girl or getting married to, he was the one in my dreams so I kind of felt like you know…I thought it was, I was the strangest person on earth, I thought I was the only person and so, until my juniors in SSS [Senior Secondary School] opened my eyes and then I knew that there were other University of Ghana http://ugspace.ug.edu.gh 183 people like me out there and so I wasn‟t the only odd one out so that is how and it‟s been like that.” (M9) “Well under normal circumstance when you are born, and you are known as straight but whilst I was growing up I realized that it wasn‟t so with me. If I listen to what people, friends say about their sexual feelings and I realized that I wasn‟t feeling that way, yeah, so I realized that well, I am different or I have a different sexual orientation. So it was through what I heard, what I read myself and what I feel about the opposite sex and the same sex. I will say it is 100% biological.” (M7) For those who thought their homosexuality was by choice, they noted that it was due to curiosity and being in an enabling environment such as an all-boys school and having friends who practiced homosexuality. As the participants put it, “I think, hmm, I don‟t know o, along the way something might have happened I don‟t…I don‟t know, I just went to secondary school and it, you know there is this thing at a particular stage in your life as a guy growing up, you fantasize about girls and all that then it altered along the way or you followed the path depending on the environment you are in. I went to a boys‟ school so I had something to do with someone and that is it, I think about the girls but…there was this girl...I had a relation with her but not sexually. We had like, in junior high, boyfriend–girlfriend yes, but not sexually. That [referring to the relationship with a boy] was sexually, so it changed everything. I think it‟s by choice. I won‟t say…I can‟t say I was born with it. Probably it‟s the environment and what you see others doing. You are in the school and there University of Ghana http://ugspace.ug.edu.gh 184 are other people doing the same thing and in a way secretly you might be looking up to those people and you accept it.” (M8) “Eii! I wouldn‟t say that I was forced into MSM but rather I got some few friends who were like that and before I realized I was also engaged in this MSM so I will say by choice.” (M6) “Emm just curiosity. I was curious to find out what it is about. Why they do what they do and what they do. It‟s by choice.” (M4)  Reactions by society and fellow students. This second theme had to do with reactions by society and their fellow students towards participants because of their homosexuality. This theme had three sub-themes which pointed to the fact that members of the society as well as fellow students had negative reactions to homosexuals and wrong perceptions about homosexuality but those who were close friends were ‗cool‘ with it. With regards to the negative reactions, participants said that those who are effeminate and can be judged by their appearance as being homosexual and those who are open about their homosexuality mostly get hooted at and called derogatory names on campus. The following extracts explain further: “People are not saying it out and loud but the general discussion of it amongst people on campus is so negative. If you come out point blank that this is what you do you don‟t expect a good welcome, you don‟t expect a positive welcome. There has been a few incidences whereby some guys who are being judged by their appearances, how they look, how feminine they look and all that and it is obvious and students like that have come out themselves to say that yes, they are really doing it and the remarks are not so good and how they are treated is not so good. They are being hooted at any time they pass by and even if you University of Ghana http://ugspace.ug.edu.gh 185 are not and you are found in their company, there and then you are being judged as one. It‟s very nasty. People who belong to the other side of this orientation are, are, are so mean. They are mean in terms of their judgment. You should listen to some of the comments they pass. Some people are like it‟s a demonic attack and it‟s a demonic spirit which is easily transferable from one soul to the other and if you should get close to them or exchange pleasantries with them you could get caught in it” (M3) “Well, I haven‟t seen anyone being physically…well, I have heard of a case, just one, one guy I know on campus who was abused by some guys just for being gay. I don‟t know if they themselves are gays or, he said they are not but they did it, that to him. About four guys” (M7). “They kind of, they act like homosexuals are demons, trust me they act like…they react not violently but negatively because when it happens that way, they see you and like you turn into a laughingstock. They don‟t want to walk with you and all that. Because they believe they will also be tagged as homosexual right. Those who make it obvious that they are, are being treated bad. Like people laugh at them, people actually laugh at them in the sense that they make you aware they are laughing at you, people call them funny names as in like „trumudifↄo‟ [one who has sex through the anus]. I have had several, not me per se, but I am with my friends and somebody is passing, like you hear people laughing and hitting drums, emm utensils sorry…those who have made it obvious are being treated badly but those who are living under cover life ok, are kind of accepted, someone like me.” (M5). University of Ghana http://ugspace.ug.edu.gh 186 With regards to the wrong perceptions about homosexuality, participants noted that fellow students thought male homosexuals had the tendency to abuse other males sexually and were the cause of HIV/AIDS. As one participant put it, “Wrong perceptions. I spoke to one guy, he goes like he thinks if he is friends with a homosexual, the homosexual will rape him. Another one tells me men who have sex with men were the cause, who through some god punishing them, they were the cause of HIV and some of them were also having sex with women so they brought HIV and they caused the death of their relatives and everything.” (M1) For the third sub-theme, some participants stated that in cases where they had been open about their sexual orientation to fellow students who were close friends, these friends had been cool with it and continued to have good relationships with them so long as it was not obvious to other students that they are homosexuals. Participants noted also that despite this cool treatment, they personally still feel uncomfortable around friends who are aware of their sexual orientation as some of these friends would still tease other students who are easily identified as homosexuals based on appearance. The following extracts explain further: “When they first found out I was even surprised myself, because they didn‟t show any signs of hatred or anything like that. They were cool with me and even up till now they are still close and when we meet. It‟s like they don‟t know anything secret about me, like, their reaction towards me is normal just as it was but the only thing is that me myself I am not able to feel comfortable around them.” (M7) University of Ghana http://ugspace.ug.edu.gh 187 “Emm, I am trying to give you an example because I have a best pal, ok, who is on campus with me, I told him about my sexual preference, I told him I am and when I am walking with him and other friends and we happen to see someone who is, who has, of course is in his full you know, walking like swaying and all that ok, even my best friend also laughs or makes fun of that person so in that case I am being tempted to believe…as in just because I am not as crazy or I haven‟t really had myself like as feminine as that person so I believe they don‟t laugh at me but he knows but he laughs at someone who is so obvious.” (M5)  Vulnerability. This theme had three sub-themes which pointed to participants‘ vulnerability due mainly to their awareness of the negative attitude of society towards homosexuality. The first sub-theme had to do with their emotional pain when fellow students react negatively towards them as depicted in the following extracts: “I feel so bad…sometimes even eating becomes very difficult when people react that way.” (M6) “I remember this instance which happened this week, I just entered into a room to ask for something…apparently one guy was in my class and something happened in my class which made me not want to go to the class any more you know, [interviewer interjects asking what happened] someone sat by me and looked at me, my face, looked at me up and down and just stood up and tried to make everyone behind me realize that like he is getting up away from me and people were giggling and it caused me mental stress and after class ended, I couldn‟t get up because I was scared, you know, so I University of Ghana http://ugspace.ug.edu.gh 188 entered the room, he saw me and then they gave me stares as if like, I don‟t have the right to step here like, if they had the right they would have murdered me or something…I feel like „this is not fair‟, it makes me so angry, so bloody angry, excuse me, but …” (M1) The second sub-theme spoke to participants‘ discomfort with their orientation due to negative perceptions about them and their own intrapersonal conflict as a result of their religious background. According to the participants, “I will say I am not. Am not comfortable because you yourself will bear witness with me that, emm, we live in a society where being a homosexual is considered as bad or immoral, illegal and so if you are living in such a society and this is your sexual orientation, personally, the reason why I don‟t feel comfortable or I don‟t want to accept my sexual orientation is because of the, because of the way people perceive it to be. Also because of my religious background.” (M7) “Am I comfortable with it? I think life would have been a little bit more easier if I was not homosexual. I think it‟s not my choice so I just have to live my life so I am living it.” (M2) With regards to the final sub-theme, participants pointed to the fact that some of them are forced to get into relationships with the opposite sex, not because they are bisexual but because they need to use heterosexuality as a cover-up for their homosexual activities so as to feel safe and accepted by society. The following extracts explain further: “Emm…on this issue, well, I am a homosexual ok, and I am living a life…I am living a strictly undercover life, because I have a girlfriend and I have a University of Ghana http://ugspace.ug.edu.gh 189 boyfriend and I am hiding things and I am just not comfortable because it is just not allowed because I believe if I should just tell my parents I am, I am gay, they could take measures that wouldn‟t favor me, you know. I am scared, seriously…I am and I am proud but not safe.” (M5) “Because apparently they are scared, some of them have girlfriends. My best buddy now is so heterosexual, not because of sexual attraction like, „I like girls, I like girls‟, you know, but it‟s…he‟s afraid of what his friends will think so because of the way I am, at certain times he is scared that I come to his room because one of his roommates knew me from Level 100 and this roommate told him that he knew me and he said, he said it in a certain way so he was worried.” (M1)  Openness about sexual orientation. This theme dealt with participants‘ openness about their sexual orientation to friends and family members. With the former, participants‘ responses pointed to the fact that they are mostly open only to friends who share their sexual orientation but not to those who do not share their sexual orientation due to wanting to avoid discrimination. The following extracts depict this: “I wouldn‟t say I am open. As at now only one person…not one person, I told one person myself but the others who know are people…they got to know themselves, I don‟t know how but I am not open, I wouldn‟t want people to know. As I said earlier because of the way people might see me. They might even isolate me like withdraw from me and it is possible that some people will even cause physical harm to me if they get to know that this is who you are.” (M7) University of Ghana http://ugspace.ug.edu.gh 190 “I am not…it depends on how you put it. I am open to those who are. I am not open. Why am I not open? Because of what the society, kind of, the norms in the society, are you getting me? I am not living in a society who accept this so I have to, you know…I think the greatest thing is the avoidance, are you getting me? You cannot just be living in a society and everybody is avoiding you because you are different. That is one thing that personally I am scared of…so why should I? I live a normal life and let‟s have few friends who are and know about it.” (M2) With regards to openness about their sexual orientation to family members, participants generally noted that it was not an option because their parents especially would not accept it and would be disappointed. Apparently, family members would rather not ask for a confirmation if they have any suspicions and would expect participants to lie when asked directly by themselves or other people within the community. According to participants, “No, I won‟t open…family members are always suspicious to begin with so they get suspicious and somebody whispered something to a family relative and the person went round telling my uncles and all that but my family particularly, it is the „don‟t ask, don‟t tell‟ thing. My sisters, if they know, they won‟t even ask me about it, yes, I think it is quite shameful to them to even ask „You, are you gay? Do you do boys?‟ No, they wouldn‟t, they don‟t want to accept it…Do you know what I think? I think when you are, they will pray you don‟t come out with the affirmative, you should deny it or say it‟s no because of…even when they know yes that is it, for my family it is your business but you yourself you should know it‟s wrong especially when there are others University of Ghana http://ugspace.ug.edu.gh 191 there and you tell, you openly come out „this is who I am‟, they will like you to rather keep it to yourself.” (M8) “My father won‟t have it, my mother will suffer, my siblings they can‟t face me…that is what I think. My father, no, no way, but my mother will suffer. She is very, very attached to me in a way and she might be very disappointed even though I think she‟s seen all that there is to see about somebody like…I think she will not really accept it, that‟s what I think. She is Christian, a very good Christian.” (M9) “I can‟t. Because I feel I will disappoint my family…I am scared of the look on their face, the disappointed look, like „what?‟ No, I just can‟t.” (M5) Figure 19 summarizes the themes and sub-themes derived from the IDIs with the male homosexual participants (see the following page). The oval shows both the group of participants interviewed and the topic discussed with each of them while the rectangles and round edge rectangles show the themes and sub-themes respectively. University of Ghana http://ugspace.ug.edu.gh 192 Figure 19. Thematic map showing summary of themes and sub-themes derived from interviews with male homosexuals. Male homosexuals: Sexual orientation and related experiences Cause of homosexuality Vulnerability Openness about sexual orientation Reactions by society and fellow students 1. Emotional pain 2. Personal discomfort with orientation 3. Heterosexuality as cover-up 1. Negative reactions 2. Wrong perceptions 3. They are cool. 1. Biological 2. Choice 1. To friends: Dependent on sexual orientation 2. To family: Not an option University of Ghana http://ugspace.ug.edu.gh 193 IDIs with female homosexuals. Analyses of transcripts of the female homosexual participants‘ interviews also had four themes emerging from them. They include:  Cause of homosexuality. This theme dealt with what participants thought were the causes of their homosexuality and it had three sub-themes. The first sub-theme pointed to the fact that some participants thought their homosexuality was as a result of circumstances including being in all-girls schools, being in the boarding house, and having friends who were practicing homosexuality. According to participants, within these circumstances, their same-sex sexual experiences just happened and this is evident from the following extracts: “I never had attraction for same sex, never. It just happened, I don‟t know. It just happened. Sometimes she comes to my room…she was very close to me so she usually comes to my room and I don‟t know, she started the initiation and before I realized it was something else.” (F3) “Me like this since infancy I‟ve been in the boarding house, girl‟s school, from class one to secondary school, girl‟s school. I went to [mentions name of school] JSS and SSS then I came here. So I started those things when I was in JSS 3. What happened was…since infancy I‟ve been dressing like this because I was playing football but I wasn‟t in a team…when I was in JSS 1, they were forcing me to be into that because I was having these two friends [mentions names of friends] who were already into that thing. They would write letters and I was like „no, I don‟t want to do this‟ and I was SU [Scripture Union] executive so I used to preach, I used to tell my friends „stop, it‟s not good, it won‟t help you‟. So I don‟t know how…we acted a play and I was the guy in the play so after the play, they started chasing me…the secondary University of Ghana http://ugspace.ug.edu.gh 194 people, training college…because I really acted that day so after that they were chasing me. So what happened was that there was this girl, [mentions name of girl], she told me she‟s sick so I should come and sit by her. So when I went there, there was a tent. Have you seen that thing? When you are in school they would put a tent around the bed. I went in and we were talking and she said I should kiss her. Then it started from there. She wasn‟t into that but she felt something for me and it started from that. She became my girlfriend. She was my first girlfriend.” (F5) The second sub-theme also pointed to the fact that participants became homosexual in order to avoid pregnancy especially because they had had prior heterosexual relationships where their male partners had maltreated them because of a real or faked pregnancy. According to one participant, “With the few relationships I have been in with guys, it has been tough, it hasn‟t been easy because the way you feel about the person, the person is just opposite minded. So if someone is to treat you that way, how would you be happy and be in that relationship? No, no, no. You just have to stay away or forget for now. You see, the reason why other people indulge in girls I think…like my situation like this, many guys have disappointed me. And ok, I‟ve been with this guy, I‟ve been with maybe Kwaku, I loved him so much but when I got pregnant Kwaku maybe he should have said „ok, give birth, then maybe we get married‟. Kwaku wouldn‟t say that. He would be like „no, abort it‟. Abortion too can affect you, it could even kill you. Kwaku didn‟t think of that, you see. So I‟ve aborted the baby, and later on Kwaku is cheating on me, forgetting all that we‟ve done. I‟ll get heartbroken. I‟ll think all University of Ghana http://ugspace.ug.edu.gh 195 guys are the same even if someone is approaching me and is bringing something new.” (F3) With the third sub-theme, participants‘ sentiments about what caused their homosexuality was that female partners are more caring than the males and more selfless during sex hence their decision to become homosexual. The following extracts explain further: “Girls are more caring. The girl always calls me. When I‟m in school and I‟m hungry, I‟ll call her and she‟ll bring me food. But that guy he wouldn‟t do it. Always giving excuses…That wasn‟t about love. The girl wasn‟t doing those things. And she makes me come [reach orgasm],the man, no, but she makes sure I come before she gets tired.” (F1) “Well, some of my friends, they really get orgasm when they sleep with girls than guys. They really get orgasm when they sleep with girls and guys too usually they think of themselves before you the lady when they are sleeping with you and ladies too they make sure as she is benefiting and enjoying you are also enjoying.” (F6)  Society disapproving. This theme had to do with the fact that society is disapproving of homosexuality as depicted by the following extracts: “You know Ghanaian society, they don‟t approve of those things. And even if they see you as a lesbian the way they would react towards you, as if you are not a human being.” (F1) University of Ghana http://ugspace.ug.edu.gh 196 “Yeah, they frown on it, unless maybe those people are into that. They, when they find out, they are cool with it but people who are not into that, eiii, for heaven‟s sake!” (F5)  Living up to societal expectations. This third had ‗marriage and childbirth‘ and ‗heterosexuality as cover-up‘ being its sub- themes. With regards to marriage and childbirth, participants stated that society expects that women would marry and have children, making it inevitable for female homosexuals to do so. According to them, “When you are ready to give birth, you can marry. Yeah, that is what society expects from me, especially northerners. Recently my father‟s brother came and was like „Rita no ↄte fie ha ↄreyɛ dɛn? W‟akyɛ. Ɔnnware na ↄnnfiri fie ha?