ABSTRACT Title of dissertation: HETEROSEXUAL PARENTS? GENDER ROLE ATTITUDES, RELIGIOUS ORIENTATION, HETEROSEXIST BELIEFS, SUPPORT GROUP EXPERIENCES, AND RELATIONSHIP FUNCTIONING WITH THEIR LESBIAN OR GAY CHILDREN Susanna Maria Gallor, Doctor of Philosophy, 2006 Dissertation directed by: Professor Ruth E. Fassinger Department of Counseling and Personnel Services The purpose of this study was to explore how support group experiences are related to parents? attitudes and relationships with their lesbian, gay, or bisexual children in an effort to understand the development of positive attitudes and relationships and what experiences might facilitate this process. This study assessed the interrelationships among parental attitudes (including gender role attitudes, religious orientation, and heterosexist beliefs), parent-child and family relationships (including current family and parent-child relationship functioning, and changes in relationship functioning following disclosure or participation in a support group), and parents? experiences with psychoeducation and social support (including overall involvement in a support group as well as involvement in support, education, and advocacy activities). Participants were contacted through chapters of Parents and Friends of Lesbians and Gays (PFLAG). The final sample consisted of 167 individuals, 110 females and 52 males, who identified as heterosexual parents of one or more same-sex oriented children. Some key findings suggested that greater parental endorsement of heterosexist attitudes was associated with less positive views of their current parent-child relationship functioning and fewer reports of positive changes in their relationships with their children since their child?s disclosure of her/his same-sex orientation. Results further indicated that the more participants were involved in PFLAG overall as well as in its support, education, and advocacy activities, the less they endorsed heterosexist attitudes. Regression results indicated that parents with higher levels of PFLAG involvement, especially involvement in advocacy, were least likely to endorse heterosexist attitudes. Also, the more parents were involved overall, and the more they were involved in support, education, and especially advocacy, the more they perceived positive changes in their relationships with their children since disclosure. The results of this study confirm previous research that explored some of the steps that parents may take as they attempt to understand, accept, and integrate their lesbian, gay, or bisexual child. It appears that accessing resources in the community, participating in supportive or self-help groups, and engaging in social advocacy are some of these important steps for parents. Additional findings and implications for future research, practice, and advocacy are discussed. HETEROSEXUAL PARENTS? GENDER ROLE ATTITUDES, RELIGIOUS ORIENTATION, HETEROSEXIST BELIEFS, SUPPORT GROUP EXPERIENCES, AND RELATIONSHIP FUNCTIONING WITH THEIR LESBIAN OR GAY CHILDREN by Susanna Maria Gallor Dissertation submitted to the Faculty of the Graduate School of the University of Maryland, College Park in partial fulfillment of the requirements for the degree of Doctor of Philosophy 2006 Advisory Committee: Dr. Ruth Fassinger, Chair Dr. Charlie Gelso Dr. Estina Thompson Dr. Marylu McEwen Dr. Pepper Philips Dr. Roy Gilbert-Higginson ?Copyright by Susanna Maria Gallor 2006 ii Acknowledgements I would like to acknowledge my committee members ? Dr. Charlie Gelso, Dr. Estina Thompson, Dr. Marylu McEwen, Dr. Pepper Philips, and Dr. Roy Gilbert-Higginson ? for all of their time, support, and feedback. Each of my committee members helped make this study stronger and more focused, and I appreciate all of their wisdom. I would like to thank my close friends Stuart D?Arcy, Jenny Ventura, Tracy Allen, Pooja Sukhwani, and Meredith Levine who have provided constant support and love, and who always make me feel like I am home. I would also like to thank Stacy Cohen who supported me through the beginning of this ambitious project and through the most difficult years of graduate school, and Nancy Hensler-McGinnis who has not only been an amazing colleague but also a true friend. I would especially like to thank my partner and best friend, Jennifer Weiner, for her endless love and patience and for teaching me what is truly significant in research and in life. And many, many thanks to my family ? Roly, Annie, Mom, and Dad ? for their unconditional support and love. I don?t know where I would be today without their encouragement and guidance. Finally, I would like to thank my advisor and mentor, Ruth Fassinger, for always encouraging me to do research and work that I feel passionate about and for teaching me the true importance of social support and advocacy. iii TABLE OF CONTENTS Acknowledgements ii Chapter I: Introduction 1 Chapter II: Review of the Literature 7 Heterosexism and Attitudes Toward Gay Men and Lesbians 7 Gender, Gender Role Ideology, and Heterosexism 10 Religious Ideology and Heterosexism 19 Disclosure/Coming Out and Parental Reactions 27 Disclosure and Parent/Child Characteristics 31 Disclosure and the Parent-Child Relationship 40 Self-Help, Psychoeducation and Social Support 51 Psychoeducation and Social Support 63 Overall Summary and Statement of the Problem 66 Chapter III: Method 73 Participants 73 Instruments 80 Demographic Questionnaire 80 PFLAG Involvement 83 Heterosexist Beliefs 84 Gender Role Attitudes 86 Religious Orientation 90 Family Functioning and the Parent-Child Relationship 93 Procedures 96 Chapter IV: Results 98 Pilot Study 98 Preliminary Analyses 101 Group Differences 103 Primary Analyses 105 Correlational Analyses 105 Regression Analyses 108 Chapter V: Discussion 123 Summary of Findings and Comparison to Existing Literature 123 Parental Attitudes and Family Relationships 123 Support Group Involvement 129 Limitations and Strengths 139 Implications 142 Research 142 Practice 145 Training and Supervision 147 iv Advocacy 149 Conclusion 150 Appendix A: Recruitment Email to PFLAG Chapter Leaders 152 Appendix B: Demographic Information 153 Appendix C: PFLAG Involvement 156 Appendix D: Attitudes about Lesbian, Gay, and Bisexual People 160 Appendix E: Attitudes about Gender Roles 161 Appendix F: Religious Experiences 163 Appendix G: Family and Parent-Child Relationships 166 Appendix H: Cover Letter 169 References 170 1 Chapter 1 Introduction ?Coming out? ? disclosing one?s same-sex sexual orientation to others ? often is considered to be an important step in developing a gay or lesbian identity (Cass, 1979, 1996). Disclosure to family, however, has the potential to affect not only the gay or lesbian individual, but it may affect family members as well (Beeler & DiProva, 1999; Holtzen & Agresti, 1990; Savin-Williams & Dube, 1998). There is little known about the parental ?coming out? process, particularly from the perspective of the parent (Savin- Williams & Dube), as well as the subsequent nature of parents? relationships with their gay or lesbian children. Popular culture writers have proposed a developmental sequence of stages that parents face upon first learning of their child?s gay or lesbian identity, but there are few empirical investigations of these ?mourning and loss? stages, including periods of shock, denial, anger, guilt, acknowledgment, and, finally, acceptance or integration. Other authors have proposed and begun to explore various developmental stage models of social justice and heterosexual ally models, and these more closely resemble what several qualitative studies, personal narratives, and a few empirical studies have illustrated as parents? struggle with but ultimate acceptance of their children?s gay and lesbian identities (e.g., Beeler & DiProva, 1999; Herdt & Koff, 2000). As noted in Matthews and Lease (2000), many authors have offered models that suggest that families move through a coming out process similar to what lesbians and gay men experience in their own identity development, and that the initial turmoil may progress to some level of acceptance, or at least acknowledgment, that allows for continued connection (e.g., 2 Dahlheimer & Feigal, 1994; Mattison & McWhirter, 1995; Strommen, 1990). The empirical research in this area generally has explored parents? initial reactions to disclosure, predictors of parents? potential to accept or reject their child (e.g., parents? attitudes toward gender roles, religious orthodoxy, age, education), and single-item, open-ended, or global ratings of the parent-child relationship. In a summary of empirical findings, Savin-Williams and Dube (1998) observed that particularly noteworthy is the indication that parents characterize themselves as more accepting than lesbian and gay youths report their parents to be, that extreme negative reactions and rejections appear to be relatively rare, that parents move toward acceptance relatively quickly after disclosure, and that a wide range of parental responses is normative. Several authors have also discussed the notion that, not only are families immersed in a heterosexist culture, but most, if not virtually all, families share many of the heterosexist assumptions and biases of society (e.g., Beeler & DiProva, 1999; Strommen, 1989). The challenge to family members upon disclosure, then, includes overcoming their own heterosexist beliefs and attitudes and contending with the heterosexism of friends, neighbors, coworkers, their community, and society at large. Some research has illustrated how some of the same factors related to negative attitudes toward gays and lesbians (e.g., attitudes toward gender roles, religious orthodoxy, age, education) may influence parents? struggle to redefine their identities as well as their relationships with their children (e.g., Beeler & DiProva, 1999; Herdt & Koff, 2000). In general, this research exploring factors associated with heterosexist attitudes has demonstrated that individuals who score high on scales of homophobia or heterosexism have had little personal contact with gay men or lesbians and tend to be older and less 3 well educated on average than those with more positive attitudes toward gays and lesbians. In addition, individuals with more negative attitudes toward lesbians and gay men are more likely to identify themselves as very religious, and are especially likely to identify with conservative or fundamentalist religions. Moreover, these individuals are more likely to hold traditional attitudes toward gender roles and have conservative attitudes toward sex in general. Finally, individuals with more negative attitudes toward gays and lesbians are more likely to be men than women ? on average, men express more homophobic attitudes overall, especially toward gay men (Bohan, 1996; Herek, 1996; Kite & Whitley, 1996). Although much has been explored related to heterosexual individuals? homophobic and heterosexist attitudes, there has been little consideration of these attitudes and their implications in the context of the relationships between gay, lesbian, or bisexual (GLB) individuals and their families of origin. Guideline 8 of the American Psychological Association?s Guidelines for Psychotherapy with Lesbian, Gay, and Bisexual Clients (2000) states: ?Psychologists strive to understand how a person?s homosexual or bisexual orientation may have an impact on his or her family of origin and the relationship to that family of origin.? Thus, psychology, and counseling psychology in particular due to its emphasis on person-environment interactions (Gelso & Fretz, 1992) and multicultural issues, is in a position to address the unique family and parenting issues associated with gay and lesbian individuals? coming out process and identity development. However, when the family of origin is mentioned in what Laird (1996) called ?the heterosexual clinical literature? and in the psychological literature on gay and lesbian identity development, it often is portrayed solely as ?the thing that one does or 4 does not ?come out to?? (p. 97). The prevailing picture that emerges is often one of disappointment, rejection, compromise, loneliness, and physical and/or emotional withdrawal of family members, particularly parents. In contrast, a few studies and narrative accounts recently have provided an emerging portrait of lesbians and gay men in family relationships that go beyond the trauma of disclosure, beyond hostility and emotional distance, to redefinition and recreation and to a much more complex view of how lesbian and gay individuals define and construct family networks (Laird). Ben-Ari (1995) summarized various concerns that parents frequently hold when their child has come out to them, distinguishing between concerns that are child-oriented and those that are parent-oriented. Child-oriented concerns include fears that the child would grow old lonely and unhappy, would never have children, would suffer prejudice and guilt, would be rejected by other family members and society, would suffer from sexually transmitted diseases, would be harassed and arrested, and would lose her/his religious beliefs. Parent-oriented concerns include having no grandchildren and, thus, no continuation of their children?s generation, feeling like a failure as a parent, feeling alienated from the child as well as from the larger family and the community, experiencing conflict between love for the child and moral and/or religious values and beliefs, having difficulties integrating a lover of the same sex into the family, and confronting their own sexuality. Some of these concerns are interrelated, such as the concerns that the child will never have children and the parent will never have grandchildren, or that the child would suffer prejudice and guilt and the parent would experience conflict between love for the child and moral/religious values and beliefs. 5 Although the current literature and empirical findings are important in understanding initial reactions, parent concerns, and several aspects of parent-child relationships, they do not directly address the process that parents experience in accepting their gay or lesbian child. What determines whether and how heterosexual parents accept their gay or lesbian children remains unclear. A few qualitative studies have begun to explore some of the steps that families, and parents in particular, take as they attempt to understand, acknowledge, accept, and integrate their gay or lesbian child. Some examples include accessing resources in the community, reintegrating their child into family rituals, including the child?s friends and partners, engaging in social advocacy around gay and lesbian issues, and participating in supportive or self-help groups. The self-help and support group literature has focused most of its attention on groups or organizations for families with a child who has a severe mental or physical disability, individuals who have experienced the loss of a loved one, caregivers of people with psychiatric and medical conditions, and individuals or family members of individuals suffering with substance addictions (e.g., Winefield, Barlow, & Harvey, 1998; Pomeroy, Rubin, & Walker, 1996). Minimal research has examined self-help or support groups for parents and families who are struggling to understand and accept the coming out of a lesbian or gay family member. Much of this literature on the efficacy of family psychoeducation and family interventions focuses on the families and support networks of individuals with chronic physical and mental illnesses (e.g., Dixon, Adams, & Lucksted, 2000). The information regarding the effectiveness of interventions such as support groups, psychoeducation, and other group interventions may be useful in exploring the potential effectiveness of these 6 types of interventions for families of lesbians and gay men even though the focus would be on markedly different populations confronted with a different set of reactions, concerns, needs, and attitudes. Indeed, some research has demonstrated that many self- help groups effectively serve their members, and results have shown how participants in self-help groups maintain the benefits and gains of the group interventions to a greater extent than control groups (Gartner, Gartner, & Ouellette-Kobasa, 1988). The purpose of the current study was to explore how support group experiences are related to parents? attitudes and relationships with their children in an effort to expand current understandings of parents? own coming out process. Based on previous research that clearly has illustrated factors associated with attitudes toward gay and lesbian individuals, this study assessed the constructs of gender role attitudes, religious orientation, and heterosexist beliefs and how these constructs relate to parents? perceptions of change over time with regard to overall family functioning and the parent- child relationship, particularly within the context of parents? involvement in organizations designed to provide support to parents and families with a same-sex oriented family member. Thus, this study sought to understand the interrelationships among parental attitudes and biases, parent-child relationship and family functioning, and parents? experiences with psychoeducation and social support specific to having a child with a same-sex identity. 7 Chapter 2 Review of the Literature In order to provide a rationale for linking the variables of interest in this study, literature is organized and reviewed in the following sections: Heterosexism and Attitudes Toward Gay Men and Lesbians (including a section on Gender, Gender Role Ideology and Heterosexism and a section on Religious Ideology and Heterosexism), Disclosure/Coming Out and Parental Reactions (including a section on Disclosure and Parent/Child Characteristics and a section on Disclosure and the Parent-Child Relationship), and Self-Help, Psychoeducation, and Social Support. It should be noted that the majority of the research in these areas has focused primarily on attitudes toward and relationships with lesbians and gay men (to the exclusion of bisexual and transgender individuals). Thus, the current study and the literature reviewed here focuses mainly on lesbian and gay male individuals, unless otherwise noted. Heterosexism and Attitudes Toward Gay Men and Lesbians ?Heterosexism? has been defined (Herek, 1990) as an ?ideological system that denies, denigrates, and stigmatizes any nonheterosexual form of behavior, relationship, or community,? (p. 316) and Fernald (1995) extended this definition to include the denial, denigration, and stigmatization of nonheterosexual persons. Although negative social attitudes toward gay men and lesbians often are referred to as ?homophobia,? several authors have argued that the term heterosexism is a more appropriate and inclusive concept (e.g., Fernald, 1995; Herek, 1990), as it includes a focus on the normalizing and privileging of heterosexuality (rather than merely a fear of homosexuality). As described in Bohan (1996), homophobia or heterosexism exists in many forms and on many levels 8 of experience. ?Personal homophobia? is reflected in an individual?s personal prejudice against lesbian and gay individuals (and traits stereotypically associated with these). ?Interpersonal homophobia? is evidenced when one?s personal attitudes are expressed in relationships with others. Interpersonal homophobia is where personal bias leads to actual discriminatory acts, ranging from homophobic jokes to rejection to physical violence. ?Institutionalized homophobia? or ?cultural heterosexism? (Herek, 1995) entails the promotion by society in general of heterosexuality as the sole, legitimate expression of sexuality and affection. This form of heterosexism includes not only the pervasiveness of this ideology in the explicit teaching of heterosexual normativity, but also the tacit communication of this ideal via society?s norms, institutions, laws, cultural forms, and even scientific practices (Bohan, 1996). According to Simoni and Walters (2001), heterosexual individuals in U.S. society possess a privileged status, and this privilege can be evidenced in marriage, custody, and adoption rights; tax and insurance benefits; anti- discrimination protection in housing and employment; and benefits and protections in terms of military service, inheritance, hospital visitation, pensions, and immigration. A privileged status allows the privileged to experience their daily life and identity as routine, all encompassing, normal, neutral, and universal (McIntosh, 1988), and