‟ [What is Rita still doing in this house? Won‟t she marry and leave this house?]. They know that all the girls will get married and go so the house is for the males so you have to get married and move out of the house. They are expecting you to do that.” (F1) “Abi [well] it would come to a time they would tell you that „marry, marry, marry‟ so that is when I would have to see to myself, to the cultural context of where I find myself. So I have to find somebody to entertain them, then I do my own stuffs to entertain myself…it‟s not my wish but since the society is trying to make me, I have to sometimes confront them and live my own life as well. I‟ll marry. I‟ll have my children because my mom needs children. That is the Ghana culture.” (F4) Heterosexuality as cover-up is related to the sub-theme of marriage and childbirth because as participants noted, female homosexuals also have boyfriends and even go to the University of Ghana http://ugspace.ug.edu.gh 197 extent of getting married to satisfy their families and avoid societal disapproval. Apparently, some female homosexuals plan to get married to male homosexuals so they can satisfy society while having the mutual understanding to satisfy their same-sex desires outside the marriage. The following extracts explain further: “I know some girl who is married yet she is still a lesbian. She told me she got married because she didn‟t want people to know she was a lesbian. This is what the society wants, so that‟s why she got married.” (F1) “My boyfriend comes home so my family…[Interviewer interjects asking how long she‟s been with the boyfriend] 2 years, just like my girl. That is why my girl also told me that „why don‟t you allow me to take a guy so that in case my family…‟ I said, „ok, fine‟. A friend also said that she wants to get married to a homosexual so that she wouldn‟t get any pressure from anywhere like, she is a lesbian and would get married to a gay…they would have children so she wouldn‟t get pressure.” (F4) “That‟s cover-up. They are using the guys as cover-ups. I have many friends who use the guys as cover-ups.” (F6)  Openness about orientation. This theme had to do with participants‘ openness about their sexual orientation to others and to their families. To others, participants noted that they can be frankly open depending on the circumstances as is evident from the following extract: “When I was on campus I used to dress like…I was a serious tomboy and I was a sports girl too so people got to know. So like my course mates, half of them, they University of Ghana http://ugspace.ug.edu.gh 198 knew that I was a strict lesbian, not bisexual, because the guys would come propose to me and I would be like „me, I fuck pussy‟. I would just tell you straightforward, „I fuck pussy, I like pussy so much‟. I don‟t care telling you.” (F5) To family, on the other hand, participants stated that they would not openly tell but that family members, particularly parents, suspected due to a lack of male friends. According to one participant, her mother got suspicious because a jilted girlfriend came home to confront her. Another participant noted that her mother sent her to a prayer camp for spiritual treatment based on suspicion that she is homosexual. This is what they said: “…recently my girl came to stay in our house. My mother was suspecting, saying yes what they said is true; the way I do my things…the way no guy comes around to look for me…so my mother even took me to some church. I went to sleep there for about a week just for the pastor to deliver me.” (F1) “My parents, no. There was this girl, I broke up with her. One time she came to my house making noise and all that. So my mum got to know and seriously up till now she keeps on talking, she keeps on praying for me but for my dad…I‟ve lied to my dad and mum but my mum believes. Even though I‟ve not told her the truth she believes it‟s true. My dad doesn‟t believe it. He, he doesn‟t believe it.” (F5) Figure 20 summarizes the themes and sub-themes derived from the IDIs with the female homosexual participants (see the following page). The oval shows both the group of participants interviewed and the topic discussed with each of them while the rectangles and round edge rectangles show the themes and sub-themes respectively. University of Ghana http://ugspace.ug.edu.gh 199 Figure 20. Thematic map showing summary of themes and sub-themes derived from interviews with female homosexuals. Female homosexuals: Sexual orientation and related experiences Cause of homosexuality Living up to societal expectations Openness about orientation Society disapproving 1. To others: Frankly open depending on circumstances 2. To family: Suspicion based 1. Result of circumstances 2. Fear of pregnancy 3. Female partners more caring 1. Marriage and childbirth 2. Heterosexuality as cover-up University of Ghana http://ugspace.ug.edu.gh 200 Discussion As noted previously, this current study was conducted in two parts, including FGDs with presumed heterosexual University of Ghana (UG) students and IDIs with known homosexual UG students. The findings made from analyses of the FGDs participants‘ transcripts and interpreted earlier are subsequently discussed topic by topic for the sake of clarity. With regard to the IDIs, since they were basically conducted to complement the FGDs, its findings are mainly discussed in comparison to findings from the latter. Religiousity and religious commitment. For the FGDs, in addition to discussing this topic in relation to Ghanaians in general and UG students in particular, participants were also asked about their personal levels of religiousity and religious commitment on a scale of 1 to 100%. Some of them declined to respond to this question stating that it was too personal. Of the participants who stated their level of religiousity, the lowest level given was 20% and the highest level given was 98% with most of them stating levels that ranged between 50% and 70%. With regards to their level of religious commitment, the lowest level given was 2% and the highest level given was 90% with most them stating levels that ranged between 30% and 50%. Considering these figures in addition to the themes that emerged under this topic, it is obvious that they confirm the 2012 Win-Gallup International‘s Global Index of Religiousity and Atheism which placed Ghana as the first among the top 10 religious populations of the world with 96% of Ghanaians stating that they are religious. Also, participants‘ thoughts about how religion permeates all aspects of the lives of Ghanaians confirm Atiemo‘s (2013) assertion that religion in Ghana is not only important because of the number of citizens that claim religious adherence but also because of its intense manifestation in the daily lives of the people. As noted previously, Atiemo stated University of Ghana http://ugspace.ug.edu.gh 201 that religious beliefs in Ghana are always part of the issues ( e . g . politics, the economy, health, education) that are debated on in public and that even at workplaces and government organized programmes, prayers are said before the start of work or the start of the programme. Additionally, as noted earlier, Gyekye (2003) stated that belonging to a particular religion requires a certain level of commitment in order to establish constant ties with whoever is the center of the particular religion that one ascribes to. Religious commitment has also been defined as the degree to which a person adheres to his or her religious values, beliefs, and practices and uses them in daily living (Worthington, 1988). Considering these, it can be concluded that the participants were quite accurate in their assertion that religious commitment means much more than just attending the religious gatherings of the religion that one is affiliated to. Obviously, it is based on this understanding on the part of participants that they felt that religious commitment is also linked to the particular religion that one is affiliated to. As the extracts stated earlier showed, they considered practitioners of the Traditional Religions and Muslims to be more committed to their religion than Christians are because the former are seen to adhere much more strictly to their religious values and beliefs by practicing and using them in their daily living. It must be noted however that the low level of religious commitment noted by participants does not really matter considering that some studies (e.g. Mohr & Sedlacek, 2000; Negy & Eiseman, 2005; Schulte & Battle, 2004) have shown religious attendance alone to be enough cause for negative attitudes toward homosexuals. Morality. Which behaviours are moral and which behaviours are not? The answer to this question is not a simple straightforward one as was shown by the themes and sub-themes that emerged upon analysis of participants‘ discussion of this topic. Notwithstanding this, participants were in agreement that morality is dictated by society, is dependent on religion, and is defined by personal action. These definitions by University of Ghana http://ugspace.ug.edu.gh 202 participants, particularly the fact that morality is dictated by society and defined by personal action, are in line with Gyekye‘s (2003) and Mbiti‘s (1989) definitions of morality. As noted earlier, according to Gyekye, morality refers to a set of social rules and norms that are based on people‘s beliefs about right and wrong conduct and good and bad character and are intended to guide the conduct of people in a society. Also, Mbiti stated that African morality is essentially ‗societary‘ and is dynamic in that it defines what a person does, rather than what he is. The idea that it is dependent on religion is however not necessarily supported by either Mbiti or Gyekye but the latter is in agreement that as participants have suggested, religion does play a role in the moral lives of individuals. It has already been noted also that, according to Gyekye (2003), procreation is a very important African cultural value which is viewed as the whole purpose of marriage. Participants apparently agreed with this view considering the theme discussed above about the unacceptability of homosexuality. Participants noted that homosexuality has implications for human life and therefore social well-being because it does not allow for procreation. In effect, homosexuality is immoral because according to Gyekye, what is morally good is that which brings about human well-being. Additionally, in noting that homosexuality is against the Ghanaian culture, participants were essentially stating again that it is immoral considering Tylor‘s (1958) definition of culture stated earlier. Among other things, culture is said to be a complex whole which includes morals and any other capabilities and habits acquired by man as a member of society. It is also supposed to be a means by which people define their world, express their feelings, and make their judgements, according to Geertz (1973). This being the case, then it is not entirely surprising that participants would make such judgements about homosexuality. In the same vein, it is not surprising that participants would note as part of the unacceptability of homosexuality the fact that it is disgusting. As was explained University of Ghana http://ugspace.ug.edu.gh 203 previously, this feeling of disgust is linked to the idea of anal sex. Going by Terrizzi Jr. et al.‘s (2010) statement that hearing the term homosexuality brings to most people‘s minds a graphic image of two men having anal sex which makes them feel disgusted, it is obvious that participants were stating exactly how they feel and why. It can be concluded therefore that, out of the domains of disgust described by Olatunji (2008), at least two main domains of disgust are related to participants‘ view that homosexuality is unacceptable, including core disgust and animal-reminder disgust. As already noted, Olatunji stated also that disgust appraisals may be more easily associated with gay sexual activity and obviously participants made this association easily. The reason for discussing specific moral issues with participants was to gauge through their opinions, the level of morality of Ghanaians generally and their personal levels of morality. As is obvious from the third theme discussed under the topic morality above, participants agreed that morality is low generally as a result of westernization and modernization, and is seen in how normal premarital sex is now, how extramarital affairs have increased, and how alcohol and hard drug use and abuse have become commonplace. Although some participants stated that they would not engage in certain immoral behaviours such as keeping overchange due to personal principles, it can be safely concluded that they are part of the problem of low morality considering that most of them would easily cheat during examinations and would intentionally avoid paying for items they have bought or keep overchange. Interestingly, they rationalized these behaviours by explaining cheating away as a form of helping behaviour and using distance, not having money, and feeling cheated as excuses. Perhaps even more interesting is the fact that when asked individually which of the immoral behaviours they are likely to engage in, a survey of their responses revealed that they are most likely to engage in examination malpractice, premarital sex, alcoholism, and bribery and corruption. It is therefore not University of Ghana http://ugspace.ug.edu.gh 204 surprising that they agreed that bribery and corruption is commonplace in Ghana. After all, as one of them noted, although they ―are all guilty of maybe keeping overchange and all that stuff‟, all these other immoral stuff ‗doesn‟t have any serious implications on human life‟ like homosexuality does which makes the latter highly immoral as it were. Homosexuality. Participants had varied views on the etiology of homosexuality which shows that they most probably had not read on the subject and were instead relying basically on what they had heard. The fact however remains that even if any of them had given facts that showed they had read, they probably would not have been any more certain of their points. This is because, as noted earlier, although scientific studies have been done to ascertain the etiology of homosexuality, there are two major schools of thought (i.e. biological theories/essentialist beliefs and psychosocial theories/constructionist views) but as Sheldon et al. (2007) noted, neither has yet received the complete approval of the scientific community. Some participants thought that homosexuality was spiritually caused and therefore believed that their interaction with homosexuals should be in the form of getting them spiritual support. This is in line with Owusu et al.‘s (2013) finding that the most prevalent view among their respondents was that homosexuality constitutes a spiritual problem because homosexual persons are demonized. The kind of interaction that respondents in Owusu et al.‘s study reported that they would have with hypothetical homosexual persons is however not similar to what participants of the current study reported as is obvious from the sub-themes on relating to homosexuals interpreted above. While respondents in the former study spoke of caning them or evicting them from their houses, participants in this study spoke of avoiding interaction with homosexual persons altogether, interacting with them cautiously, offering social support, or offering spiritual support. The only exception University of Ghana http://ugspace.ug.edu.gh 205 for respondents in the former study was the fact that they would be less harsh in their interaction with hypothetical homosexual persons who are relatives. One interesting thing that emerged in this study is the fact that any potential interaction with homosexual persons is dependent on the genders of both the homosexual persons and the participants. This is consistent with Anarfi and Gyasi-Gyamerah‘s (2014) finding. Although they found no gender differences in the survey part of their study, with the FGDs, female participants stated that they would be comfortable interacting with male homosexuals while male participants stated that they would be comfortable interacting with female homosexuals. Participants in the current study expressed similar sentiments. This is also consistent with Lim‘s (2008) finding that the women in their study generally reported that they would feel more comfortable working closely with male homosexuals while men reported that they would feel more comfortable working closely with female homosexuals. The fact that participants in the current study also indicated that they would not want to interact with male homosexuals because of their feelings of disgust regarding anal sex is consistent with Arndt and de Bruin‘s (2006) finding that both male and female students were relatively more accepting of lesbians than gay men. Considering all these, it seems quite unlikely that even where contact with homosexuals is fostered, it can generate positive attitudes as confirmed by some studies (e.g. Herek & Capitanio, 1996; Newman, 2007; Sakalli, 2002; Sakalh & Ugurlu, 2002). With regards to identifying homosexuals, as stated previously, Kramer (2003) and Banks (2012) both noted some identification markers noted by the homosexuals they interviewed. Kramer‘s interviewees spoke of two types of homosexual males in Ghana, the classics and the locals, stating that the latter tend to dress stereotypically feminine while Banks‘ interviewees also spoke of two types of homosexual males, those identified as ɛsor and those identified as ase, stating that the latter are usually allowed to perform University of Ghana http://ugspace.ug.edu.gh 206 varying degrees of femininity. From the extracts interpreted above in relation to identifying homosexuals, it is obvious that participants in the current study were basing their identification on the same appearance markers as Kramer‘s locals and Banks‘ ase. The only difference is that Kramer‘s locals as per her interviewees‘ explanations were generally uneducated while participants of the current study were actually speaking of fellow students. So perhaps they were describing male homosexuals who basically fall within Banks‘ description of those who are ase. Considering all these themes in relation to homosexuality, it was noteworthy that some participants felt that since homosexual persons are human beings, they should have the freedom to choose their sexual orientation. Vignettes. As explained previously, the vignettes were designed to depict the persons