is, therefore, generally invisible to those experiencing the privilege. Herek (1996) described three purposes he believed attitudes endorsed about lesbians, gay men, and bisexual people by heterosexuals may fulfill: (a) Value-expressive, wherein the attitude serves to reinforce important values and concepts of self; (b) social- expressive, wherein the attitude serves to strengthen one?s sense of group identity and gain group acceptance; and (c) ego-defensive, wherein the attitude serves to diminish 9 anxiety resulting from conscious or unconscious conflicts surrounding sexuality and gender identity. This functional framework indicates that no single variable will explain most of the variance in attitudes held by a heterogeneous group. Rather, several variables will be significant because of their importance to the social and psychological needs of different subsets of individuals in the group. Herek (1994) summarized several findings with the Attitudes Toward Lesbians and Gays scale (ATLG; Herek, 1984b), one of the most widely used scales that reliably and validly measures attitudes toward lesbians and attitudes toward gay men. ATLG scores are consistently associated with several social psychological variables. These include 1) attitudes about gender and family roles, 2) religiosity, 3) political ideology, and 4) the extent and quality of interpersonal contact with lesbians and gay men (Herek, 1988; Herek & Glunt, 1993b). In short, heterosexuals are more likely to hold positive attitudes toward lesbians and gay men to the extent that they accept nontraditional roles for men and women, are not religious (defined by three religiosity variables: liberal denomination, church attendance, and fundamentalist beliefs), describe themselves as politically moderate or liberal, and have had positive interpersonal experiences with gay men or lesbians (Herek, 1994). In the current study, the heterosexist attitudes of parents who have a son or daughter with a same-sex orientation were assessed in order to explore the relationships among parental attitudes, parental responses to their child?s gay or lesbian identity, the parent-child relationship, and parents? experiences with and utilization of supportive and educational resources in their own ?coming out? process. Heterosexist attitudes were assessed using a new scale developed by Jeeseon Park (in progress) called the 10 Heterosexism Scale. As noted above, homophobia and heterosexism exist in several forms, and the present study sought to capture individual and personal aspects of homophobia/heterosexism as well as the cultural and institutional aspects of homophobia/heterosexism, as captured by the Heterosexism Scale. Other parental attitudes this study explored in relationship to heterosexism and parents? coming out experience included gender role attitudes and attitudes toward religion. Gender, Gender Role Ideology, and Heterosexism Anselmi and Law (1998) define gender roles as ?socially and culturally defined prescriptions and beliefs about the behavior and emotions of men and women? (p. 195). Anselmi and Law further state that gender roles influence people?s perceptions of others as well as their own behavior and feelings. Greene (2003) argued that heterosexist beliefs are based on gendered ideas about what men and women are supposed to be and on what constitutes normal behavior for them. According to Greene, heterosexist thinking leads to a range of inaccurate assumptions that are commonly held about gay men and lesbians. For example, one false belief is that to be gay or lesbian is to want to be a member of the other sex (Kite & Deaux, 1987). Another erroneous assumption that connects heterosexism and gender dichotomies is that there is a direct connection between sexual orientation and a person?s conformity or failure to conform to traditional gender stereotypes of roles and physical appearance (Kite & Deaux). As Greene notes, these assumptions, in addition to various other mistaken assumptions, ?suggest that to understand the meaning and reality of being a lesbian or gay man requires a careful exploration and understanding of the importance of cultural gender roles, the nature of the culture?s traditional gender role stereotypes, the relative fluidity or rigidity of those 11 roles, their range, rewards for conformity, and punishments for failure to conform? (p. 368). Kite and Whitley (1998) offered one analysis of attitudes toward homosexuality based on the assumption that heterosexuals? evaluations of gay men and lesbians are rooted in a broader belief system about women, men, and their appropriate roles (Deaux & Kite, 1987; Kite, 1994). According to Kite and Whitley, this belief system has two implications for attitudes toward homosexuality. First, people?s gender-associated beliefs (e.g., stereotypes, gender role attitudes) appear to be inextricably linked. For example, many people believe that individuals who possess stereotypically masculine traits also adopt stereotypically masculine roles and possess stereotypically masculine physical characteristics and, similarly, that those who possess feminine characteristics on one dimension are likely to be feminine on other dimensions. Evidence that this belief system is tied to heterosexuals? perceptions of lesbian and gay individuals comes from demonstrations that men who are described as having feminine characteristics are judged likely to be gay whereas women described as having masculine characteristics are judged likely to be lesbian. People similarly infer that gay men have the gender-associated characteristics of heterosexual women and that lesbians have the gender-associated characteristics of heterosexual men (e.g., Kite & Deaux, 1987). Second, it appears that people?s evaluations of those who contradict traditional gender roles (e.g., people who engage in role behaviors associated with the other sex or who possess characteristics associated with the other sex) generally are not positive (e.g., Jackson & Cash, 1985; Laner & Laner, 1980). Kite and Whitley (1998) state that this may be particularly true for gay people, who, as discussed above, are stereotypically 12 perceived as having cross-sex traits, roles, and physical characteristics and who are apparently disliked as a result, particularly by those with traditional gender role attitudes. For example, Sandnabba and Ahlberg (1999) sought to explore parents? attitudes toward cross-gender boys (boys who adopt traditionally feminine characteristics) and cross- gender girls (girls who adopt traditionally masculine characteristics), including the possible reasons for different perceptions of cross-gender girls and boys. The authors found that predicted psychological adjustment and perceptions of likelihood of same- gender sexual behavior in adulthood were found to be related to the degree to which children were believed to adopt cross-gender characteristics. In addition, adults? attitudes toward cross-gender boys and cross-gender girls were found to differ significantly. Boys who did not adhere to traditional gender-roles were more negatively regarded than girls who adopted cross-gender characteristics. Both men and women predicted cross-gender boys to be less psychologically well-adjusted as adults than cross-gender girls and men predicted cross-gender boys to be more likely to grow up to be gay than ?typical? boys (Sandnabba & Ahlberg). Kite and Whitley?s (1998) discussion of gender-associated beliefs and their relevance to antigay attitudes also may be one explanation for sex differences in attitudes toward homosexuality, a result that has been found consistently in the literature, including Sandnabba and Ahlberg?s (1999) study. Kite and Whitley explain that research has shown that expectations for consistency in others? gender-associated characteristics appear to be more firmly held for men than for women. For example, gender-associated traits and gender-associated physical characteristics are more narrowly defined for men than for women (Hort, Fagot, & Leinbach, 1990). In addition, people appear to react 13 more negatively to violations of the traditional male gender role than violations of the traditional female gender role (Herek, 1986; Kite & Whitley, 1996). One study (Herek, 1988) sought to investigate the basis for differences among heterosexuals in their reactions to gay people and centered this inquiry around the issue of gender differences. Herek proposed three questions. First, is the intensity of heterosexuals? attitudes toward gay people consistently affected by gender of the (heterosexual) respondent, the (gay) target, or both? Second, what is the relative contribution of other social psychological variables to heterosexuals? attitudes? Third, do these correlates exert a differential influence on the attitudes of heterosexual men and women, and do they have a differential effect on attitudes toward lesbians and toward gay men? In Herek (1988), three separate studies with student samples were reported that address these three questions. In the first study, a measure of attitudes toward lesbians and gay men was administered to a sample of 368 undergraduate students (249 females, 119 males), along with measures of related variables. The dependent measure, attitudes toward lesbians and gay men, was assessed using the ATLG scale. Independent measures were assessed using five sets of variables that previous research had suggested might be important in explaining ATLG scores. The first variable was sex-role attitudes, assessed with the 25-item short form of the Attitudes toward Women Scale (AWS; Spence & Helmreich, 1972). The other variables included authoritarianism, perceived social support, personal contact (assessed by asking respondents how many of their close male and female friends during the past two years were gay or lesbian and whether their past interactions with gay males or lesbians were generally positive or generally negative), 14 and religiosity, again defined by liberal denomination, church attendance, and fundamentalist beliefs. Scores for male and female respondents were compared, and the role of the five independent variables in predicting attitudes toward lesbians and gay men was assessed through a multiple regression analysis. The second study was conducted in order to assess the stability of the findings in the first study with different samples in different geographic locales. The same instruments and items were administered to 405 student volunteers (226 females, 179 males) at six different universities. In these first two studies, results indicated that heterosexual males consistently expressed more negative attitudes than did heterosexual females. Further, it was found that heterosexual males hold more negative attitudes toward gay men than toward lesbians. As expected, the regression analyses revealed that attitudes are affected by respondents? adherence to traditional values concerning gender and family, degree of religious orthodoxy, and past experiences with lesbians and gay men. These variables were associated with a substantial amount of the variance in ATLG scores; however; none of them appeared to be more influential than the others in determining attitudes, nor did these variables differentially influence heterosexual males? and females? attitudes (Herek, 1988). Herek (1988) next conducted a third study to replicate these findings and to assess the effects of an additional category of variables. Measures of variables related to intrapsychic conflicts (defensiveness, personal insecurities about one?s own gender identity hypothesized to be associated with hyperconformity to gender roles, and personal insecurities related to differences/similarities between themselves and gay or lesbian individuals) were added to the questionnaire battery to examine their influence on 15 attitudes toward lesbians and gay men. A new sample of heterosexual undergraduate volunteers (n = 149) from the same university as in the first study completed the ATLG and a battery of questionnaires. Results indicated that males once again displayed more negative attitudes than did females, with the effect more pronounced for attitudes toward gay men. Further, seeing differences between oneself and gay people on gender-related attributes was important only for ATG scores; male respondents with negative attitudes were likely to perceive greater differences between themselves and gay men. As in the previous regression analyses, ATL and ATG scores were associated with traditional views of sex roles and conservative religious ideology. In addition, tolerant attitudes seemed to be associated with perceptions of oneself as not fitting stereotypes of either masculinity or femininity. In summary, Herek (1988) found that attitudes toward lesbians and gay men are influenced by at least four separate sources: Hostility is associated with traditional attitudes about gender- and family-roles; perceptions that one?s friends hold similarly negative attitudes; strong adherence to an orthodox religious ideology (see below for a more in-depth review of religious ideology and attitudes toward gays and lesbians); and past negative experiences with gay people. Herek (1988) proposed that sex differences in attitudes might be understood best by considering cultural constructions of gender, since men and women are likely to have different experiences associated with the principal correlates of homophobia (perceived attitudinal norms, religiosity, personal contact with gay people, and ideologies of family and gender). Further exploring gender differences in attitudes toward gay men and lesbians, Kerns & Fine (1994) first sought to replicate earlier findings that heterosexual men held 16 more negative attitudes toward gay men and lesbians than did heterosexual women. The second purpose of the study was to assess whether traditional gender role attitudes mediated the relationship between gender and negative attitudes toward gay men and lesbians. In order to identify a plausible mediator variable, the authors utilized a gender role socialization model in which children and adolescents develop their gender role attitudes from socialization experiences with their parents and/or other influential individuals. Kerns and Fine hypothesized that because gay men and lesbians deviate from cultural expectations for the roles that men and women should perform in their lives (Kite, 1994), one would expect that heterosexual individuals who are more accepting of those who deviate from traditional gender roles would have more positive attitudes toward gay men and lesbians. Participants were 155 students in an introductory psychology course (59% female; 97% Caucasian). In addition to demographic information, the ATLG scale (Herek, 1988) was used to measure the extent of negative attitudes toward gay men and lesbians, and Kurdek?s (1988) Traditional Attitudes Toward Men, Women, and the Equality Between Men and Women Scale was used to measure gender role attitudes. Consistent with previous research, results indicated that heterosexual males had more negative attitudes toward gay men than did heterosexual females. Both male and female participants, however, had more negative attitudes toward gay men than toward lesbians, and there were no gender differences in attitudes toward lesbians. Finally, results indicated that traditional gender role attitudes did, in fact, mediate the relationship between gender and negative attitudes toward gay men (Kerns & Fine, 1994). 17 In Whitley (2001), the author?s first goal was to compare predictions derived from gender schema theory (Bem, 1981) and multifactor gender identity theory (Spence, 1993) concerning the strength of the relationships of gender-role self-concept and gender-role beliefs to attitudes toward homosexuality. A second goal of this research was to examine the strength of the relationships of adherence to traditional beliefs concerning male versus female gender-roles to attitudes toward homosexuality. Study 1 was a meta-analysis of the results of previous research on this topic and Study 2 collected new data to determine the degree to which several gender-role variables were independently related to attitudes toward homosexuality as well as antigay behavior. In Study 2, participants were 211 women and 183 men in an introductory psychology course (88% White). Participants? attitudes toward lesbians and gay men were assessed, along with their endorsement of stereotypes of lesbians and gay men and their level of antigay behavior. Other variables assessed included traditional gender-role beliefs, endorsement of traditional male role norms, hostile and benevolent sexism, modern sexism, and hyper-gender-role orientation (including hypermasculinity for men and hyperfemininity for women). The results of both studies clearly demonstrated the relationship of gender-role beliefs to attitudes toward homosexuality (Whitley, 2001). The meta-analysis in Study 1 indicated that traditional gender-role beliefs and adherence to ?old fashioned? or traditional sexism are closely linked to negative attitudes toward homosexuality. Endorsement of traditional male role norms was no more closely related to attitudes toward homosexuality than endorsement of traditional female norms, and the results of the meta-analysis did not provide either clear support or lack of support for Bem?s (1981) 18 gender schema theory. In Study 2, the major predictors of negative attitudes were participant gender, endorsement of traditional male role norms, traditional sexist beliefs, and benevolent sexism. In addition, Study 1 found hypermasculinity to be related to negative attitudes and Study 2 found both hypermasculinity and hyperfemininity to be correlated with both attitudes and antigay behavior, although they had unique relationships only with behavior. In summary, several authors have discussed how gendered ideas about what men and women are supposed to be and about what constitutes normal behavior for them, including gender stereotypes and attitudes toward gender roles, influence people?s attitudes toward gay men and lesbians (e.g., Deaux & Kite, 1987; Greene, 2003). Several empirical studies have explored and consistently found relationships among gender, gender role attitudes, and heterosexist beliefs (e.g., Herek, 1988; Kerns & Fine, 1994). This literature, however, is not without limitations. First, external validity and generalizability are limited given that the majority of participants in many of the studies were White and college students. Moreover, some studies did not differentiate between attitudes/behavior toward lesbians and attitudes/behavior toward gay men, which have been shown to differ (Herek & Capitanio, 1999). Finally, the measures utilized in many of the studies need to be reconsidered in terms of their appropriateness and usefulness. For example, many of the measures used to assess homophobia or antigay behavior tend to assess overt and blatant acts of discrimination and fail to consider more subtle, ?invisible? acts of discrimination and/or avoidance. Also, in terms of gender role attitudes, one measure often used, the Attitudes Toward Women Scale (Spence & Helmreich, 1972), is limited in its scope in that attitudes toward male roles and/or a 19 comparison of people?s attitudes toward both women?s and men?s roles are not assessed. For the current study, attitudes toward gender and gender roles were measured, in part, by the Sex Role Egalitarianism Scale-Short Form (SRES; King & King, 1997), a scale developed to measure attitudes toward the equality of women and men in marital, parental, employment, socio-interpersonal-heterosexual, and educational roles. In addition, items related to attitudes toward sexual initiative and casual sex were included. These eight items were composed of items from two subscales of the Gender Attitudes Inventory (GAI; Ashmore, Del Boca, & Bilder, 1995) and parallel items regarding attitudes toward male roles constructed by the primary investigator. Religious Ideology and Heterosexism Gordon Allport proposed a religious typology in which some people have a religious orientation that is primarily extrinsic, a self-serving, instrumental approach conforming to social conventions. Others, in contrast, have an intrinsic religious orientation; religion provides them with a framework through which all life is understood and given meaning. Allport and Ross (1967) summarized this distinction: ?The extrinsically motivated person uses his [sic] religion, whereas the intrinsically motivated lives his [sic] religion? (p. 434). Thus, these authors shifted