described in them as homosexual either due to genetics or due to choice (see Table 11). One thing that becomes immediately obvious from the interpretation of the themes relating to them is the fact that while for vignettes 1, 2, and 4, there were themes that pointed to the participants‘ view that the respective vignette persons were not justified in the choices they made, for vignette 3, the issue of justification did not come up at all. Since the vignette person described in vignette 3 had to undergo sex-change to correct a genetic aberration, participants felt that both she and her parents made the right decision given the circumstances. There was not a single reference to the fact that the vignette person can be considered homosexual even though the vignette talks of the person wondering if she can be considered homosexual (see Appendix E). It can be concluded therefore that participants‘ positive sentiments with regards to this vignette person is due to the fact that her circumstances are justifiable and has nothing to do with homosexuality. This is because if it was a matter of genetics solely, then the person described in vignette 1 should also have had participants showing positive University of Ghana http://ugspace.ug.edu.gh 207 sentiments. However, it was the person described in vignette 2 who got some admiration from participants for being strong willed even though his choice to be homosexual was not seen as justified. The pattern of participants‘ sentiments on the various vignette persons is not consistent with Haider-Markel and Joslyn‘s (2008) finding that the perception of homosexuality as controllable (not biological) tended to increase negative affect while the perception of homosexuality as uncontrollable (biological) increased positive affect towards gays and lesbians. As is obvious from the themes under vignette 1, participants felt that if the biological explanations of the vignette person‘s homosexuality were factual then steps should be taken to give him treatment that would correct the problem. The fact that participants felt it would have been more acceptable for the person described in vignette 1 to have collected bribes than to be homosexual shows how negative their attitude is to homosexuals. In addition, given the fact that participants had more positive sentiments towards the person described in vignette 3, it is probable that majority of them also had negative attitudes toward homosexuals. As Boysen and Vogel‘s (2007) findings indicated, participants with initial negative attitudes tended to see biological explanations of homosexuality as more persuasive evidence against the legitimacy and acceptability of homosexuality. It seems this is the case with participants in this study hence the results. It is interesting to note participants‘ views with regards to female homosexuals or lesbians. As stated previously, one of the themes relating to the person described in vignette 2 had to do with the fact that the person would not have had to justify his homosexual behaviour if he were female because lesbianism does not attract as much negative sentiment in the Ghanaian society as gayism. Another reason proffered was that guys enjoy watching lesbians being romantic towards each other or having sexual relations. Additionally, in explaining why they thought sex-change surgery was too University of Ghana http://ugspace.ug.edu.gh 208 drastic a measure for the person described in vignette 4 to have taken, one participant noted that he would rather have chosen to be lesbian. These views point to the fact that attitudes toward female homosexuals are not as negative as those toward male homosexuals. Incorporating findings from Herek‘s (2000) study on sexual prejudice and gender, it is evident that this finding from the current study is consistent with it. Herek found that heterosexuals‘ attitudes toward lesbians differed from their attitudes toward gay men in overall intensity with men‘s attitudes toward gay men being consistently more hostile than their attitudes toward lesbians and also more hostile than women‘s attitudes towards homosexuals of either gender. Based on participants‘ sentiments, this appears to be the case here also considering that these sentiments were extracted from transcripts of all-male groups (2A and 4B). Comparison between findings of IDIs and FGDs. Considering the themes and sub-themes interpreted earlier with regards to both the IDIs and the FGDs, some similarities were found. One such similarity can be seen between FGDs participants‘ extracts under the sub-theme that dealt with total avoidance of interaction with homosexuals and the male IDIs participants‘ extracts under the theme that dealt with reactions by society and fellow students. While the FGDs participants noted that they would avoid interacting with homosexuals because they do not want to be branded homosexual themselves by society, the IDIs male participants gave the same reason for fellow students avoiding them. Another reason given by the FGDs participants was that they wanted to avoid contracting the ‗disease‘ of homosexuality because it is spiritually contagious and the male IDIs participants also confirmed this to be reasons why some fellow students and other members of society would avoid them. Another similarity can be seen between FGDs participants‘ extracts relating to the sub-theme of cautious interaction and the male IDIs participants‘ extracts relating to the University of Ghana http://ugspace.ug.edu.gh 209 sub-theme of wrong perceptions. FGDs participants‘ reasons for being cautious mainly had to do with fear that the homosexual person might entice them or even abuse them sexually. The IDIs participants confirmed this, stating the same fears on the part of fellow students. Interesting similarities can also be found between the FGDs participants‘ extracts relating to the theme that had to do with their uncertainty about the cause of homosexuality and both the male and female IDIs participants‘ thoughts on the cause of their homosexuality. FGDs participants spoke of avoiding pregnancy, history of bad heterosexual relationships, and non-availability of opposite sex relating to single-sex schools and all these were found in the IDIs participants‘ account of the cause of their homosexuality. The main difference was that while some FGDs participants stated emphatically that homosexuality has a spiritual causation, none of the IDIs participants stated that he/she thought their homosexuality is due to any spiritual affliction. There were however some differences also in relation to the cause of homosexuality being biological. FGDs participants generally appeared to be disbelieving about a biological cause, stressing that if there is indeed a biological cause, then a cure can be sought to deal with it physiologically as can be seen from some of the themes interpreted previously under vignettes. Some of the male IDIs participants however noted that they felt their homosexuality was biologically caused although some of them also noted that they did not believe they were born homosexual. It is important to note that none of the female IDIs participants stated that they thought their homosexuality was biologically caused. In this wise, there was a difference between them and the male IDIs participants. Differences between findings from male homosexuals and female homosexuals. Apart from the difference noted earlier, another difference between the male and female IDIs participants was that while the former spoke of their vulnerability (i.e. emotional University of Ghana http://ugspace.ug.edu.gh 210 pain) due to negative reactions of fellow students and their personal discomfort with their homosexuality, the female participants did not share these sentiments. The latter were rather concerned with living up to the societal expectations of marriage and childbirth although they agreed with the male IDIs participants that they often have to use heterosexuality as a cover-up, a fact that is consistent with Sabin et al.‘s (2013) findings. Perhaps this is because as noted previously, female homosexuals are not easily noticeable as male homosexuals since, according to Kramer (2003), the fact that girls/women are affectionate in public whether they are homosexual or not makes it difficult to tell who is, while boys/men are easily noticed if they display the same kind of affection towards each other. As the male IDIs participants noted themselves, a male homosexual is even more noticeable if he appears feminine in mannerism and appearance (i.e. how he dresses). With regards to the issue of heterosexuality as cover-up, it is also consistent with Kramer‘s (2003) finding that some married men also have extramarital affairs with same- sex partners and that some of her interviewees also had heterosexual relationships which for them served to provide the cover they needed to keep their homosexuality a secret. Additionally, having heterosexual relationships ensures that homosexual persons are able to get married in the traditional sense and bear children as expected by society. This is also consistent with Morgan and Wieringa‘s (2005) finding among their Kenyan interviewees who noted that they are pressurized by society to marry and have children. For one of the female homosexual participants in the current study who noted that she preferred not to be labelled, it is most probable that her feelings regarding the inevitability of marriage and having to use heterosexuality as a cover-up might be the reason for avoiding labels. This is consistent with Crowley‘s (2010) finding that among those who preferred no label within the groups she identified on MySpace, one of the reasons they University of Ghana http://ugspace.ug.edu.gh 211 gave for this was that although they experienced attractions to women, they did not want to exclude future relationships with men. With the issue of openness about their orientation, the general sentiment among both male and female homosexual participants was that openness to their families was not an option. Both groups of participants noted that parents, particularly mothers, had their suspicions. For the male participants, these mothers had not asked them directly but for the female participants, the mothers had confronted them and had turned to prayer to deal with what they obviously saw as a spiritual problem. This finding is not entirely surprising considering that Maguen et al.‘s (2002) findings showed that friends were the most cited as persons that their participants would disclose their sexual orientation to. Likewise, as is evident from the above interpretations, although the general sentiment was that they would only disclose to those who share the same orientation, some male participants had disclosed to very close friends who were heterosexual without it affecting their relationship. Conclusion As is evident from the findings of this study, Ghanaians are indeed very religious but are not necessarily committed to their religion. According to the FGDs participants, although religious commitment is generally low within the Ghanaian community as a whole and amongst UG students also, it is linked to religious affiliation in a way that puts the religious commitment of Muslims and practitioners of traditional religions on a higher level than that of Christians. The FGDs participants admitted that their own personal levels of religious commitment were lower than their level of religiousity. Their definition of religious commitment was very much consistent with the definition of Worthington University of Ghana http://ugspace.ug.edu.gh 212 (1988) and their idea that religious commitment means more than being religious was very much consistent with Gyekye‘s (2003) ideas. The participants also noted that morality is generally low, pointing to the fact that cheating in examinations, engaging in premarital and extramarital affairs, alcohol/drug abuse, and bribery and corruption, among others, are so commonplace in Ghana that they are virtually normal. In their definition of morality, participants were again consistent with Gyekye (2003). In spite of their admission of a general and personally low religious commitment and morality, their evaluation of homosexuals was generally unfavourable. This was most evident in the themes that emerged from their discussion of the vignettes. Participants were only sympathetic toward the vignette person described in vignette 3 obviously because they felt that his circumstances were understandable and justifiable while that of the vignette persons described in vignettes 1, 2, and 4 were not. It is probably due to this feeling that participants asked to imagine they were the vignette persons could not easily empathize with the other vignette persons as they did with the vignette 3 person. With regards to the IDIs, the homosexual participants generally confirmed that they are always faced with negative evaluations from the Ghanaian society and their fellow students. The fear of the negative reactions they have received and anticipate receiving from society makes it difficult for them to disclose their sexual orientation to their family members and others with the exception of close friends and friends who are themselves homosexual. The male participants especially noted the emotional pain they feel when they face such negative reactions and how some of them sometimes wish they were not homosexual. Both male and female participants noted that they have to use heterosexuality as a cover-up so as to be accepted by society and avoid undue pressure from family until such time that they cannot avoid the inevitable marriage and childbirth University of Ghana http://ugspace.ug.edu.gh 213 prescribed by society. Finally, comparing findings from both the FGDs and IDIs in relation to the etiology of homosexuality, the following conceptual model (Figure 21) was developed as a proposal. University of Ghana http://ugspace.ug.edu.gh 214 Figure 21. Conceptual model showing a proposed sexual orientation continuum pertaining to Ghanaians. Tendency to be bisexual Exclusive heterosexuality Exclusive homosexuality History of bad heterosexual relationships Single-sex school environment Societal expectations Socialization process Sexual Orientation Continuum Demonic possession Personal characteristics Need to belong and be loved First sexual experience Technology Westernization Peer Pressure Religion Culture Curiosity Fear of Pregnancy Need for Pleasure Characteristics of potential sexual partner University of Ghana http://ugspace.ug.edu.gh 215 Figure 21 demonstrates that the sexual orientation of Ghanaians is not just a straight line continuum with people falling on various points on it but is rather one that sits on the tip of a broad based triangle like a see-saw. The extreme ends of this sexual orientation continuum, exclusive heterosexuality and exclusive homosexuality, are shown in round edge rectangles. The continuum also has some factors hanging on either side of its triangle base like pendulums. These pendulums serve as push or pull factors depending on the circumstances and are shown in big thick-lined rectangles at the ends of dashed lines. They include the socialization process and societal expectations on the one side and the single-sex school environment and history of bad relationships on the other. As can be seen from the figure also, other factors serve as impetus for the push or pull factors, including religion, culture, curiosity, and fear of pregnancy. These are shown in small thin line rectangles hanging on lines joining the pair of the push or pull factors on either side of the triangle base. Additionally, there are some floating factors which serve as added impetus to the swing of the pendulums and are shown freely within the model as their name depicts. Among these floating factors are such things as the need to belong and be loved, personal characteristics, technology, need for pleasure, peer pressure, westernization, demonic possession, first sexual experience, and characteristics of a potential sexual partner. This conceptual model points to the fact that if an individual is brought up through the socialization process and learns what the Ghanaian society expects from him/her, imbibes these quite well with religion and culture serving as the impetus, he/she will have these factors pulling him/her to the side of exclusive heterosexuality. However, if this individual meets certain circumstances including the single-sex school environment and bad heterosexual relationships with curiosity and fear of pregnancy as the impetus, he/she will have these pulling him/her to the side of exclusive homosexuality. Also, if the University of Ghana http://ugspace.ug.edu.gh 216 individual should have all these factors plus all those that serve as impetus on both sides pushing and pulling him/her at the same time, then he/she is likely to be centered at the midpoint of the continuum and therefore have the tendency to be bisexual. It must be noted however that the force of the push or pull factors and the factors that give them impetus are magnified or reduced based on the number of the floating factors that might combine to give them added impetus. It is this added impetus that finally determines any of the outcomes noted above. University of Ghana http://ugspace.ug.edu.gh 217 Chapter Six General Discussion Overview Homosexuality has been a topic for debate for quite some time in Ghana and a lot of negative sentiments have been expressed in the public domain about it. Such debates have occurred mostly in the media, particularly on the Ghanaian airwaves and drawing on the generally negative sentiments expressed, one could surmise that the attitudes of Ghanaians toward homosexuals are negative. Arguments against homosexuality have mainly been that it is immoral, against religious laws and against the Ghanaian culture. According to the literature (see Chapter 2), attitudes toward homosexuality and homosexuals are generally negative across the world and a couple of studies conducted in Ghana confirm this. This study, which was set in UG, was therefore conducted with the main aim of ascertaining whether Ghanaians indeed have negative attitudes toward homosexuals and determining the underlying structure of these attitudes. To achieve this, the methodology for collecting relevant data was designed based largely on Haddock et al.