the focus away from the content of faith (i.e., specific beliefs) or the number of times that a person attended religious services to the meaning faith had for an individual. Allport and Ross reported that an extrinsic orientation tends to be positively associated with prejudice, whereas an intrinsic orientation tends to be negatively correlated. The relationship between religious orientation and attitudes of intolerance is found consistently in the literature specifically with regard to racial prejudice (Larson, Cate, & Reed, 1983). Less research has been conducted regarding non- 20 racial discrimination, specifically with lesbian and gay populations. However, there is evidence that a relationship between intolerant attitudes and religious orientation exists for lesbian and gay populations as well (e.g., Fulton, Maynard, & Gorsuch, 1999; Herek, 1987). For example, Herek (1987) sought to examine the relationship between intergroup prejudice and religiosity by comparing White heterosexuals? attitudes toward lesbians and gay men with their racial attitudes. Herek suggested that perhaps ?an intrinsic motivation may be less a force for tolerance than an extrinsic motivation is a force for socially fostered prejudice? (p. 35). According to Herek, because most denominations strongly condemn racial bigotry, intrinsics? tolerance may reflect simple compliance rather than internalization of an ethic of love for one?s neighbor. Herek proposed a study of the influence of religious orientation on racist attitudes as well as on attitudes toward a group that is itself condemned by many Western religious denominations ? gay and lesbian individuals. In Herek (1987), extrinsics were expected to manifest more racism than were intrinsics. Further, because hostility toward gay people is generally the norm, extrinsics were expected to be hostile toward them as well. For intrinsics? sexual attitudes, however, the outcome was not predicted. Questionnaires were distributed by instructors to students at four universities. Because antiblack racism and attitudes toward gay and lesbian individuals were conceptualized as intergroup attitudes, respondents were included in the analysis only if they reported being White and having engaged exclusively in heterosexual behaviors since age 16, using the behavioral scale described by Kinsey, Pomeroy, and Martin (1948). This left a total of 126 respondents (80 females and 46 21 males). The Religious Orientation Scale (ROS; Allport & Ross, 1967) was used to assess intrinsic-extrinsic religious orientation, the ATLG was used to assess attitudes toward lesbians and gay men, seven items measuring symbolic racism (conceived to be a subtle form of racism, in contrast to overt expressions of racial hostility) were used to assess racist attitudes, and the Religious Ideology Scale (RIS; adapted from Putney & Middleton, 1961) was used to assess religious ideology. Results suggested that the influence of religious orientation on prejudice depends on the out-group in question. When religious teachings encourage tolerance, intrinsic persons report less prejudice than do extrinsics. When religious teachings do not encourage tolerance or are themselves persecutory, however, intrinsics appear to be no less prejudiced than extrinsics and perhaps more so. In that case, conservatism of religious beliefs better predicts prejudice than does religious orientation. Intrinsically oriented persons who adhere to religions that condemn homosexuality are more likely to be hostile to gay men and lesbians than are those belonging to denominations with a more accepting stance. Thus, according to Herek (1987), an intrinsic orientation is not associated with tolerance per se. Rather, it is associated with tolerance toward groups identified as deserving of tolerance by one?s religious philosophy. Furthermore, although earlier researchers found a simple positive association between racism and hostility based on sexual orientation, the data presented in Herek (1987) show that, for some people, the relationship might be negative. Using Allport and Ross?s (1967) model of religious orientation as a guide, McFarland (1989) sought to further differentiate between extrinsic religion (using religion for ulterior motives such as security, comfort, status, or social support) and 22 intrinsic religion (using religion as the orienting center of one?s life and motivation). A factor analysis of the Intrinsic and Extrinsic scales (Kirkpatrick, 1989) revealed the Extrinsic scale to have three subscales: (a) Social Extrinsic (reflecting the social benefits of religion), (b) Personal Extrinsic (reflecting the personal comfort, protection, and relief that religion provides), and (c) Religious Seriousness. One of the purposes of McFarland?s study was to investigate whether the entire Extrinsic scale or one of its subscales was primarily responsible for the positive relationship with racial discrimination. McFarland posited that the Social Extrinsic subscale would be predictive of discrimination. He proposed that individuals who enjoy social relationships would be likely to avoid discomfort in relationships by avoiding people who are different from them. More recently, the study of the relationship between religion and psychological functioning has expanded to include two other forms of religious orientation: fundamentalism and quest. An additional purpose of McFarland?s (1989) study was to explore the concept of quest, which McFarland predicted would relate negatively to homophobia. Quest religion involves an openness to change and a questioning approach to religion; views religious doubt as positive; and resists simple, clear-cut answers to complex existential questions (Batson, Schoenrade, & Ventis, 1993; Sciarra & Gushue, 2003). Previous research has demonstrated that a quest orientation is independent of both extrinsic and intrinsic orientations (Batson, 1976), and several studies have demonstrated that quest is indirectly related to discriminatory or prejudicial attitudes (e.g., Batson, Flink, Schoenrade, Fultz, & Pych, 1986). Finally, McFarland hypothesized that fundamentalism would correlate with several kinds of discrimination including sexual 23 orientation and a general tendency to discriminate. Altemeyer and Hunsberger (1992) provided the following definition of fundamentalism: ?The belief that there is one set of religious teachings that clearly contains the fundamental basic, intrinsic, essential, inerrant truth about humanity and deity; that this essential truth is fundamentally opposed by forces of evil which must be vigorously fought; that this truth must be followed today according to fundamental, unchangeable practices of the past; and that those who believe and follow these fundamental teachings have a special relationship with the deity? (p. 118). Participants were 247 undergraduate White students who identified themselves as Christian (McFarland, 1989). The following measures were utilized: The Intrinsic and Extrinsic Religion Scales (Allport & Ross, 1967); the revised Quest Scale (Batson & Ventis, 1982); a six-item scale measuring fundamentalism; a scale adapted from Herek?s (1987) Attitudes Toward Gay Men and Attitudes Toward Lesbians scales; a single-item 5-point scale measuring religious seriousness; and a measure reflecting discrimination against Blacks. Results showed that Intrinsic Religion was positively correlated with discrimination against gays and lesbians, but not against Blacks. Conversely, Extrinsic Religion was not related to discrimination against gays and lesbians, but was positively related to discrimination against Blacks. These results were consistent with previous research that showed differences in the influence of religious orientation on prejudice depending on the out-group in question (Herek, 1987). Fundamentalism was associated not only with discrimination against gays/lesbians and Blacks, but was also associated with a general tendency to discriminate. As predicted, quest was related to less discriminatory attitudes 24 on the basis of sexual orientation, race, gender, or belief system. However, it was most strongly associated with a general anti-discriminatory attitude (McFarland, 1989). One additional study attempted to understand further the distinctions among each type of religious orientation and their relationships with discriminatory attitudes. Fulton, Maynard, and Gorsuch (1999) noted that because intrinsic and extrinsic religious orientations are religious motivations, they may relate inconsistently to other variables (e.g., racial discrimination, antigay discrimination) unless the nature of that to which a person is committed is taken into account. The authors hypothesized that if this is so, partialling fundamentalism, a belief measure, from the intrinsic-antigay correlations should change them in a more tolerant direction. Thus, it was hypothesized that: 1) Fundamentalism would be correlated with antigay sentiment; 2) Intrinsic religious motivation would a) correlate negatively with antipathy toward Blacks and positively with antipathy toward homosexuals and b) be uncorrelated or negatively correlated with antipathy toward homosexuals after statistically controlling for fundamentalism; 3) Extrinsic (including Extrinsic-social and Extrinsic-personal) religious motivation would correlate positively with antipathy toward both Blacks and homosexuals; and 4) Quest religious orientation would correlate negatively with antipathy toward both Blacks and homosexuals. Participants were 257 students from a conservative Christian college. Only White heterosexual students were selected for the study, yielding 110 female participants and 66 male participants. Participants were administered scales assessing I/E religious orientation, Quest religious orientation (Q), and Fundamentalism (F). Scales assessing antiblack attitudes as well as antigay attitudes were also administered. For the antigay 25 attitudes, Fulton et al. differentiated ?morally rationalized antigay items? (e.g., ?Homosexuality is a perversion?) from ?nonmorally rationalized items? (e.g., ?A person?s homosexuality should not be a cause for job discrimination?). As predicted, F correlated little with antiblack prejudice but highly with antigay attitudes, and F correlated with both the moral