‘s (1993) study but after the initial study (Study 1), results obtained necessitated a second and a third study (Study 2 & Study 3) that have all contributed to progress in the process of gaining a fuller understanding of Ghanaians‘ attitudes toward homosexuals. This three-in-one (3-in-1) study has also given insights into some of the experiences of Ghanaian homosexual persons, providing a background for future research. Summary and Discussion Summary of findings. The first of this 3-in-1 study was designed as a survey and gathered data to test five (5) hypotheses (see Chapter 3). Findings showed that University of Ghana http://ugspace.ug.edu.gh 218 respondents had negative attitudes toward homosexuals based on how they evaluated them and that these attitudes had nothing to do with differences in the respondents‘ personal characteristics. Additionally, respondents‘ stereotypic beliefs were associated with their symbolic beliefs and affect but that this association was proof of only a partial overlapping of information elicited by all three variables. It was deduced based on respondents‘ most elicited responses regarding these variables that the overlaps among them are due to the link between their thoughts about the morality or lack of it on the part of homosexuals and their feeling of disgust towards them. Findings of Study 1 further showed that only respondents‘ symbolic beliefs are the best predictor of their evaluation of homosexuals, explaining the bulk of the variance in their evaluation. It was again deduced based on the respondents‘ most elicited responses that symbolic beliefs are probably the best predictor of respondents‘ unfavourable evaluation of homosexuals because of their view that homosexuals block respect for certain aspects of the Ghanaian culture. Finally, the findings showed that although respondents‘ religious commitment and moral values are associated with their evaluation of homosexuals, they did not moderate the relationship between their attitude components and their evaluation of homosexuals. These variables however had the effect of increasing the extent to which respondents‘ stereotypic and symbolic beliefs together explained the variance in their evaluation of homosexuals. Study 2 was conducted as a follow-up to Study 1 in order to ascertain whether findings of the first study will be replicated and also to determine the role of attributions in participants‘ attitudes toward homosexuals. It was designed as a pre-test post-test experiment (see Chapter 4) and seven (7) hypotheses were tested under it. From analyses of the data gathered, findings indicated that participants evaluated homosexuals more unfavourably than the other target groups and that no significant gender differences University of Ghana http://ugspace.ug.edu.gh 219 existed in both the pre-test and post-test conditions. The findings also indicated that a positive correlation existed among all of the participants‘ stereotypic beliefs, symbolic beliefs, and affect and that these variables altogether explained sixteen percent (16%) of the variance in evaluation of homosexuals. Additionally, the findings indicated that participants‘ religious commitment and moral values moderated the relationship between their attitude components and their evaluation of homosexuals. Finally, the findings of Study 2 indicated that participants‘ evaluation of homosexuals was relatively more favourable in the post-test condition than in the pre-test condition. And, contrary to expectation, only participants in the biological transgendered and control groups had more favourable behavioural intentions towards the vignette persons. It was however inferred that this finding was probably as a result of the combined effects of attributions and biased assimilation. Finding answers for why participants in the biological group had the most unfavourable behavioural intentions therefore formed the basis of conducting Study 3. Study 3 was qualitative in nature (see Chapter 5), consisting of FGDs with presumed heterosexual students and IDIs with known homosexual students. Some of the themes that emerged from analyses of the FGDs transcripts pointed to the fact that Ghanaians are very religious but are not necessarily committed to their religion. Participants indicated that religious commitment is generally low within the Ghanaian society and also amongst UG students and they noted that Muslims and practitioners of the traditional religions are more committed to their religion than Christians. The FGDs participants also indicated that their personal levels of religious commitment were lower than their level of religiousity. Additionally, themes emerged which indicated that morality is generally low with such things as cheating in examinations, engaging in premarital and extramarital affairs, abusing alcohol/hard drugs, and bribery and corruption University of Ghana http://ugspace.ug.edu.gh 220 being said to be commonplace and virtually normal in Ghana now, most of them due to justifiable reasons. Despite their admission of a general and personally low religious commitment and morality, participants‘ evaluation of homosexuals was generally unfavourable, a fact that was most evident in the themes that emerged from their discussion of the vignettes. Only the person described in vignette 3 won the sympathy of participants due to the fact that they obviously felt that his circumstances were understandable and justifiable than that of the other vignette persons. This negative evaluation of homosexuals by the FGDs participants was generally confirmed by the IDIs participants. Some of the themes that emerged from analyses of the IDIs participants‘ transcripts pointed to the fact that they are always faced with negative evaluations from the Ghanaian society and their fellow students. Apparently, anticipation of such negative reactions from society makes it difficult for them to disclose their sexual orientation to their family members and others with the exception of homosexual friends and close heterosexual friends. Some of the themes that emerged from analysing transcripts of the male homosexuals spoke of the emotional pain they feel when faced with such negative reactions and how they sometimes wished they were not homosexual. Although the female homosexual participants did not speak of experiencing such emotional pain, they agreed with the male participants that they often resort to using heterosexuality as a cover-up so as to be accepted by society. Such cover-ups are apparently also necessitated by the need to avoid being pressurized by family members, especially parents, to get married and thereby delay the inevitable marriage and childbirth that is prescribed by the Ghanaian society. Discussion and relatedness of findings to the theoretical framework. How do all the findings summarized above link with each other in fulfilling the aims and University of Ghana http://ugspace.ug.edu.gh 221 objectives of this 3-in-1 study and how do they relate to the theoretical framework for the whole study? This is the focus of this part of the general discussion of the findings of this study. Linkages between studies 1 and 2. As noted earlier, one of the main reasons for conducting Study 2 was to determine whether the results of Study 1 will be replicated and so the first four hypotheses under it were set for this purpose. Considering the above summary, it is obvious that participants in Study 2 evaluated homosexuals more unfavourably than the other target groups thereby replicating this finding in Study 1. However, a closer look at the means of respondents‘ evaluations in Study 1 (see Table 2) and participants‘ evaluations in Study 2 (see Table 14) reveals the fact that there were some marginal differences that were lost in approximately plotting the means on the evaluation thermometer (see Figures 2 & 4). For example, homosexuals were evaluated a little more unfavourably in Study 2 than in Study 1 while Orthodox Church Priests were evaluated a little less favourably in Study 2 than in Study 1. Another finding in Study 2 that replicated that of the first is the finding that there is a positive correlation between respondents‘ stereotypic beliefs, symbolic beliefs, and affect but for this too, there was a difference. The difference is that while for Study 1 a positive correlation was found among respondents‘ stereotypic beliefs and their symbolic beliefs and affect but not among their symbolic beliefs and their affect, for Study 2, all three attitude components were positively correlated with each other. As such, although the results of Study 1 was in contradiction to Haddock et al.‘s (1993) study where all three variables were positively correlated with each other, that of Study 2 was consistent with their finding. It was however noted that this does not mean that the variables elicited information that were completely overlapping in both studies. Neither of these findings is improbable considering that as discussed in University of Ghana http://ugspace.ug.edu.gh 222 Chapter 2, the Multicomponent Model of Attitudes states that the affective, cognitive and behavioural components of attitudes are interrelated and yet are independent of each other to some extent. In addition to testing for correlations between the attitude components, for both studies 1 and 2, participants‘ answers to questions regarding these attitude components were collated and it was found that the most frequently elicited responses for stereotypic beliefs were that homosexuals are not religious, are immoral, and evil. This was the same for Study 1. For symbolic beliefs, participants in Study 2 perceived homosexuals as blocking morality, blocking marriage, and promoting respect for human rights while in Study 1 respondents perceived homosexuals as blocking respect, blocking morality, and promoting materialism. Obviously, the two studies have only morality as the common response between them and even so while Study 2 had it as the first most elicited response, Study 1 had it as the second most elicited response. With respect to participants‘ affect, the feelings or emotions stated as most elicited by homosexuals when they see, meet, or think about them were feelings of anger, unhappiness, and disgust for both studies although not in the same order (see Tables 6 & 17). Despite these differences, it can be said that the responses elicited in both studies show a support for the conclusion that it is not a totally improbable outcome that although the attitude components were correlated with each other, they do not elicit completely overlapping information. Additionally, morality or the lack of it on the part of homosexuals, was stated by participants for both stereotypic beliefs and symbolic beliefs in Study 2. As already established in Chapter Three, the fact that morality underlies participants‘ feelings of disgust goes beyond mere probability, considering that the link between these factors may be the point at which the attitude components partially overlap and considering also that the literature supports this assumption (e.g. Chapman et al., 2009; Olivera La Rosa & University of Ghana http://ugspace.ug.edu.gh 223 Rossellό-Mir, 2013; Schnall et al., 2008). Although morality and disgust do not occur in the same pattern in both studies, the same assumption can be said to be at play in Study 2 due especially to the fact that positive correlations were found amongst all the attitude components. As such, it can be concluded that the overlaps in information elicited by the various attitude components in both studies are driven mainly by participants‘ stereotypic beliefs and therefore where participants felt that homosexuals were immoral, they likewise saw them as blocking morality and therefore felt disgusted at them. The third hypothesis tested in determining whether findings of Study 1 replicate in this second study is the hypothesis that stereotypic beliefs, symbolic beliefs, and affect are likely to predict respondents‘ evaluation of homosexuals. In the first study, the variance explained by all the predictor variables was 10% while it was 16% in the second study and so there is a difference of 6% between findings from the two studies. Added to this difference is the fact that while the bulk of the variance in respondents‘ evaluation of homosexuals in Study 1 was explained by their symbolic beliefs, in Study 2 it was explained by participants‘ stereotypic beliefs and their affect. This is because only symbolic beliefs predicted respondents‘ evaluation of homosexuals in Study 1 to the exclusion of stereotypic beliefs and affect while in the second study, both stereotypic beliefs and affect rather predicted participants‘ evaluation of homosexuals to the exclusion of symbolic beliefs. So, unlike was the case for Study 1, findings of Study 2 was consistent with that of Haddock et al.‘s (1993) study in that with the addition of the evaluative implications of affect to stereotypic beliefs, the variance in evaluation of homosexuals was explained 6% more than symbolic beliefs did in Study 1. In Study 1, it was expected that respondents‘ religious commitment, moral values, and personal morals will moderate the relationship between their attitude components and their evaluation of homosexuals. This was however not the case. Nonetheless, the same University of Ghana http://ugspace.ug.edu.gh 224 hypothesis was tested again in Study 2 and interestingly, it was confirmed with the exception of personal morals. One might however wonder why the same effect was not found in Study 1 but the answer to that is quite obvious. As explained previously, the moderated regression analysis done to test this hypothesis in Study 2 was done a bit differently. Each attitude component was analyzed with each of the suspected moderator variables separately and all the variables were included in the analyses whether they were associated with the evaluation of homosexuals or not (see Chapter 4). The reason for this change was because of the suspicion that the moderating effects of the moderator variables might not have been fully realized in the first study because they might have had some control effects on each other. Therefore, if making this change produced these results, then the earlier suspicion that the moderator variables might have had some control effects on each other is no longer a suspicion but a fact. Although Study 1 did not find any gender differences in attitudes toward homosexuals, the effect of gender was still tested under Study 2 and it was found that in spite of the experimental conditions, findings of Study 1 was replicated in this instance also. In discussing the lack of gender differences in Study 1, the question was raised as to whether the finding is similar to that of Ben-Ari (2001) in that the lack of gender differences is because respondents are psychology students. Participants in Study 2 were also psychology students but as noted earlier, it is more probable that the lack of gender differences in both studies is rather consistent with Anarfi and Gyasi-Gyamerah‘s (2014) assertion that this is most likely due to differences in culture. As was noted in Chapter Two, according to the Social Learning Theory, as humans we acquire or learn our social behaviour, in other words our culture, including prejudicial and religious attitudes, by observing significant others and imitating them. We then keep exhibiting the behaviours we have learnt or refrain from them depending on whether we are rewarded or punished University of Ghana http://ugspace.ug.edu.gh 225 for these behaviours. Considering that both male and female Ghanaians are socialized within the same cultural environment, it is really not surprising that there are no differences in their attitudes. Linkages between all three studies. The socialization process can also be said to be the reason why in Study 1 it was found that respondents generally had negative attitudes toward homosexuals regardless of religion. As noted earlier, Spilka et al. (2003) stated that religiousness is typically strongly influenced by one‘s immediate environment through both modeling and reinforcement processes with parents especially being the main agents. Obviously, this explains some findings of Study 3 also and gives an understanding of why their attitudes, considering their reactions towards the various vignette persons, were generally negative towards homosexuals. The FGDs participants noted that Ghanaians, including themselves, are very religious and that it permeates all aspects of life in Ghana so much so that it can be said to be part of the culture. According to Spilka et al. (2003), many other external influences have the potential to affect people‘s religiousness including peers and school so it is not surprising that participants in Study 3 attested to the fact that students on UG campus are also religious. They however noted that both religious commitment and morality are low and considering these, one would have imagined that the FGDs participants would be less judgmental of homosexuals. The fact that they were not points to the power of the socialization process. The IDIs participants confirmed this with their references to the negative reactions they get from society and the cultural demands on them to be what their families and society expects them to be, which is being heterosexual and eventually getting married and having children. As noted in Chapter 4, Study 2 also tested the effects of attributions on behavioural intentions to examine the behavioural component of attitudes and thus completing the University of Ghana http://ugspace.ug.edu.gh 226 circle of enquiry as put forth by the Multicomponent Model of Attitudes. Contrary to expectation, the finding was that participants in the biological transgendered group had the most favourable behavioural intentions but those in the biological group had the most unfavourable behavioural intentions. The effect of attributions were tested again in Study 3 by making the FGDs participants read the same vignettes and discuss them and the findings confirmed that of Study 2. Participants in both studies were empathetic towards the person described in vignette 3 (see Appendix E). Considering these findings casually, one would say that it is contrary to what the Attribution theory of Controllability states (Weiner, 1979, 1985; Weiner et al., 1988). The attribution theory, as previously noted, states that stigmatized behaviours that are believed to have biological origins will be evaluated positively while those that are believed to be due to individual choice will be evaluated negatively. The theory further explains stigmatized behaviours believed to have biological origins as those that are seen to be caused through circumstances beyond a person‘s control. Since this is the case, and considering the obvious fact that participants in both studies thought the vignette 3 person definitely had circumstances beyond his/her control because he/she had ‗genuine‘ biological reasons, such participants thought that he/she deserved a positive evaluation and therefore had favourable behavioural intentions towards this person. In effect, findings from both Study 2 and Study 3 are technically consistent with the dictates of the attribution theory. The reason is that, going by the findings made in Study 3 regarding the same vignette (i.e. Vignette 1; see Appendix E) used for the biological group in Study 2, the FGDs participants felt that if the vignette person‘s homosexuality is biological, then efforts should be made to correct his/her anomalies so that he/she can become ‗normal‘. FGDs participants in Study 3 virtually refused to accept that the vignette 1 person‘s homosexuality is biologically caused, a fact that is further University of Ghana http://ugspace.ug.edu.gh 227 evidenced by the themes that emerged from their discussion of the etiology of homosexuality. They noted that the vignette 1 person had no justification for his/her homosexuality (see Chapter 5). If biological causation is not believable, then the vignette 1 person‘s situation is similar to that of the vignette 2 person who chose to be homosexual. In that case, it follows then that the vignette 1 person should be treated as one who is believed to have caused his stigma, which participants in both studies did, and this is consistent with the attribution theory‘s explanation that individuals who are believed to have caused their stigma are evaluated negatively. Strengths and Limitations of the Study Strengths. One of the main strengths of this 3-in-1 study is the fact that it made use of mixed methods in its quest to realize its aims and objectives as discussed above under summary of the findings. When the related literature reviewed in Chapter 2 is juxtaposed against the methodology of the current study, it highlights the strengths of this study more prominently. For instance, it was noted previously that Haddock et al.