and nonmoral antigay statements. Intrinsic orientation (I) was only positively correlated with morally legitimated homosexual antipathy on the attitude scale; nonmorally legitimated attitudes were uncorrelated with I. Fulton et al. (1999) stated that these findings suggest that, despite significant morally based homosexual antipathy, people who score high on I are able, in the vocabulary of Christianity, to ?love the sinner, but hate the sin.? In addition, consistent with previous research (McFarland, 1989), I was strongly correlated with Fundamentalism. When Fundamentalism was controlled for, the relationship between I and nonmoral antigay attitudes went from uncorrelated to negatively correlated. Moreover, even the positive relationship between I and morally based antigay attitudes decreased to essentially zero when Fundamentalism was controlled. As Fulton et al. noted, these findings support those of earlier studies (Herek, 1987; Kirkpatrick, 1993; McFarland, 1989) which found positive correlations between I and prejudice against gays and lesbians. In terms of Extrinsic religious orientation, Fulton et al. (1999) found that Extrinsic-social was associated only with nonmorally legitimated antihomosexual attitudes. This suggests that the homosexual antipathy of persons high in Es (as was found in previous studies, e.g., Kirkpatrick, 1993) has less to do with maintaining theological consistency, and more to do with maintaining effective social boundaries. Finally, Quest was either negatively correlated or uncorrelated with measures of 26 antipathy with all groups, and uncorrelated with Fundamentalism. These findings support the finding of McFarland (1989) that showed negative correlation between Quest and all discriminatory attitudes. Fulton et al. concluded by suggesting that the relationship of tolerance to religious orientation is not a simple function of commitment to one?s religion. Rather, the nature of the religious commitment also must be considered. In summary, various studies have illustrated the complex relationship between religion and homophobia. Orthodox and fundamental religious orientations both are related directly to homophobic attitudes. In general, low scores on religiosity indicate more tolerant attitudes toward same-sex orientations. More specifically, intrinsic religion has been found to be related directly to antigay prejudice and discrimination against GLB individuals, except when fundamentalism is controlled. Extrinsic religion as a whole has been shown to be unrelated to discriminatory attitudes against same-sex orientations, although there are mixed results with regard to Extrinsic-Social and Extrinsic-Personal religious orientations. Finally, a quest religious orientation has been shown to be negatively correlated or uncorrelated with discriminatory attitudes toward GLB individuals. As with the literature exploring gender role attitudes, the literature exploring religious orientation is limited in terms of external validity and generalizability. The majority of the studies consist of samples with White college students who are predominantly Christian. The current study assessed religious orientation using several measures that have been used and validated previously; however, some of the scales? items and/or instructions were slightly modified in an effort to make the scales more widely applicable to individuals of different religions. Intrinsic and extrinsic religious orientations were assessed with the Age Universal I/E Scale-Revised (Maltby & Lewis, 27 1996). Quest religious orientation was assessed with the Age Universal Quest Scale- Revised (Maltby & Day, 1998). Disclosure/Coming Out and Parental Reactions As Savin-Williams and Dube (1998) noted, popular culture writers have proposed a developmental sequence of stages that parents face upon first learning of their child?s homosexuality, a sequence similar to that experienced by individuals undergoing grief and mourning. Personal narratives, popular advice-giving tracts, and ?self-help? books written by and for parents often present the act of disclosure by youths of their gay/lesbian identity as necessarily creating a crisis within the family. In contrast to previous publications, Savin-Williams and Dube redirected the focus from youths onto parents and reviewed empirical research that addresses the purported developmental stages that characterize parental reactions to disclosure. According to Savin-Williams and Dube (1998), many writers have proposed paradigms by which parents evolve through a series of stages from initial shock to eventual acceptance of their child. These proposed models follow the delineation of stages originally proposed by Kubler-Ross (1969) that characterize individuals facing imminent death. Prior to stage one, shock is an initial parental reaction to the disclosure by a child that she or he is gay/lesbian. In stage one, denial and isolation, parents realize the severity of their new knowledge, and then deny the reality of their child?s homosexuality. The denial is usually an anxious one, but it provides a time for parents to regain their bearings and equilibrium. In stage two, anger, parents react with agitation, dismay, or rage at their child; sometimes it leads to rejection or physical abuse. As a defensive maneuver, parents search for an external cause for their child?s homosexuality 28 and become irrationally angry at this perceived perpetrator. Bargaining, stage three, is an attempt to regain equilibrium by working out ?a deal.? Religious parents may beseech God for a conversion of the child?s sexual orientation. As parents move toward acceptance, they may bargain with the child to tell absolutely no one, sometimes including the other parent, and to never again discuss the issue. The fourth stage is characterized by depression and, in some cases, resigned tolerance. This may include a guilt that parents feel for not recognizing their child?s ?condition? early enough to change the outcome or for being the kind of parent that ?causes? a child to be gay or lesbian. In this stage, many parents are saddened by the life they believe their child must now endure. Consistent with social stereotypes, they only imagine the sexual aspects of being gay, believe that their child will inevitably be lonely in her or his old age, or face discrimination that will result in the child leading a clandestine life. Support groups, educational materials, and mass media presentations are thought to be instrumental in moving parents toward acceptance, the fifth and final stage, which implies that they essentially have completed their mourning, and their son?s or daughter?s homosexuality is no longer a family secret or source of shame (Savin-Williams & Dube). According to DeVine (1984), parental desires to have a heterosexual child have become less significant in this final stage of acceptance; however, because of previous teachings and stereotypes imposed by the social norm of heterosexuality, few individuals attain the status of a proud, self-professed parent of a gay, bisexual, or lesbian child. Similarly, DeVine (1984) described certain stages that parents might experience as they deal with the news that their son or daughter is gay: (a) Subliminal awareness ? involving the existence of some vague suspicions, often based on a-typical gender role 29 characteristics and behaviors; (b) Impact ? involving the actual discovery of the child?s identity, best characterized as involving a ?crisis,? accompanied by such reactions as shock, denial, confusion, blame, anger, and guilt; (c) Adjustment ? involving attempts to deal with the crisis, which might be to get their son or daughter to change or to keep the child?s sexual identity a secret, so as to maintain the social, religious, and professional respectability of the family; (d) Resolution ? involving a ?working-through process,? consisting of mourning the wish to have a heterosexual child and the dreams and plans associated with this as well as learning more about homosexuality and modifying one?s own stereotypes about what it means for someone to be gay; and (e) Integration ? involving integrating whatever mourning and learning that needed to occur into an acceptance of one?s daughter or son for who the child is, and as the child is. DeVine further discussed that the movement through these stages is governed by three aspects of the family as a system: the ?cohesion? or closeness of family members, the ?regulative structures? or rules that govern family member behavior, and the ?family themes,? or defining values and behaviors that dictate the family members? views of themselves and their interaction with the larger community. In a discussion of this stage model and other literature on reactions of family members, Strommen (1989) suggested that the traumatic nature of family member reaction consists of two related processes: (a) the application of negative values about homosexuality to the disclosing member, and (b) a perception that homosexual identity negates or violates previous family roles. Empirical research seldom has tested the validity of grieving models against the real-life reactions parents have on learning about the sexual-minority status of their child (Savin-Williams, 2001). Indeed, some limited evidence runs counter to these models. In a 30 review of the limited empirical research regarding parental reactions to a child?s disclosure, Savin-Williams and Dube (1998) called into question the appropriateness of a stage model based on grief reactions to describe the experiences of parents. From the perspectives of both parents and sexual-minority youths, the range of documented first parental reactions to disclosure is vast. Some parents react with shock, displays of anger, and denial; others react with knowing inevitability, acceptance, and unconditional love. As Savin-Williams and Dube discussed, oftentimes previous suspicion heightens the parents? awareness that they have a lesbian or gay child, and this suspicion typically arises from the child?s gender atypical behavior or interests. Moreover, Matthews & Lease (2000) pointed out that because individuals come out at different stages in their lives, the issues related to family may vary depending on when this occurs. For instance, the issues for an adolescent who is still living at home and financially and otherwise dependent on parents may be different from those for a self-sufficient middle-aged adult with relatively little day-to-day contact with the family of origin. Similarly, the families are likely to experience such situations differently as well (Matthews & Lease). In addition, Savin-Williams (2001) noted that factors that predict how parents react once their child discloses to them are poorly understood. One speculation is that parents? response depends on how they discover the child?s same-sex attractions ? whether they are told directly by the child or indirectly by others. Another speculation is that the child?s age when the discovery occurs determines parental reactions, parents being more likely to deny the stability of the child?s sexual identity if their child discloses to them at a young age. Alternatively, the age of the parents might influence their reactions to having a gay or lesbian child. Other factors that have been proposed to affect 31 parental reactions to gay, lesbian and bisexual family members include the family culture, parenting style, religion, ethnicity, and sex-role orientation (Savin-Williams). As Savin- Williams noted, none of these variables has been systematically pursued. Disclosure and Parent/Child Characteristics Much of the research exploring disclosure and parental reactions has focused on characteristics of parents that seem to be related to negative attitudes towards gays and lesbians more generally and differences in knowledge of or reaction to their son?s or daughter?s same-sex identity more specifically. One characteristic that has demonstrated some interesting differences is parent and/or child gender. Much research has found differences between level of outness to mothers and level of outness to fathers. This is apparent in studies of adult children (Ben-Ari, 1995; Cramer & Roach, 1988) and youths (Herdt & Boxer, 1993; Savin-Williams, 1990). In addition, some studies have found more specific differences among parent-child dyads; however, virtually no empirical research has systematically explored differences among mother-daughter, mother-son, father- daughter, and father-son relationships. In one study with a sample of 194 lesbians, gays, and bisexuals (aged 14-21) attending social, recreational, and support groups in fourteen metropolitan areas, D?Augelli, Hershberger, and Pilkington (1998) found that among the very few (9%) youths whose first disclosure was to a parent, all chose mothers. Overall, of those who told parents, 65% revealed to mothers, 9% to fathers, and 25% to both. In general, very few mothers did not know or suspect their children?s sexual orientation, and most mothers who knew about their children had discussed the issue with them. More mothers of lesbians did not suspect (18%) than mothers of gay males (7%). The fathers of lesbians were even less knowledgeable: only 42% knew, compared to 59% of the gay 32 males? fathers; nearly half (46%) of fathers of lesbians did not know or suspect, compared to 16% of fathers of gay males. In an earlier study with a sample of 202 lesbians, gays, and bisexuals (aged 14 to 21; 147 males and 55 females; 30% Black, 40% White, 12% Hispanic, 3% Asian American, 15% of mixed ethnic background) from a youth group in metropolitan Chicago, Boxer, Cook, and Herdt (1991) found that a larger number of girls reported that their mothers and fathers were aware of their sexual orientation (mothers, 63%; fathers, 37%) than did boys (mothers, 54%; fathers, 28%). Participants more frequently disclosed their sexual orientation to their mothers than to their fathers, regardless of their gender. Boxer et al. (1991) also found that both males and females reported better relations with mothers than with fathers, a difference that was statistically significant. They noted, however, that relations with fathers were not necessarily more conflictual, but were sometimes less emotionally salient. The age at which gay and lesbian individuals come out to their parents also may be an important factor in parents? reactions to disclosure. According to Boxer et al. (1991), emerging evidence suggests that the age of self-identification as gay or lesbian has been lowering over the last 50 years. These authors further noted that self- identification as gay or lesbian during adolescence may be a unique developmental process found only in current cohorts of some gay or lesbian youth and carry different consequences for the development of parent-child relations over the life course. According to Savin-Williams (2001), youths are increasingly coming out to parents at an earlier absolute age, often while still in high school or, occasionally, while in junior high or middle school. Research has not tested the hypothesis that reactions of parents are more severe if the child is not an adult (Savin-Williams & Dube, 1998), but some 33 research has explored family members? reactions and the parent-child relationship in the context of the child?s age of disclosure and place of residence. For example, Boxer et al. found that the youth?s age was a significant predictor of the quality of the parent-child relationship. Older youth reported better relations with mothers and with fathers than younger youth. In addition, several youth reported changes in relationships with mothers (64%) over the past year, as well as changes in relationships with fathers (50%). For mothers, age of the youth was the most important and the only significant predictor of the direction of change in the relationship. Older youth reported more positive changes in relationships than did younger youth. The same result was found for quality of change in relationships with fathers. In another study, D?Augelli et al. (1998) found that family members were not seen as safe confidants for 16-year-olds who were residing at home. D?Augelli noted that the risks of disclosure occur at a time at which the young person?s family relationships are undergoing changes associated with adolescence. Many of the youth in this study used avoidance to maintain harmony, likely an important concern for adolescents still dependent on their families. Moreover, although many of the youths in this study eventually disclosed their same-sex identity to at least one immediate family member, only half of mothers and siblings, and one-quarter of fathers, were fully accepting; another one-quarter of fathers and 10% of mothers were actively rejecting. In many cases, disclosure to family members was associated with threats and with verbal and physical abuse. Consideration of racial and ethnic identity, another important parent and child characteristic, in the context of coming out experiences and parent-child relationships is 34 almost nonexistent. Greene (2003) discussed the conflated nature of heterosexism in that heterosexism is not a singular or isolated experience or event. As such, heterosexism cannot be disconnected from the broader context of an individual?s development or existence. Although there is virtually no empirical research exploring these factors, some work has explored coming out experiences within multicultural contexts. For example, the assertion that young gay people of color can encounter additional problems and obstacles while coming out to family members due to their ethnic, racial, and cultural backgrounds has been documented (e.g., Merighi & Grimes, 2000; Savin-Williams, 1996). In a series of studies examining sexual identity development among 83 ethnic- minority youths, Dube and Savin-Williams (1999) found that fewer than half of ethnic- minority youths reported disclosure to family members. The authors suggested that this may reflect their greater fear of rejection and disappointment from family members, a lack of emotional closeness, or cultural norms discouraging talk about sex. Rust (2003) suggested that because there are cross-cultural differences in the ways families function, coming out to one?s family raises different issues for individuals belonging to different racial/ethnic groups. For example, in some strong family-centered cultures, part of being a woman or man is fulfilling one?s family role by marrying and parenting children (e.g., Chan, 1989). Parents may interpret their children?s coming out as lesbian, gay, or bisexual as a rejection of the family and of their ethnicity. Thus, parental pressure toward heterosexuality must be understood as partially motivated by a desire to preserve ethnic identity (Rust). In a qualitative study of 57 African-, European-, Mexican-, and Vietnamese- American gay men, aged 18 to 24, Merighi and Grimes (2000) found that the reactions of 35 family members ranged from being very supportive to becoming disengaged. In particular, four general types of responses emerged from the data: 1) support through action (e.g., involvement in gay rights demonstrations); 2) support that preserves a kinship bond; 3) avoidance; and 4) distancing and disengagement. In addition, the majority of the respondents discussed how their culture influenced the manner in which they came out to their families (i.e., directly or indirectly). Although many of the stories that were told centered on how cultural norms and ideals worked against disclosure (e.g., the potential for adversely affecting the perceptions of others toward the family), several participants viewed the importance of family loyalty and preservation as a foundation upon which to feel less fearful of being rejected. How young gay people of color or their heterosexual family members negotiate the disclosure of their same-sex identity is an area in need of further investigation. Some studies have explored various characteristics more broadly and in conjunction with each other, including gender and age, as they relate to parents? responses to their GLB son or daughter. For example, Holtzen and Agresti (1990) explored parental characteristics such as sex-role attitudes, homophobia, and self-esteem in an attempt to explore the complex interaction of cognition and affect in attitude development and change. Measures of homophobia, self-esteem, and sex-role attitudes were completed by heterosexual parents of a gay or lesbian child to determine if differences were evident between those parents who had known for some time