‘s (1993) study is different in the sense that it does not make use of attitude scales like most studies on attitudes toward homosexuality or homosexuals do. The current study used a similar methodology especially in terms of using a single attitude measure (the attitude thermometer) and presenting homosexuals as part of a list of five groups that participants had to evaluate (see Appendix A). There are however some important differences that were aimed at getting information thought to be important in this setting and for this study (see methodology sections under Chapters 3, 4, & 5). While Haddock et al. looked at RWA and how it relates to the components of attitudes toward homosexuals, the current study looked at how religious commitment, morality, and attributions relate to these components. University of Ghana http://ugspace.ug.edu.gh 228 This study is also similar to that of Negy and Eiseman‘s (2005) study in that it was interested in student‘s affective reactions to homosexuals as part of unraveling the structure of prejudicial attitudes toward homosexuals in Ghana but unlike theirs, formal hypotheses were stated in this study. To test these hypotheses, primary data had to be collected through a carefully designed methodology that was thought to be the best way to get the necessary information, given the constraints that were faced during the process (see methodology sections under Chapters 3, 4, & 5). Some of the studies reviewed in Chapter 2 did not collect primary data but rather reviewed existing literature and made deductions based on their review (e.g., Haider-Markel & Joslyn, 2008; Loftus, 2001; Ohlander et al., 2005; Sherkat, 2002). The fact that this study not only gathered primary data but did so using different methods to get a broader understanding of the variables also shows its strength. Additionally, in consistence with Boysen and Vogel‘s (2007) recommendation that future research should use other sociocultural explanations of homosexuality so as to allow for greater generalizability of findings, the second part of this study made participants read various vignettes with different explanations of homosexuality showing controllability or uncontrollability after an initial assessment of their attitudes toward homosexuals. It is worth noting that it also made use of experimental manipulations although the variables of interest do not include popular culture images as was recommended by Thompson (2008) (see Chapter 4). Finally, although some studies rather analyzed the interactions of sexual minority women on certain social media (e.g., Crowley, 2010) instead of conducting actual interviews with them and some could not recruit any sexual minority person (e.g., Owusu et al., 2013), the third part of the current study was able to do this. Both male homosexuals (gays) and female homosexuals (lesbians) were interviewed and this afforded the opportunity to glean information about University of Ghana http://ugspace.ug.edu.gh 229 same-sex behaviours and some related issues to complement findings on attitudes of presumed heterosexuals toward homosexuals. It is therefore safe to say that as much as was possible, this study gathered data in a scientific manner to ensure that findings made are a true reflection of the actual situation with regards to the attitudes of Ghanaians toward homosexuals. Limitations and recommendations. In spite of these strengths, the study has some limitations. Firstly, since the setting for the study was the University of Ghana (UG) and only students were interviewed, there is the possibility that the findings may not reflect the attitudes of other members of the Ghanaian society. UG students were used for the sake of proximity and because numerous other studies have used university or college students (see Chapter Two). Additionally, considering that UG is a public university, it was thought that it would most likely have a fair representation of people from all parts of Ghana schooling there. However, one cannot overlook the fact that having only students from one setting albeit from different parts of the country as participants may not necessarily equal having a cross section of the general population as participants. It is therefore recommended that future research looks at getting a more representative sample as this might bring out some findings that might have been lost in this study. Secondly, the instruments used for data collection in Studies 1 and 2 were all developed in the West and therefore their reliability and validity in the Ghanaian context can be called into question. This is despite the fact that where necessary, they were modified to suit the demands of this study, were pretested and were found to have high Cronbach‘s alphas with the local sample (see the methodology sections in chapters 3 and 4). It is therefore recommended that future studies look at standardizing the instruments. When this is done, subsequent studies can use the instruments with the certainty that their reliability or validity cannot be called into question. University of Ghana http://ugspace.ug.edu.gh 230 Thirdly, in Study 3, a distinction was made between gays and lesbians but this was not done Studies 1 and 2. Some differences in participants‘ attitudes toward gays and toward lesbians were noted in Study 3. Considering this, it is probable that if this distinction had also been made in Studies 1 and 2, some differences in attitudes might have been noted. It is therefore recommended that future studies examine attitudes toward gays and lesbians separately as there might even be differences in the structure of these attitudes. Future studies could also do an initial assessment of potential participants to determine their sexual orientation instead of making presumptions about their heterosexuality as was done in this study. By so doing, findings made with regards to their attitudes can be said to be strictly that of heterosexuals toward homosexuals. Future studies should as well explore the issue of the effect of attributions and biased assimilation on attitudes toward homosexuals into more detail using quantitative methods. Finally, although this study also interviewed both male and female homosexuals as part of the third study, this was done solely in order to determine if they would confirm findings gotten from the presumed heterosexuals. Future studies could therefore explore into more detail the lived experiences of gays and lesbians not only on campus, but in the general Ghanaian society using both quantitative and qualitative methods. Additionally, future studies can explore into more detail the issue of sexuality in all its forms in the same manner. By so doing, such studies would be able to test the proposed model on the formation of sexual orientation of Ghanaians offered based on the findings of this study (see Figure 21). Other Recommendations It was noted in Chapter Two that in Lim and Johnson‘s (2001) assessment of the levels of homophobia among social work students in the Republic of South Korea, they University of Ghana http://ugspace.ug.edu.gh 231 found that compared to U.S. samples, the students had high levels of homophobia but that in-class discussion of homosexuality was found to be significantly associated with lower levels of homophobia. Based on this and the findings from the third part of this study, it is recommended that UG take a look at developing courses, especially in the humanities, that would deal with the issue of human sexuality. This would afford students the opportunity to learn facts about human sexuality that would at least ensure that sexual minority students are not victimized in any way even if it would not affect attitudes toward homosexuality positively. Such courses would also help students, whatever their sexual orientation, learn how to be sexually responsible for their own health and for the health of their partners. In line with this, it is also recommended that a sexuality forum be organized periodically, every other academic year, where both staff and students can have the opportunity to discuss issues relating to human sexuality including physical health, psychological wellbeing, religion, morality, gender discrimination, sexual molestation, and sexual harassment. In the meantime, it might be prudent for the university to enforce rules and regulations governing the halls and hostels of residence on campus so that students would not get the opportunity to engage in certain behaviours within their rooms that would infringe on the rights and sensibilities of their fellow roommates. Conclusion and Final Observed Model Considering the findings from this study and the above discussion, it can be concluded that Ghanaians, particularly UG students, indeed have prejudicial attitudes toward homosexuals. These prejudicial attitudes are moderated by moral values and religious commitment and are also affected by attributions and biased assimilation. Based University of Ghana http://ugspace.ug.edu.gh 232 on this conclusion, the following conceptual model, Figure 22, was developed to reflect all the findings of this 3-in-1 study. University of Ghana http://ugspace.ug.edu.gh 233 Figure 22. Observed model showing findings of all parts of this 3-in-1 study. Moral Values Affect Behavioural Intentions Prejudicial Attitudes toward Homosexuals Stereotypic Beliefs Symbolic Beliefs Religious Commitment University of Ghana http://ugspace.ug.edu.gh 234 As seen from Figure 22, all the variables are shown in rectangles of different sizes that are solely meant to accommodate the variable names. The interaction effects between variables are depicted by solid double-tipped arrows having both a diamond tip and a triangular tip at opposite ends. The existence of a correlation or association between variables is depicted by a normal double-tipped arrow while solid double-tipped arrows depict a two-way effect. As noted previously, this study was conducted with the expectation that stereotypic beliefs, symbolic beliefs, affect, personal characteristics, morality and religious commitment would have direct effects on attitudes toward homosexuals. There was also the expectation that this study would show that while controlling for the effect of personal characteristics, moral values, personal morals and religious commitment would have moderating effects on the effects of affect, stereotypic beliefs, and symbolic beliefs on attitudes toward homosexuals (see Figure 1). This final observed model (Figure 22) demonstrates that some aspects of the conceptual model (Figure 1) were confirmed while others were not. As the figure shows, participants‘ prejudicial attitudes toward homosexuals is mainly driven by their stereotypic beliefs with the added evaluative implications of their symbolic beliefs and affect plus the moderating effects of moral values and religious commitment. It also shows that participants‘ behavioural intentions are influenced by their prejudicial attitudes while prejudicial attitudes in turn also do influence behavioural intentions. As noted previously, participants‘ behavioural intentions are due to the attributions they make about the causes of homosexuality. The effect of the attributions is such that if biological explanations for homosexuality are evidenced by anatomical deformities, then it is acceptable especially when the person can take steps to correct it. However, if there are no obvious anatomical deformities, then homosexuality is not justifiable and therefore is University of Ghana http://ugspace.ug.edu.gh 235 unacceptable especially when the person refuses to take steps to help him/her conform to societal expectations. University of Ghana http://ugspace.ug.edu.gh 236 References Akers, R. (1998). Social learning and social structure: A general theory of crime and deviance. Boston: North Eastern University Press. Akers, R., Krohn, M., Lanza-Kaduce, L., & Radosevich, M. (1979). Social learning and deviant behavior: A specific test of a general theory. American Sociological Review, 44, 636-655. Altemeyer, B. (2001). Changes in attitudes toward homosexuals. Journal of Homosexuality, 42, 63–75 Anarfi, J. K., & Gyasi-Gyamerah, A. A. (2014). Religiosity and attitudes toward homosexuality: Views of Ghanaian university students. Research in the Social Scientific Study of Religion, 25, 173-212. Atiemo, A. O. (2013). Religion and the Inculturation of Human Rights in Ghana. Bloomsbury: London. Attipoe, D. (2004, June 24). MSM and HIV in Ghana. The Gully. Retrieved May 9, 2012 from http://www.thegully.com/essays/gaymundo/0403_gaymen_hiv_ghana/msm_ghana_intro_ summary.html Arndt, M., & De Bruin, G. (2006). Attitudes toward lesbians and gay men: Relations with gender, race and religion among university students. Psychology in Society, 33, 16-30. Bailey, J. M. & Pillard, R. C. (1991). A genetic study of male sexual orientation. Archives of General Psychiatry, 48, 1089-1096. Baron, R. A., Branscombe, N. R., & Byrne, D. (2009). Social Psychology. (12th ed.). Boston: Allyn and Bacon. Banks, W. D. (2012). Remembering Ɔkɔmfo Kwabena: ‗Motherhood‘, spirituality, and queer leadership in Ghana. African Historical Review, 44(2), 1-17. University of Ghana http://ugspace.ug.edu.gh 237 BBC NEWS AFRICA, (2011). Uganda Fury at David Cameron aid threat over gay rights. Retrieved March 19, 2012 from http://www.bbc.co.uk/news/world/africa/ Ben-Ari, A. T. (2001). Homosexuality and heterosexism: Views from academics in the helping professions. British Journal of Social Work, 31, 119 – 131. Bem, D. J. (1996). Exotic becomes erotic: A developmental theory of sexual orientation. Psychological Review, 103, 320-335. Boysen, G. A., & Vogel, D. L. (2007). Biased assimilation and attitude polarization in response to learning about biological explanations of homosexuality. Sex Roles: A Journal of Research, 57, 755-762. Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative Research in Psychology, 3(2), 77-101. Briñol, P., & Petty, R. E. (2003). Overt head movements and persuasion: a self-validation analysis. Journal of personality and social psychology, 84(6), 1123. Brown, W. M., Fin, C. J., Cooke, B. M., & Breedlove, S. M. (2002). Differences in finger length ratios between self-identified ―butch‖ and ―femme‖ lesbians. Archives of Sexual Behavior, 31, 123-127. Byer, C. O., Shainberg, L. W., & Galliano, G. (1999). Dimensions of human sexuality (5th ed.). Boston: McGraw-Hill. Cacioppo, J. T., Priester, J. R., & Berntson, G. G. (1993). Rudimentary determinants of attitudes: II. Arm flexion and extension have differential effects on attitudes. Journal of personality and social psychology, 65(1), 5. Carter, D. (2005). Stonewall: The rebellion that Sparked the Gay Revolution (First ed.). New York: Macmillan. ISBN 0-312-34269-1. University of Ghana http://ugspace.ug.edu.gh 238 Cass, C. V. (1979). Homosexual identity formation: A theoretical model. Journal of Homosexuality, 4, 219-235. Chaiken, S. & Baldwin, M. W. (1981). Affective-cognitive consistency and the effect of salient behavioural information on the self-perception of attitudes. Journal of Personality and Social Psychology, 41, 1-12. Crowley, M. S. (2010). Experiences of young bisexual women in lesbian/bisexual groups on MySpace. Journal of Bisexuality, 10(4), 388–403. Criminal Code (Amendment) Act of 1960, 646 Acts of Ghana Code, §104 (2003). Davis, C. M., Yarber, W. L., Bauserman, R., Schreer, G., & Davis, S. L. (Eds.). (1998). Handbook of sexuality-related measures. California: Sage Publications, Inc. Dover, K. J. (1989). Greek Homosexuality. Cambridge, MA: Harvard University Press. Drescher, J. (2009). Queer diagnoses: Parallels and contrasts in the history of homosexuality, gender variance, and the Diagnostic and Statistical Manual. Archives of Sexual Behavior, doi:10.1007/s10508-009-9531-5. Drobac, J. (1999). Pansexuality and the law. William and Mary Journal of Women and Law, 5(2), 297-308. Eagly, A. H. & Chaiken, S. (1993). The psychology of attitudes. Fort Worth, TX: Harcourt Brace Jovanovich. Esses, V. M., Haddock, G., & Zanna, M. P. (1993). Values, stereotypes, and emotions as determinants of intergroup attitudes. In D. M. Mackie & D. L. Hamilton (Eds.), Affect, cognition, and stereotyping: Interactive processes in group perception (pp. 137-166). New York: Academic Press. University of Ghana http://ugspace.ug.edu.gh 239 Féray, J.-C., & Herzer, M. (1990). Homosexual Studies and Politics in the 19th Century: Karl Maria Kertbeny. Journal of Homosexuality, 19(1), 23-48. Festinger, L. (1957). A theory of cognitive dissonance. Stanford, CA: Stanford University Press. Fox, R. C. (1995). Bisexual identities. In L. D. Garnets & D. C. Kimmel (Eds.), Psychological perspectives on lesbian, gay, and bisexual experiences (pp. 86-129). New York: Columbia University Press. Gallo, M. (2006). Different Daughters: A History of the Daughters of Bilitis and the Rise of the Lesbian Rights Movement. Carrol & Graf Publishers. ISBN 0-7867-1634-7 Geertz, C. (1973). The Interpretation of Cultures. New York: Basic Books. GhanaToGhana.com, (2012). Angry youth destroy same sex marriage ceremony in James Town and chase the two lesbians. Retrieved March 20, 2012 from http://www.ghanatoghana.com/Ghanahomepage/ Gulia, K. K., & Mallick, H. N. (2010). Homosexuality: a Dilemma in discourse! Indian Journal of Physiological Pharmacology, 54 (1), 5 – 20. Gyekye, K. (2003). African Cultural Values: An Introduction. Accra: Sankofa Publishing Company. Haddock, G. & Maio, G. R. (2008). Attitudes: Content, structure and functions. In M. Hewstone, W. Stroebe, & K. Jonas (Eds.), Introduction to Social Psychology: A European Perspective (4th ed.) (pp. 112-133). Oxford, United Kingdom: Blackwell Publishing Ltd. Haddock, G., Zanna, M. P., & Esses, V. M. (1993). Assessing the structure of prejudicial attitudes: The case of attitudes toward homosexuals. Journal of Personality and Social Psychology, 65(6), 1105 – 1118. University of Ghana http://ugspace.ug.edu.gh 240 Haider-Markel, D. P. & Joslyn, M. R. (2008). Beliefs About the Origins of Homosexuality and Support for Gay Rights: An Empirical Test of Attribution Theory. Public Opinion Quarterly 72, 291-310. Haslam, N., & Levy, S. R. (2006). Essentialist beliefs about homosexuality: Structure and implications for prejudice. Personality and Social Psychology Bulletin, 32, 471–485. Hay, H., with W. Roscoe (ed.) (1996). Radically Gay: Gay Liberation in the Words of its Founder. Boston: Beacon Press. ISBN 0-8070-7080-7. OCLC 33333896. Herek, G. M. (2000). Sexual prejudice and gender: Do heterosexuals‘ attitudes toward lesbians and gay men differ? Journal of Social Issues, 56(2), 251-266. Herek, G. M. (2000). The psychology of sexual prejudice. Current Directions in Psychological Science, 9(1), 19-22. Herek, G. M. (2007). Confronting sexual stigma and prejudice: Theory and practice. Journal of Social Issues, 63, 905-925. Herek, G. M. & Capitanio, J. P. (1996). ―Some of my best friends‖: Intergroup contact, concealable stigma, and heterosexuals‘ attitudes toward gay men and lesbians. Personality and Social Psychology Bulletin, 22(4), 412-424. Hofstee, E. (2006). Constructing a Good Dissertation. Johannesburg: EPE. Hooker, E. (1957). The adjustment of the male overt homosexual. Journal of Projective Techniques, 21, 18-31. Katz, J. (1995). The Invention of Heterosexuality. New York: Dutton. Kawakami, K., Dion, K. L. & Dovidio, J. F. (1998). Racial prejudice and stereotype activation. Personality and Social Psychology Bulletin, 24, 407-416. University of Ghana http://ugspace.ug.edu.gh 241 Kessler, T. & Mummendey, A. (2008). Prejudice and intergroup relations. In M. Hewstone, W. Stroebe, K. Jonas (Eds.), Introduction to Social Psychology: A European Perspective (4th ed.) (pp. 290-314). Oxford, United Kingdom: Blackwell Publishing Ltd. Kottak, C. P. (2004). Anthropology: The Exploration of Human Diversity (10th ed.). New York: McGraw-Hill Co. Inc. Kramer, I. B. (2003). The things we don‘t talk about: Homosexuality among university students at Legon and Cape Coast (Unpublished undergraduate dissertation). Cornell University, Ithaca, New York. Krosnick, J. A., Betz, A. L., Jussim, L. J., & Lynn, A. R. (1992). Subliminal conditioning of attitudes. Personality and Social Psychology Bulletin, 18, 152-162. LeVay, S. (1991). A difference in hypothalamic structure between heterosexual and homosexual men. Science, 253, 1034-1037. LeVay, S. & Valente, S. M. (2003). Human Sexuality. Sunderland: Sinauer Associates, Inc. Licht, H. & Dawson, L. H. (Ed.) (1969). Sexual Life in Ancient Greece (J. H. Freese, Trans.). Panther/Harper Collins Distribution. Lim, V. K. G. (2002). Gender differences and attitudes toward homosexuality. Journal of Homosexuality, 43(1), 85-97. Lim, H. S., & Johnson, M. M. (2001). Korean social work students‘ attitudes toward homosexuals. Journal of Social Work Education, 37(3), 545-554. Loftus, J. (2001). America‘s liberalization in attitudes toward homosexuality, 1973 to 1998. American Sociological Review, 66(5), 762-782. University of Ghana http://ugspace.ug.edu.gh 242 Maguen, S., Floyd, F. J., Bakeman, R., & Armistead, L. (2002). Developmental milestones and disclosure of sexual orientation among gay, lesbian, and bisexual youths. Applied Developmental Psychology, 23, 219–233. Manusov, V., & Spitzberg, B. H. (2008). Attributes of attribution theory: Finding good cause in the search for theory. In D. O. Braithwaite & L. A. Baxter (Eds.), Engaging theories in interpersonal communication (pp. 37-49). Thousand Oaks, CA: Sage. Mbiti, J. S. (1989). African Religions and Philosophy (2nd Ed.). Doubleday: New York. Mish, F. C., Gilman, E. W., Lowe, J. G., McHenry, R. D., Pease, Jr. R. W., Bollard, J. K., … Perrault, S. J. (Eds.) (1990). Webster‟s Ninth New Collegiate Dictionary (9th ed., Vols. 1- 9). Massachusetts: Merriam-Webster Inc. Mohr, J. J., & Sedlacek, W. E. (2000). Perceived barriers to friendship with lesbians and gay men among university students. Journal of College Student Development, 41, 70–80. Morgan, R., & Wieringa, S. (Eds.) (2005). Tommy boys, lesbian men, and ancestral wives: Female same-sex practices in Africa. Jacana Media. Murphy, S. T. & Zajonc, R. B. (1993). Affect, cognition, and awareness: Affective priming with optimal and suboptimal stimulus exposures. Journal of Personality and Social Psychology, 64, 723-739. Myers, D. G. (2002). Social Psychology. Boston: McGraw-Hill Co. Inc. Negy, C., & Eisenman, R. (2005). A comparison of African American and White college students‘ affective and attitudinal reactions to lesbian, gay, and bisexual individuals: An exploratory study. Journal of Sex Research, 42(4), 291-298. Newman, B. S. (2007). College students' attitudes about lesbians: What difference does 16 years make?. Journal of homosexuality, 52(3-4), 249-265. University of Ghana http://ugspace.ug.edu.gh 243 Ohlander, J., Batalova, J., & Treas, J. (2005). Explaining educational influences on attitudes toward homosexual relations. Social Science Research, 34, 781–799. Olatunji, B. O. (2008). Disgust, scrupulosity and conservative attitudes about sex: Evidence for a meditational model of homophobia. Journal of Research in Personality, 42, 1364- 1369. Olivera La Rosa, A., & Rossellό-Mir, J. (2013). On the relationships between disgust and morality: A critical review. Psicothema, 25(2), 222-226. Owusu, A. Y., Anarfi, J. K., & Tenkorang, E. Y. (2013). Attitudes and views on same-sex sexual behavior in Ghana. Global Advanced Research Journal of Social Science (GARJSS), 2(8), 176-186. Peplau, L. A. & Cochran, S. D. (1990). A relationship perspective on homosexuality. In D. P. McWhirter, S. A. Sanders, & J.M. Reinisch (Eds.), Homosexuality/heterosexuality: Concepts of sexual orientation (pp. 321-49). New York, NY: Oxford. Pickett, B. (2011). Homosexuality. In E.N. Zalta (Ed.), The Stanford encyclopedia of philosophy (Spring 2011 edition). Retrieved from http://plato.stanford.edu/archives/spr2011/entries/homosexuality Rettig, S., & Pasamanick, B. (1959). Changes in moral values among college students: A factorial study. American Sociological Review, 24, 856 – 863. Robertson, P. K. (2004). The historical effects of depathologizing homosexuality on the practice of counseling. The Family Journal, 12, 163-169. Sabin, L., Adu-Sakordie, Y., Agyarko-Poku, T., Ashigbie, P., Hollenbeck-Pringle, D., DeSilva, M. B.,… Beard, J. (2013). Attitudes and behaviours among older MSM in Ghana. Research Report, Boston, Massachusetts. University of Ghana http://ugspace.ug.edu.gh 244 Sakalh, N., & Ugurlu, O. (2002). Effects of social contact with homosexuals on heterosexual Turkish university students' attitudes towards homosexuality. Journal of homosexuality, 42(1), 53-62. Sakalli, N. (2002). The relationship between sexism and attitudes toward homosexuality in a sample of Turkish college students. Journal of Homosexuality, 42(3), 53-64. Schnall, S., Haidt, J., Clore, G. L., & Jordan, A. H. (2008). Disgust as embodied moral judgment. Personality and Social Psychology Bulletin, 34, 1096-1109. Schulte, L. J., & Battle, J. (2004). The relative importance of ethnicity and religion in predicting attitudes towards gays and lesbians. Journal of Homosexuality, 47(2), 127-142. Sheldon, J. P., Pfeffer, C. A., Jayaratne, T. E., Feldbaum, M., & Petty, E. M. (2007). Beliefs about the etiology of homosexuality and about the ramifications of discovering its possible genetic origin. Journal of Homosexuality, 52, 11-150. Sherkat, D. E. (2002). Sexuality and religious commitment in the United States: An empirical examination. Journal for the Scientific Study of Religion, 41(2), 313-323. Skinner, W., & Fream, A. (1997). A social learning theory analysis of computer crime among college students. Journal of Research in Crime and Delinquency, 34, 495-518. Smith, C. T., & Nosek, B. A. (2011). Affective focus increases the concordance between implicit and explicit attitudes. Social Psychology, 42, 300-313. Spilka, B., Hood, Jr. R. W., Hunsberger, B., Gorsuch, R. (2003). The Psychology of Religion: An Empirical Approach (3rd ed.). New York: The Guilford Press. Strong, B., DeVault, C., Sayad, B. W., & Yarber, W. L. (2002). Human Sexuality. Boston: McGraw-Hill Co. Inc. University of Ghana http://ugspace.ug.edu.gh 245 Søgaard, M. (2013). Consequences of imposing the homo/hetero binary and the prospect for decriminalisation of MSM in contemporary Ghana (Unpublished master‘s thesis). University of Copenhagen, Denmark. Terrizzi Jr., J. A., Shook, N. J., Ventis, W. L. (2010). Disgust: A predictor of social conservatism and prejudicial attitudes toward homosexuals. Personality and Individual Differences, 49(6), 587-592. Thompson, E. M. (2008). Girl Friend or Girlfriend? Journal of Bisexuality, 6(3), 47-67. Urban Dictionary. Retrieved September 3, 2014 from http://www.urbandictionary.com/define.php?term=humansexual Tylor, E. B. (1958). Primitive Culture. New York: Harper Torchbooks. Whitley Jr., B. E. (2009). Religiosity and attitudes toward lesbians and gay men: A meta- analysis. International Journal for the Psychology of Religion, 19(1), 21-38. Worthington, E. L., Jr. (1988). Understanding the values of religious clients: A model and its application to counseling. Journal of Counseling Psychology, 35, 166–174. Worthington, Jr., E. L., Wade, N. G., Hight, T. L., Ripley, J. S., McCullough, M. E., Berry, J. W., …O‘Connor, L. (2003). The religious commitment inventory—10: Development, refinement, and validation of a brief scale for research and counseling. Journal of Counseling Psychology, 50(1), 84 – 96. Weiner, B. (1985). An attributional theory of achievement motivation and emotion. Psychological review, 92(4), 548. Weiner, B., Perry, R. P., & Magnusson, J. (1988). An attributional analysis of reactions to stigma. Journal of Personality and Social Psychology, 55,738-748. University of Ghana http://ugspace.ug.edu.gh 246 WIN-Gallup International (2012). Global index of religiosity and atheism. Retrieved August 8, 2013 from http://www.wingia.com/web/files/news/14/file/14.pdf Zanna, M. P., & Rempel, J. K. (1988). Attitudes: A new look at an old concept. In D. Bar-Tal & A. W. Kruglanski (Eds.), The social psychology of knowledge (pp. 315-334). Cambridge, England: Cambridge University Press. University of Ghana http://ugspace.ug.edu.gh 247 Appendices Appendix A: Group Evaluation Questionnaire (GEQ) SECTION A: DEMOGRAPHIC INFORMATION Instructions: Circle the option that applies to you or write out your response, depending on what a particular question demands. 1. Gender 1. Female 2. Male 2. Age: ___________years 3. Level 1. 100 2. 200 4. Religion 1. None 2. Christianity (indicate the denomination) __________________ 3. Muslim/Moslem 4. Traditional 5. Other (please specify) ________________________ SECTION B: RELIGIOUS COMMITMENT AND MORALITY 1. Please read each of the following statements relating to religious activities carefully. Using the scale below, CIRCLE the response that best describes how true each statement is for you. You can choose ONLY ONE of the options for each statement. Not at all true of me - 1 Somewhat true of me - 2 Moderately true of me - 3 Mostly true of me - 4 Totally true of me - 5 University of Ghana http://ugspace.ug.edu.gh 248 a. I often read books and magazines about my faith. b. I make financial contributions to my religious organization. c. I spend time trying to grow in understanding of my faith. d. Religion is especially important to me because it answers many questions about the meaning of life. e. My religious beliefs lie behind my whole approach to life. f. I enjoy spending time with others of my religious affiliation. g. Religious beliefs influence all my dealings in life. h. It is important to me to spend periods of time in private religious thought and reflection. i. I enjoy working in the activities of my religious affiliation. j. I keep well informed about my local religious group and have some influence in its decisions. 2. Below are twenty-one (21) acts or situations. Please note that you will need to go over them twice. The first time, please evaluate each of them in terms of “rightness” or “wrongness” ranging from one (1) to ten (10). Write ‗one‘ in BOX „A‟ for each item if the item seems least ―wrong‖ or ―not wrong at all‖, and ‗ten‘ if the item is judged ―most wrong‖. Use the in- between numbers (2 - 9) for in-between degrees of wrongness; the higher the number, the more wrong it becomes. The scale is therefore as follows: 1 2 3 4 5 6 7 8 9 10 LEAST MOST WRONG WRONG When you are done with the first round, kindly go over the items again for the second time. This time, please rate your own likelihood of engaging in these acts or situations using the following scale: Not at all likely - 1 Seldom likely - 2 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 3 3 3 3 3 3 3 3 3 3 4 4 4 4 4 4 4 4 4 4 5 5 5 5 5 5 5 5 5 5 University of Ghana http://ugspace.ug.edu.gh 249 Sometimes likely - 3 Most Often likely - 4 Highly likely - 5 Write the number corresponding to your choice in BOX „B‟. A B 1. Kidnapping and killing a child for juju. 2. Having sexual relations while unmarried. 3. Forging a cheque. 4. Smoking cigarettes. 5. Testifying falsely in court when under oath. 6. Having illicit sexual relations after marriage. 7. Attacking and robbing a person. 8. Having an abortion. 9. Not giving to charity when able. 10. Keeping over-change mistakenly given you by a shop attendant. 11. Copying from another‘s paper in a school examination. 12. Speeding away after one‘s car knocks down a pedestrian. 13. Getting a divorce. 14. Accepting a bribe in the course of duty. 15. Engaging in homosexuality. 16. Using profane or blasphemous speech. 17. Taking money for one‘s vote in an election. University of Ghana http://ugspace.ug.edu.gh 250 18. Avoiding a fare on public transport. 19. Engaging in commercial sex work. 20. Taking marijuana or any hard drug. 21. Killing someone in self-defense. SECTION C: EVALUATION THERMOMETER Instructions: Please provide a number between zero degrees and 100 degrees to indicate your overall evaluation of typical members of the following groups of people using the thermometer below: POSITIVE 100o Extremely Favourable 90o Very Favourable 80o Quite Favourable 70o Fairly Favourable 60o Slightly Favourable 50o Neither Favourable nor Unfavourable 40o Slightly Unfavourable 30o Fairly Unfavourable 20o Quite Unfavourable 10o Very Unfavourable NEGATIVE 0o Extremely Unfavourable a. Orthodox Church Priests _________________ b. Commercial Sex Workers _________________ c. Spiritual Church Leaders _________________ University of Ghana http://ugspace.ug.edu.gh 251 d. Homosexuals _________________ e. Traditional Priests _________________ SECTION D: COGNITIVE AND AFFECTIVE INFORMATION 1. For each of the groups stated below, first list the characteristics you would use to describe typical members of the group. You have the option to list as many characteristics as you can. Then, using the following scale, rate each characteristic listed based on the extent to which you think it is a positive or negative characteristic: -2 -1 0 +1 +2 VERY VERY NEGATIVE POSITIVE Finally, state the percentage of group members you believe possess each characteristic. a. Orthodox Church Priests CHARACTERISTIC RATING PERCENTAGE __________________________ ______ _______ __________________________ ______ _______ __________________________ ______ _______ b. Commercial Sex Workers CHARACTERISTIC RATING PERCENTAGE __________________________ ______ _______ __________________________ ______ _______ __________________________ ______ _______ c. Spiritual Church Leaders CHARACTERISTIC RATING PERCENTAGE __________________________ ______ _______ __________________________ ______ _______ __________________________ ______ _______ University of Ghana http://ugspace.ug.edu.gh 252 d. Homosexuals CHARACTERISTIC RATING PERCENTAGE __________________________ ______ _______ __________________________ ______ _______ __________________________ ______ _______ e. Traditional Priests CHARACTERISTIC RATING PERCENTAGE __________________________ ______ _______ __________________________ ______ _______ __________________________ ______ _______ 2. For each of the groups stated below, first list the values, customs, or traditions that you believe are blocked or facilitated by typical members of the group. You have the option to list as many as you can. Then, using the following scale, rate them based on the extent to which you believe each value, custom, or tradition is blocked or facilitated: -2 -1 0 +1 +2 ALMOST ALMOST ALWAYS BLOCKED ALWAYS FACILITATED Finally, state the percentage of group members you believe block or facilitate each value. a. Orthodox Church Priests VALUE RATING PERCENTAGE __________________________ ______ _______ __________________________ ______ _______ __________________________ ______ _______ b. Commercial Sex Workers VALUE RATING PERCENTAGE __________________________ ______ _______ University of Ghana http://ugspace.ug.edu.gh 253 __________________________ ______ _______ __________________________ ______ _______ c. Spiritual Church Leaders VALUE RATING PERCENTAGE __________________________ ______ _______ __________________________ ______ _______ __________________________ ______ _______ d. Homosexuals VALUE RATING PERCENTAGE __________________________ ______ _______ __________________________ ______ _______ __________________________ ______ _______ e. Traditional Priests VALUE RATING PERCENTAGE __________________________ ______ _______ __________________________ ______ _______ __________________________ ______ _______ 3. For each of the groups stated below, first list the feelings or emotions you experience when you see, meet, or think about typical members of the group. Then, using the following scale, rate each feeling or emotion listed based on the extent to which you think it is a positive or negative emotion: -2 -1 0 +1 +2 VERY VERY NEGATIVE POSITIVE Finally, state the percentage of group members who evoke each emotion in you. a. Orthodox Church Priests University of Ghana http://ugspace.ug.edu.gh 254 FEELING/EMOTION RATING PERCENTAGE __________________________ ______ _______ __________________________ ______ _______ __________________________ ______ _______ b. Commercial Sex Workers FEELING/EMOTION RATING PERCENTAGE __________________________ ______ _______ __________________________ ______ _______ __________________________ ______ _______ c. Spiritual Church Leaders FEELING/EMOTION RATING PERCENTAGE __________________________ ______ _______ __________________________ ______ _______ __________________________ ______ _______ d. Homosexuals FEELING/EMOTION RATING PERCENTAGE __________________________ ______ _______ __________________________ ______ _______ __________________________ ______ _______ e. Traditional Priests FEELING/EMOTION RATING PERCENTAGE __________________________ ______ _______ __________________________ ______ _______ __________________________ ______ _______ THANK YOU! University of Ghana http://ugspace.ug.edu.gh 255 Appendix B: Group Evaluation and Behavioural Intention Form (GEBIF) SECTION A: EVALUATION THERMOMETER Instructions: Please provide a number between zero degrees and 100 degrees to indicate your overall evaluation of typical members of the following groups of people using the thermometer below: POSITIVE 100o Extremely Favourable 90o Very Favourable 80o Quite Favourable 70o Fairly Favourable 60o Slightly Favourable 50o Neither Favourable nor Unfavourable 40o Slightly Unfavourable 30o Fairly Unfavourable 20o Quite Unfavourable 10o Very Unfavourable NEGATIVE 0o Extremely Unfavourable a. Orthodox Church Priests _________________ b. Commercial Sex Workers _________________ c. Spiritual Church Leaders _________________ d. Homosexuals _________________ e. Traditional Priests _________________ University of Ghana http://ugspace.ug.edu.gh 256 SECTION B: BEHAVIOURAL INTENTIONS Instructions: Imagine that you know the person you just read about in the vignette personally. Using the space provided on the next page, please write out how you are likely to behave towards the person and why: ______________________________________________________________________________ ______________________________________________________________________________ ______________________________________________________________________________ ______________________________________________________________________________ ______________________________________________________________________________ ______________________________________________________________________________ ______________________________________________________________________________ ______________________________________________________________________________ ______________________________________________________________________________ ______________________________________________________________________________ ______________________________________________________________________________ ______________________________________________________________________________ ______________________________________________________________________________ ______________________________________________________________________________ ______________________________________________________________________________ ______________________________________________________________________________ ______________________________________________________________________________ ______________________________________________________________________________ ______________________________________________________________________________ THANK YOU!!! University of Ghana http://ugspace.ug.edu.gh 257 Appendix C: Focus Group Discussion Guide A: OPENING QUESTIONS 1. Let‘s talk about religion. - Would you say Ghanaians are religious? - Which denominations/religious organizations are you aware of? - On the scale of 0 – 100%, how many followers do these religious organizations have among Ghanaians? - What about here on campus? Would you say students are religious? - Which denominations/religious organizations are present on campus? - On the scale of 0 – 100%, how many student followers do these religious organizations have? - How committed are people in Ghana to their religion? What about the students on campus? - What are some of the things that indicate how committed someone is to his/her religion? - What about you personally? What is your religious affiliation? (Ask respondents to each answer this and the next three questions in turn). - Would you describe yourself as a religious person? Why? Why not? - What are some of the things that you do as part of your religion? - Based on what you do, would you say you are very committed to your religion? 2. Let‘s take a look also at morality. - How would you define morality? Would you say it is synonymous to being religious or having a high religious commitment? - To what extent do you think people are likely to engage in the following:  Bribery and Corruption  Premarital and Extramarital Affairs  Avoiding fare on public transport, keeping overchange, cheating during exams  Smoking cigarettes, using hard drugs,  Commercial Sex Work, Homosexuality - On a scale of 0 – 100%, how many people do you think are likely to engage in this? (Ask for each of the items above, right after respondents have answered the previous question relating to the item). - How open are people about their engagement in this? (Ask for each of the items above, right after respondents have answered the previous question relating to the item). - What about you? To what extent are you likely to engage in any of these? University of Ghana http://ugspace.ug.edu.gh 258 B: HOMOSEXUALITY 1. The issue of homosexuality is now common knowledge in Ghana. Do you think there are Ghanaians who are homosexuals? Do you know if it exists here on campus? (Ask each of the respondents also about where they live outside of campus). - Can you tell us what you know about homosexuality? Which of the sexes are more involved in it? - How do you think people/students get to become homosexuals? - Do you know anyone personally who is gay or lesbian? How did you know? - Are those practicing homosexuality as open about their relationship as heterosexuals? Why? Why not? - How do people/students here relate to those who are homosexuals? - What about you? Personally, how would you describe homosexuals? - What values, customs or traditions do you believe are blocked by homosexuals? - What values, customs or traditions do you believe are facilitated by homosexuals? - How do you feel when you see, meet or think about homosexuality? (Ask specifically if any of them have had contact with a homosexual). C: VIGNETTES Give out one of the 5 vignettes to each of the participants to read and then ask them to discuss. The tone of the discussions depends on which set of groups is being moderated. For the first set of groups, the following questions should be asked: 1. Vignette One: “LIFE IS GOOD” - What do you think about the whole scenario presented by this vignette? (Probe to get at how respondents think and feel about the vignette person). 2. Vignette Two: “MY LIFE IS MINE TO LIVE” - What do you think about the whole scenario presented by this vignette? (Probe to get at how respondents think and feel about the vignette person). 3. Vignette Three: “A NEW LEASE ON LIFE” - Do you think the vignette person and her parents had any other choice than what is presented in the story? - What do you think about the vignette person‘s worry about her feelings that she‘s a man in a woman‘s body and the fact that she‘s in a relationship with a man now? - What do you think also about the decision not to tell her fiancé about her past? - What do you think about the whole scenario presented by this vignette? (Probe to get at how respondents think and feel about the vignette person). 4. Vignette Four: “LIFE IS COMFORTABLE” - What do you think about Joseph‘s choices and his philosophy about life? - What about his parents? What do you think about their decision to disown him and the fact that they have forbidden his siblings to see him? University of Ghana http://ugspace.ug.edu.gh 259 - What about the siblings? Do you think they should obey the dictates of their parents? - What do you think about the whole scenario presented by this vignette? (Probe to get at how respondents think and feel about the vignette person). For the second set of groups, ask the respondents the following question for all 5 vignettes and let them take turns answering: - If you were the vignette person, what would you do given these same circumstances and why? THANK YOU!!! University of Ghana http://ugspace.ug.edu.gh 260 Appendix D: In-Depth Interview Guide SEXUAL ORIENTATION AND RELATED EXPERIENCES 1. Personally, what would you say is your sexual orientation? Why do you consider yourself of this orientation? - Usually, society socializes us from birth to be heterosexual. What would you say are the factors that precipitated your sexual orientation? - Would you say your orientation is by choice or is it biological? Are you comfortable with this orientation? - Is this usually the case with other people that you know who share your orientation? - How do people who do not share your orientation react to those who do? What are the reasons for this? How do you feel about it? - Are you open about your orientation? - If yes, how do you handle the reactions of people considering that the society usually would like to pretend that other sexual orientations do not exist? - If no, why are you not open about it? - What about your family members? Do they know about your sexual orientation? - If yes, how did they react the first time they found out? How did they find out? How do they relate to you now? - If no, why haven‘t you told them? Do you think that your relationship with them will be negatively affected if they were to find out? 2. You definitely aren‘t the only homosexual person here on campus. What percentage of the student population do you think are homosexual? - Are you certain of this number? - What percentage of this number are gays and what percentage are lesbians? - Do you have contact with other homosexuals on campus? What form does this contact take? - How are other homosexual persons on campus treated? Are they open about their orientation? Why? Why not? THANK YOU!!! University of Ghana http://ugspace.ug.edu.gh 261 Appendix E: Vignettes VIGNETTE 1 - Life is Good!!! I am in my twenties and work with one of the big companies here in Ghana. I love my job and try to work to the best of my ability. My position ensures that I meet a lot of people which helps my company but also poses a lot of temptations for me because I‘m often faced with people wanting to bribe their way into getting services from my company when they really do not deserve it. Since I was brought up in a religious home and have a love for my country and my company, I‘ve always been able to overcome this temptation. I try to be law-abiding always and I dare say that I‘ve succeeded very well enough in my short life. I can say that I am very satisfied with my life as it is and even more so because I have a very loving and supportive same-sex partner. Initially, I was concerned about my attraction to others of my own sex but now I understand that this is just how I am. Like almost all children, I was born after 9 months with all my fingers and toes accounted for. My parents were elated. That is until I started showing signs of being a homosexual during my teenage years. But all that is behind us now. My parents, siblings and I have all come to understand that I am the way I am because of my biological makeup which is something I have no control over. According to our family doctor (whom my parents took me to see because they were worried about me), adult sexual orientation depends on the level of androgens to which the brain is exposed during a sensitive period of development of the foetus in the womb. The doctor explained that this is known as the prenatal hormonal theory which has a solid basis in animal research. He told us that according to LeVay & Valente (2003), research in this area has shown that most males and few females exceed some threshold of androgen exposure and therefore become attracted to females. Most females and a few males fall below that threshold level and therefore become attracted to males. This means that females who exceed the threshold of androgen exposure and males who fall below that threshold while in their mothers‘ wombs are born homosexual and are therefore sexually attracted to members of their own sex. Having read more on the issue since then, my family and I understand that I am who I am and nothing can be done about it. I‘ve made my peace with it and so have they. They are very University of Ghana http://ugspace.ug.edu.gh 262 supportive of me and are happy I have a loving and supportive partner. I am very happy myself. Life is good!!! VIGNETTE 2 - My Life is Mine to Live I am so tired of people trying to determine how I‘m to live my life!!! I‘m well over 18 and definitely well over the age of consent. In fact, I‘m 28 years old and therefore a full grown adult. If I want to be homosexual, then that is just what I‘ll be. Like most young people, I started having romantic relationships when I was a teenager and I‘ve been in several relationships with people of the opposite sex since then. What good did it do me? All I‘ve gained is disappointment and heartaches over and over again until my current partner and best friend since childhood offered me a shoulder to cry on during my last ‗broken heart‘. As they say, one thing lead to another and here we are today. I never knew love could be so sweet and the sex is great!!! Much much better than I‘ve ever had with someone of the opposite sex. I don‘t care what people think. This is who I am and who I want to be. I‘m homosexual and I‘m proud to be so. It is my human right to live my life anyway I want to live it and no one can deny me that right, not even my parents. They are not happy with me and my father is always threatening to disown me. What do I care? They can‘t live my life for me. The fact that they brought me into this world doesn‘t mean they have the right to control my life. After all, I didn‘t ask to be born, did I? My philosophy is ―live and let live‖. That is how I‘ve approached all aspects of my life: school, relationships, work. There are too many rules anyway and it‘s all couched in religion. Don‘t do this, don‘t do that!! That‘s all you hear from people in the society but most of them are doing the supposedly bad things when they think no one is watching. I‘ve done my own share of those supposedly bad things in my school days – cheating (in exams, with bus fare e.t.c), drinking alcohol, smoking (even a little wee) – and I probably will go on doing them so long as it makes me happy. Like I said, it‘s my life. I‘m working now and like always I look out for myself. Most companies in Ghana today do not pay as much as workers deserve so I try to make ends meet by whatever means possible. There‘s nothing free under the sun so customers have to pay a little extra if they want quick and expert service. You know what I mean. Everybody understands that this is how things work. I‘m University of Ghana http://ugspace.ug.edu.gh 263 enjoying life because it‘s meant to be enjoyed. No more heartache for me and nothing can get in the way of my happiness. My life is mine to live!!! VIGNETTE 3 - A New Lease on Life I have two reasons to be happy today. The first is that 3 years ago, I finally came back to Ghana from the U.S.A after the sex-change surgery that I so badly needed. The second is that a year ago, I met my wonderful fiancé sitting across the dinner table from me right now. We met in this same restaurant that day and intend to keep celebrating our anniversary here because on this same date next year, we will get married. Wondering why I needed a sex-change surgery? Let me tell you. I was born with a condition known as gonadal intersexuality. It is a rare kind of intersexuality in which a single individual possesses both testicular and ovarian tissue although the external genitals often look like one of either sexes. So, you might say that I was born a hermaphrodite. My parents didn‘t know this and so I was raised as a boy because I had genitals that made me male. The problems started when I reached adolescence as I began developing breasts and rounder hips while my genitals remained like that of a boy instead of a teenager on the verge of manhood. I was terrified and my parents were devastated because I am their only child. We had to see a doctor and upon medical examination, the problem was realized. Although I had male genitals, my ovarian tissues were more developed than the testicular tissues and according to the doctor, considering that I had also developed breasts and a more feminine figure, it would be best if I had a sex- change and became fully female. So, obviously, I needed surgery but how were we going to deal with a society that knew me as a boy from childhood, now having to see me as a girl. After considering our options, my parents decided we should leave the country immediately for the United States of America, where I was to get the surgery done anyway. As soon as we moved there, arrangements were made for the surgery and thanks to God it was successful. I‘m now fully female. I went through school eventually without too much trouble and finished my university education. When we came back home though, we had to pretend I was a few years younger than I actually am because while in the U.S. my parents called to tell our relatives and old friends that I had died but called again in about a year and a half that they had adopted a girl University of Ghana http://ugspace.ug.edu.gh 264 who had also lost her parents but who somehow looked very much like me that they couldn‘t just let her go. Everyone was sympathetic to this and thought my parents were very lucky to get someone like that to help them overcome loosing me. This helped to make my transition back into Ghana very smooth. Nathaniel doesn‘t know about all these and I don‘t intend to tell him as it is in the past and I have a new life now. It‘s just that sometimes I can‘t shake the feeling that I‘m a man in a woman‘s body. This makes me wonder about my feelings for Nathaniel and whether it can be considered homosexual. I try not to dwell on these negatives too much though and I must say I‘m a very happy young woman because I have a new lease on life. I intend to enjoy it!!! VIGNETTE 4 - Life is Comfortable Everybody deserves to be comfortable in life. That is what I believe. So, the way I see it, if something or someone makes you feel uncomfortable, then you need to deal with it so that you can be as comfortable as you should be. Right now, I am very comfortable with my life but especially in my body, after being uncomfortable almost two-thirds of my life. My name is Joseph Owusu and I‘m 32 years old but I‘ve been known by this name only since I was 21 years old. Before then, my family and friends knew me as Josephine Owusu. Are you asking how that came about? The answer is simple. I had sex-change surgery and before you ask, let me tell you why. For all of my 21 years of being the girl Josephine Owusu, I was never comfortable with my female body. For some reason, I felt that I should have been a boy instead of a girl. I hated the feminine clothes and all the things that a female is supposed to do. I hated even more that I had to have periods every month. What‘s more, I realized that I was attracted to girls instead of boys. This confirmed for me that I indeed wasn‘t in the right body because I felt totally male. As soon as I turned 18, I gathered courage and told my parents about this but as one would expect, they took my complaints as silly talk. It was obvious that I wasn‘t going to get any help there so, I decided to take matters into my own hands. After all, it‘s my body. I already knew that one could have a sex-change so I set out to find out where I could get it done, how much it will cost and all the necessary information and planned towards getting it done. It meant money University of Ghana http://ugspace.ug.edu.gh 265 and the only way I could see that could get me money fast was for me to get a rich boyfriend. I must say I was a very beautiful young woman so it was too difficult to hook a very rich sugar daddy who agreed to send me to the U.S.A. Of course he didn‘t know what I intended to do there and I didn‘t tell my parents or siblings that I was leaving the country. As soon as I got to there, I cut all ties with him and found a hospital that could do the surgery for me without delay. It was very successful, hence my new handsome self now. I‘m back in Ghana now and no one recognizes me, not even my former sugar daddy, so I‘m safe. My parents and siblings know all about my sex-change because I called and told them what I intended to do after I got to the States. My parents had a fit and they‘ve decided that I no longer exist which is fine with me. My siblings have been forbidden to see me or have anything to do with me, which is fine too. I‘ve made new friends and I have a good job. I don‘t need anybody from my past anyway. My friends and colleagues believe I‘m an orphan so that takes care of having to explain why I don‘t have a good relationship with my family. I have a fiancée now and we hope to get married within a year but she also doesn‘t know. I intend to keep it that way. I finally feel that I‘m in the right body so life is comfortable and that is as it should be. VIGNETTE 5 - Just Five Minutes of Pleasure It seems like ages now since we felt we were so in love and ready to take on the world. Now, sitting here staring angrily at each other, we both wonder how we could have been so foolish as to get ourselves into this mess. It all started innocently enough. We share common friends so it was at a friend‘s birthday party that we met for the first time and took it for granted that we‘ve found a new friend in each other because as it‘s often said, ―The friend of my friend is my friend‖. I thought she was the most beautiful girl I had ever seen. Later, I found out that she also thought me the most handsome guy she had ever seen. Of course, I didn‘t propose right away and when I finally gathered the courage to do so, she also didn‘t agree to my proposal right away. The time she took to ―stretch‖ me, like they say, only added to the excitement of the whole ―chase‖. As soon as Nana Akua said yes though, everything started moving fast. We wanted to spend all our time together, alone with each other. Before long, we rarely joined our whole circle of University of Ghana http://ugspace.ug.edu.gh 266 friends. The time spent alone quickly moved from being time to just talk, hold hands, take walks and kiss goodbye at the end of the day to wanting to have total privacy so we can engage in heavy petting. Make no mistake, Nana set her limits and being a gentleman, I respected those limits but somehow, we just couldn‘t control ourselves anymore. One day we eventually went all the way. It‘s funny but that was the day we forgot to kiss goodbye. We also didn‘t see each other for two days afterwards, each of us finding an excuse not to meet. This time apart, as we learnt later on from each other, was spent alternating between feeling guilty and justifying our actions. After all, we were in love with each other so it was only natural that we would consummate that love. We didn‘t even get the chance to do it again and now this!! It‘s been two weeks since Nana told me she has missed her period and today she took the home pregnancy test using a kit we got from the pharmacy. The stick is showing blue like it‘s supposed to if one is pregnant. We‘ve just had our first major argument and blamed each other for causing this but the big question is, ―What are we going to do‖? We are both 18 and have just completed SHS. My name is Paa Kwasi Addo and I can‘t believe how much trouble we are in right now for just about 5 minutes of pleasure!!! University of Ghana http://ugspace.ug.edu.gh 267 Appendix F: Informed Consent Forms FORM 1 Title: ATTITUDES TOWARDS HOMOSEXUALS: ASSESSING THE STRUCTURE OF PREJUDICIAL ATTITUDES AND THE MODERATING EFFECTS OF RELIGIOUS COMMITMENT AND MORALITY. Principal Investigator: ANGELA ANARFI GYASI-GYAMERAH Address: DEPARTMENT OF PSYCHOLOGY, UNIVERSITY OF GHANA, LEGON Introduction This Consent Form contains information about the research named above. In order to be sure that you are informed about being in this research, we are asking you to read (or have read to you) this Consent Form. You will also be asked to sign it (or make your mark in front of a witness). We will give you a copy of this form. This consent form might contain some words that are unfamiliar to you. Please ask us to explain anything you may not understand. Reason for the Research You are being asked to take part in a research to investigate people‘s attitudes towards homosexuals. General Information about Research This research will mainly involve the completion of questionnaires which is in three sections and deals with issues relating to people‘s religiousity, morality, and general attitudes towards typical members of certain social groups including homosexuals. Your Part in the Research If you agree to be in the research, you will have to complete one questionnaire. Your part in the research will last for 1 hour. About 200 persons (100 men and 100 women) in all will take part in this research at the Department of Psychology, University of Ghana, Legon. Possible Risks No risks, whether physical or psychological, are anticipated in this research as the questionnaire is meant to simply seek your opinion on the issue of homosexuality. Possible Benefits Your contribution will be very valuable in that it will help towards getting a greater understanding of attitudes towards homosexuals in Ghana. It will also contribute to the body of knowledge in this area of research. University of Ghana http://ugspace.ug.edu.gh 268 If You Decide Not to Be in the Research You are free to decide if you want to be in this research or not. It is strictly voluntary. Confidentiality The data from this study is for academic purposes only and will be kept strictly and completely confidential. Your personal information will not be associated with the data nor with any written reports, presentation, or publications that may develop from this study. Any future use of the data will be for the same purposes and will be subjected to the same confidentiality guidelines. However, the staff of the Department of Psychology, University of Ghana, Legon, may sometimes look at your research records. Someone from the IRB might want to ask you questions about being in the research, but you do not have to answer them. Compensation You will not be paid for participating in this research since it is strictly voluntary. Leaving the Research You may leave the research at any time. If you choose to take part, you can change your mind at any time and withdraw. Also, you may be asked to leave the research if:  you are not able to follow the research procedures, or  the research is stopped. If You Have a Problem or Have Other Questions In case you encounter any problem during this research or have any questions about the research, please call Mrs. Angela A. Gyasi-Gyamerah, Principal Investigator (Tel.: 0264719165) or Dr. Charity Akotia, Head, Department of Psychology, University of Ghana, Legon (Tel.: 028- 9550463). Your rights as a participant This research has been reviewed and approved by the IRB of Noguchi Memorial Institute of Medical Research of the University of Ghana. An IRB is a committee that reviews research studies in order to help protect participants. If you have any questions about your rights as a research participant you may contact Rev. Dr. Ayete-Nyampong, Chairperson, NMIMR-IRB, on mobile number 0208152360. University of Ghana http://ugspace.ug.edu.gh 269 VOLUNTEER AGREEMENT The above document describing the benefits, risks and procedures for the research title ―Attitudes towards Homosexuals: Assessing the Structure of Prejudicial Attitudes and the Moderating Effects of Religious Commitment and Morality.‖ has been read and explained to me. I have been given an opportunity to have any questions about the research answered to my satisfaction. I agree to participate as a volunteer. ___________________________ __________________________________ Date Signature or mark of volunteer If volunteers cannot read the form themselves, a witness must sign here: I was present while the benefits, risks and procedures were read to the volunteer. All questions were answered and the volunteer has agreed to take part in the research. _______________________ __________________________________ Date Signature of Witness I certify that the nature and purpose, the potential benefits, and possible risks associated with participating in this research have been explained to the above individual. _______________________ __________________________________ Date Signature of Person Who Obtained Consent University of Ghana http://ugspace.ug.edu.gh 270 FORM 2 Title: ATTITUDES TOWARDS HOMOSEXUALS: ASSESSING THE STRUCTURE OF PREJUDICIAL ATTITUDES AND THE MODERATING EFFECTS OF RELIGIOUS COMMITMENT AND MORALITY. Principal Investigator: ANGELA ANARFI GYASI-GYAMERAH Address: DEPARTMENT OF PSYCHOLOGY, UNIVERSITY OF GHANA, LEGON Introduction This Consent Form contains information about the research named above. In order to be sure that you are informed about being in this research, we are asking you to read (or have read to you) this Consent Form. You will also be asked to sign it (or make your mark in front of a witness). We will give you a copy of this form. This consent form might contain some words that are unfamiliar to you. Please ask us to explain anything you may not understand. Reason for the Research You are being asked to take part in a research to investigate people‘s attitudes towards homosexuals. General Information about Research This research will involve the completion of two questionnaires. The first one is in three sections and deals with issues relating to people‘s religiousity, morality, and general attitudes towards typical members of certain social groups including homosexuals. The second questionnaire is in two sections and deals with attitudes towards typical members of certain social groups and behavioural intentions towards certain individuals. Your Part in the Research If you agree to be in the research, you will be one of a group of 20 persons who will take part in an experiment. The first part of the experiment will involve the completion of the first questionnaire. This should take about 45 minutes. After you are finished completing this questionnaire, you will be given a story about a particular individual to read (or have read to you) within 10 minutes. As soon as you finish reading the story, you will fill out the second questionnaire which should take about 35 minutes. Your part in the research will therefore last for about 1 hour and 30 minutes. In all, 200 persons (100 men and 100 women) will take part in this research at the Department of Psychology, University of Ghana, Legon. University of Ghana http://ugspace.ug.edu.gh 271 Possible Risks No risks, whether physical or psychological, are anticipated in this research as the questionnaire is meant to simply seek your opinion on the issue of homosexuality. Possible Benefits Your contribution will be very valuable in that it will help towards getting a greater understanding of attitudes towards homosexuals in Ghana. It will also contribute to the body of knowledge in this area of research. If You Decide Not to Be in the Research You are free to decide if you want to be in this research or not. It is strictly voluntary. Confidentiality The data from this study is for academic purposes only and will be kept strictly and completely confidential. Your personal information will not be associated with the data nor with any written reports, presentation, or publications that may develop from this study. Any future use of the data will be for the same purposes and will be subjected to the same confidentiality guidelines. However, the staff of the Department of Psychology, University of Ghana, Legon, may sometimes look at your research records. Someone from the IRB might want to ask you questions about being in the research, but you do not have to answer them. Compensation You will not be paid for participating in this research since it is strictly voluntary. Leaving the Research You may leave the research at any time. If you choose to take part, you can change your mind at any time and withdraw. Also, you may be asked to leave the research if:  you are not able to follow the research procedures, or  the research is stopped. If You Have a Problem or Have Other Questions In case you encounter any problem during this research or have any questions about the research, please call Mrs. Angela A. Gyasi-Gyamerah, Principal Investigator (Tel.: 0264719165) or Dr. Charity Akotia, Head, Department of Psychology, University of Ghana, Legon (Tel.: 028- 9550463). Your rights as a participant University of Ghana http://ugspace.ug.edu.gh 272 This research has been reviewed and approved by the IRB of Noguchi Memorial Institute of Medical Research of the University of Ghana. An IRB is a committee that reviews research studies in order to help protect participants. If you have any questions about your rights as a research participant you may contact Rev. Dr. Ayete-Nyampong, Chairperson, NMIMR-IRB, on mobile number 0208152360. University of Ghana http://ugspace.ug.edu.gh 273 VOLUNTEER AGREEMENT The above document describing the benefits, risks and procedures for the research title ―Attitudes towards Homosexuals: Assessing the Structure of Prejudicial Attitudes and the Moderating Effects of Religious Commitment and Morality.‖ has been read and explained to me. I have been given an opportunity to have any questions about the research answered to my satisfaction. I agree to participate as a volunteer. ___________________________ ___________________________________ Date Signature or mark of volunteer If volunteers cannot read the form themselves, a witness must sign here: I was present while the benefits, risks and procedures were read to the volunteer. All questions were answered and the volunteer has agreed to take part in the research. _______________________ __________________________________ Date Signature of Witness I certify that the nature and purpose, the potential benefits, and possible risks associated with participating in this research have been explained to the above individual. _______________________ __________________________________ Date Signature of Person Who Obtained Consent University of Ghana http://ugspace.ug.edu.gh 274 FORM 3A Title: ATTITUDES TOWARDS HOMOSEXUALS: ASSESSING THE STRUCTURE OF PREJUDICIAL ATTITUDES AND THE MODERATING EFFECTS OF RELIGIOUS COMMITMENT AND MORALITY. Principal Investigator: ANGELA ANARFI GYASI-GYAMERAH Address: DEPARTMENT OF PSYCHOLOGY, UNIVERSITY OF GHANA, LEGON Introduction This Consent Form contains information about the research named above. In order to be sure that you are informed about being in this research, we are asking you to read (or have read to you) this Consent Form. You will also be asked to sign it (or make your mark in front of a witness). We will give you a copy of this form. This consent form might contain some words that are unfamiliar to you. Please ask us to explain anything you may not understand. Reason for the Research You are being asked to take part in a research to investigate people‘s attitudes towards homosexuals. General Information about Research This research will mainly involve interviewing 180 persons (90 men and 90 women) in groups of 6 on issues relating to people‘s religiousity, morality, sexuality and general attitudes towards homosexuals. The groups will either be made up of persons who are all men, all women, or a mix of both men and women (3 of each). The interviews will be tape-recorded. Your Part in the Research If you agree to be in the research, you will be expected to take part in the discussion of the issues stated above. It is expected that the discussion should last for 1 hour. Possible Risks No risks, whether physical or psychological, are anticipated in this research as the questionnaire is meant to simply seek your opinion on the issue of homosexuality. Possible Benefits Your contribution will be very valuable in that it will help towards getting a greater understanding of attitudes towards homosexuals in Ghana. It will also contribute to the body of knowledge in this area of research. University of Ghana http://ugspace.ug.edu.gh 275 If You Decide Not to Be in the Research You are free to decide if you want to be in this research or not. It is strictly voluntary. Confidentiality The data from this study is for academic purposes only and will be kept strictly and completely confidential. The recorded discussions will be transcribed and the cassette erased. Your personal information will not be associated with the data nor with any written reports, presentation, or publications that may develop from this study. Any future use of the data will be for the same purposes and will be subjected to the same confidentiality guidelines. However, the staff of the Department of Psychology, University of Ghana, Legon, may sometimes look at your research records. Someone from the IRB might want to ask you questions about being in the research, but you do not have to answer them. Compensation You will not be paid for participating in this research since it is strictly voluntary. Leaving the Research You may leave the research at any time. If you choose to take part, you can change your mind at any time and withdraw. Also, you may be asked to leave the research if:  you do not contribute to the discussion at all or  the research is stopped. If You Have a Problem or Have Other Questions In case you encounter any problem during this research or have any questions about the research, please call Mrs. Angela A. Gyasi-Gyamerah, Principal Investigator (Tel.: 0264719165) or Dr. Charity Akotia, Head, Department of Psychology, University of Ghana, Legon (Tel.: 028- 9550463). Your rights as a participant This research has been reviewed and approved by the IRB of Noguchi Memorial Institute of Medical Research of the University of Ghana. An IRB is a committee that reviews research studies in order to help protect participants. If you have any questions about your rights as a research participant you may contact [Rev. Dr. Ayete-Nyampong, Chairperson, NMIMR-IRB, mobile 0208152360] University of Ghana http://ugspace.ug.edu.gh 276 VOLUNTEER AGREEMENT The above document describing the benefits, risks and procedures for the research title ―Attitudes towards Homosexuals: Assessing the Structure of Prejudicial Attitudes and the Moderating Effects of Religious Commitment and Morality.‖ has been read and explained to me. I have been given an opportunity to have any questions about the research answered to my satisfaction. I agree to participate as a volunteer. ___________________________ ____________________________________ Date Signature or mark of volunteer If volunteers cannot read the form themselves, a witness must sign here: I was present while the benefits, risks and procedures were read to the volunteer. All questions were answered and the volunteer has agreed to take part in the research. _______________________ __________________________________ Date Signature of Witness I certify that the nature and purpose, the potential benefits, and possible risks associated with participating in this research have been explained to the above individual. _______________________ __________________________________ Date Signature of Person Who Obtained Consent University of Ghana http://ugspace.ug.edu.gh 277 FORM 3B Title: ATTITUDES TOWARDS HOMOSEXUALS: ASSESSING THE STRUCTURE OF PREJUDICIAL ATTITUDES AND THE MODERATING EFFECTS OF RELIGIOUS COMMITMENT AND MORALITY. Principal Investigator: ANGELA ANARFI GYASI-GYAMERAH Address: DEPARTMENT OF PSYCHOLOGY, UNIVERSITY OF GHANA, LEGON Introduction This Consent Form contains information about the research named above. In order to be sure that you are informed about being in this research, we are asking you to read (or have read to you) this Consent Form. You will also be asked to sign it (or make your mark in front of a witness). We will give you a copy of this form. This consent form might contain some words that are unfamiliar to you. Please ask us to explain anything you may not understand. Reason for the Research You are being asked to take part in a research to investigate people‘s attitudes towards homosexuals. General Information about Research This research will mainly involve interviewing 20 persons (10 men and 10 women) into some detail about people‘s religiousity, morality and sexuality, including their own sexual orientation and experiences relating to it. The interviews will be tape-recorded. Your Part in the Research If you agree to be in the research, you will be interviewed alone and the whole session will be tape-recorded. It is expected that the interview should last for 45 minutes. Possible Risks No risks, whether physical or psychological, are anticipated in this research as the questionnaire is meant to simply understand your experiences as a homosexual and to get your views on people‘s attitude towards you and the issue of homosexuality. Possible Benefits Your contribution will be very valuable in that it will help towards getting a greater understanding of attitudes towards homosexuals in Ghana. It will also contribute to the body of knowledge in this area of research. University of Ghana http://ugspace.ug.edu.gh 278 If You Decide Not to Be in the Research You are free to decide if you want to be in this research or not. It is strictly voluntary. Confidentiality The data from this study is for academic purposes only and will be kept strictly and completely confidential. The recorded interviews will be transcribed and the cassette erased. Your personal information will not be associated with the data nor with any written reports, presentation, or publications that may develop from this study. Any future use of the data will be for the same purposes and will be subjected to the same confidentiality guidelines. However, the staff of the Department of Psychology, University of Ghana, Legon, may sometimes look at your research records. Someone from the IRB might want to ask you questions about being in the research, but you do not have to answer them. Compensation You will not be paid for participating in this research since it is strictly voluntary. Leaving the Research You may leave the research at any time. If you choose to take part, you can change your mind at any time and withdraw. Also, you may be asked to leave the research if:  you do not contribute to the discussion at all or  the research is stopped. If You Have a Problem or Have Other Questions In case you encounter any problem during this research or have any questions about the research, please call Mrs. Angela A. Gyasi-Gyamerah, Principal Investigator (Tel.: 0264719165) or Dr. Charity Akotia, Head, Department of Psychology, University of Ghana, Legon (Tel.: 028- 9550463). Your rights as a participant This research has been reviewed and approved by the IRB of Noguchi Memorial Institute of Medical Research of the University of Ghana. An IRB is a committee that reviews research studies in order to help protect participants. If you have any questions about your rights as a research participant you may contact [Rev. Dr. Ayete-Nyampong, Chairperson, NMIMR-IRB, mobile 0208152360] University of Ghana http://ugspace.ug.edu.gh 279 VOLUNTEER AGREEMENT The above document describing the benefits, risks and procedures for the research title ―Attitudes towards Homosexuals: Assessing the Structure of Prejudicial Attitudes and the Moderating Effects of Religious Commitment and Morality.‖ has been read and explained to me. I have been given an opportunity to have any questions about the research answered to my satisfaction. I agree to participate as a volunteer. ___________________________ ___________________________________ Date Signature or mark of volunteer If volunteers cannot read the form themselves, a witness must sign here: I was present while the benefits, risks and procedures were read to the volunteer. All questions were answered and the volunteer has agreed to take part in the research. _______________________ __________________________________ Date Signature of Witness I certify that the nature and purpose, the potential benefits, and possible risks associated with participating in this research have been explained to the above individual. _______________________ __________________________________ Date Signature of Person Who Obtained Consent University of Ghana http://ugspace.ug.edu.gh 280 Appendix G: Ethical Clearance University of Ghana http://ugspace.ug.edu.gh 281 University of Ghana http://ugspace.ug.edu.gh 282 University of Ghana http://ugspace.ug.edu.gh