about their child?s sexuality and those parents who were newly informed about their child?s sexuality. The five hypotheses tested were: 1) Homophobia will correlate negatively with measures of nontraditional sex role attitudes, with measures of social and global self-esteem, as well as with the length 36 of time a parent has known about his or her child?s sexuality; 2) Homophobic parents will have known about their child?s sexuality for a shorter time, will evidence significantly more sex-role stereotypes, lower social self-esteem, and lower global self-esteem than non-homophobic parents; 3) Recently informed parents will show a significantly higher degree of homophobia, more traditional sex-role stereotyping, lower social self-esteem, and lower global self-esteem than parents who have known for a longer time; 4) Recently informed homophobic parents will evidence a decrease in their homophobia and changes in other measures at a four month follow-up; and 5) Length of time since disclosure coupled with sex-role stereotyping will predict a parent?s degree of homophobia. Participants in Holtzen and Agresti (1990) included 55 heterosexual parents who were knowledgeable regarding their child?s sexuality. Participants were contacted through seven chapters of Parents and Friends of Lesbians and Gays (PFLAG) and seven college organizations for gay and lesbian students. PFLAG members received their materials directly from a contact person within their chapter, and students? parents received their materials through their children. Of the 55 participants, 91% (n = 50) agreed to participate in a follow-up mailing. However, only 18 participants (36%) responded to the second mailing and of these only two had known of their child?s sexuality for a short period of time. Thus, hypothesis four could not be tested. Parents agreeing to participate in the study represented several geographic areas. Sixteen of the total number of respondents were fathers and 38 were mothers. Thirty-three parents had gay sons and 22 had lesbian daughters. All respondents were White with a mean educational level that included some college. The Index of Homophobia (IHP; Hudson & Ricketts, 1980) was used as a measure of both affective and cognitive domains of 37 homophobia. A short version of the Attitudes towards Women Scale (AWS; Spence & Helmreich, 1972) was used to measure sex-role stereotyping. Two scales, the Self- Esteem Inventory for Adults (SEI; Coopersmith, 1982), and Short Form A of the Texas Social Behavior Inventory (TSBI; Helmreich & Stapp, 1974) were used to measure self- esteem. Results indicated that homophobic parents had knowledge of their child?s gay or lesbian sexuality for a relatively shorter period of time, had a lower sense of social self- esteem and endorsed more traditional sex-role stereotypes than relatively non- homophobic parents. Significant differences in homophobia occurred between recently informed parents (less than two years) and parents who had had knowledge of their child?s sexuality for many years. Whether there occurred significant changes over time in this sample is unknown because there were insufficient follow-up data. Interestingly, significant differences were found between those parents who agreed to participate in the follow-up mailing and those who did not wish to take part in it. The former group was significantly less homophobic and endorsed significantly fewer sex-role stereotypes than the latter group. Robinson, Walters, and Skeen (1989) sought to obtain another a parental perspective regarding relationships with their gay or lesbian children, particularly in terms of their responses to the initial knowledge that their child was gay or lesbian, the process of adjusting to their child?s same-sex identity, their feelings of responsibility, and their attitudes toward the AIDS epidemic. Participants were 402 parents of gay and lesbian children, 105 fathers and 298 mothers. The ages of the parents ranged from 37 to 82; 97% were Caucasian, most were well educated, and most reported middle to upper 38 level incomes. Also, most considered themselves participants in organized religion, with about 1/3 considering themselves very religious. They were recruited through two national organizations (302 from the Federation of Parents and Friends of Lesbians and Gays [PFLAG] and 99 from the National Federation of Parents and Friends of Gays [PFOG]). In addition, a snowball procedure was used in which (a) gay and lesbian young adults were asked to supply the names and addresses of their parents and (b) participants were asked to supply names and addresses of other parents. In PFOG, chapter leaders were sent questionnaires and consent forms, along with separate postage-paid envelopes, and were asked to distribute the materials during a regular meeting. Members of PFLAG received explanation letters, request forms, and postage-paid return envelopes through their regular newsletter. Both group leaders and individual parents who requested questionnaires received them through the mail. Questions were written by the authors and pretested with a group of 20 parents of gay or lesbian children. Items also were evaluated for face validity by national officers of PFLAG. Initial reaction was assessed by asking parents to circle all words in a list (e.g., sad, guilty, embarrassed, glad) or write in other words that described the way they felt when they first learned for sure that their child was homosexual. Parents then were asked about the adjustment process they might have experienced in accepting their child?s homosexuality in terms of the following stages: shock, denial, guilt, anger, and acceptance. In addition, parents? feelings of responsibility for the child?s homosexuality, feelings about the AIDS epidemic and its influence on responses to their children, and parents? level of liberalism versus conservatism were assessed on 4-point Likert-type scales (Robinson et al., 1989). 39 The majority of the parents in this study described themselves as somewhat liberal or very liberal (48% and 30%, respectively). Almost all of the parents, however, indicated some form of negative initial reaction to learning of their child?s same-sex identity. Most reported some form of regret; confusion, disbelief, sadness, and guilt also were reflected in many reports. Although 2% said that they were glad to learn that their child was gay, these parents wrote notes indicating that it was a relief to confirm what they had suspected, not that they were glad the child was gay or lesbian (Robinson et al., 1989). As they responded over time to knowledge that their child was gay, 64% (63% of fathers, 66% of mothers) reported that they experienced a five-stage progression of mourning and loss in dealing with their emotions: shock, denial, guilt, anger and acceptance. Despite initial difficulty, 97% of mothers and fathers indicated that they had worked through earlier stages and had arrived at the level of acceptance. Some parents, however, had not progressed that far and, as suggested by Robinson et al., may have gotten stuck in one of the stages or regressed to an earlier stage from time to time. A majority of parents in this sample did not attribute their child?s homosexuality to inadequate relationships with themselves or their spouses. In addition, the vast majority of parents (87%) believed their children were born gay or lesbian. Finally, the majority of the parents (71%) said they worried that their children might catch AIDS. Parents of lesbians were not as worried as parents of gay men. Most parents (89%), however, indicated that the AIDS outbreak had not altered their feelings toward their children or their views of homosexuality in general. Robinson et al. (1989) and Holtzen & Agresti (1990) were two of the first few empirical studies that explored disclosure and coming out issues from the perspective of 40 parents. However, several limitations warrant caution in interpretation and generalization of the findings. First, data come from parental support groups, and parents of gays and lesbians who join these types of groups and are available to participate in research have identified themselves publicly. Parents of gays and lesbians who are not members of a support group may have a different set of concerns. The high acceptance level of the parents in this study indeed may be a function of the fact that so many were participants in the support groups from which they were recruited. Second, generalizability of the findings is limited because of overrepresentation of mothers, and, in one study, gay male children. Moreover, all respondents were White and well-educated. Finally, internal validity may be limited for both studies. In the Holtzen and Agresti study, the homophobia scale and gender role scale are limited in scope in that the homophobia scale does not assess more subtle heterosexist beliefs, and the gender role scale does not assess attitudes toward men?s roles. Also, Robinson et al. developed all of the items for their questionnaire, and psychometric properties were not reported. Disclosure and the Parent-Child Relationship Savin-Williams (1998) noted that few researchers have addressed the relationships that sexual minority youths have with their parents other than the initial disclosure reactions or a global rating of how important the parents are to the child. Limited empirical evidence suggests that parent-child relationships are positive and satisfying, especially with the mother, but this conclusion is incongruent with many of the personal narratives of youths and parents presented in the popular literature (Savin-Williams). One study that has delved into an empirical investigation of the parent-child relationship in the context of a child disclosing his same-sex identity is that of Cramer and Roach (1988). 41 The first goal of this study involved investigating a parent?s potential to reject a gay son when disclosure occurs. Several predictors previously shown to be related to homophobia were explored: religious orthodoxy, belief in traditional sex-roles, authoritarianism, age, education, and being raised in an urban or rural setting. The second goal was to gather more information regarding how parents respond over time to a gay son. A questionnaire was designed to retrospectively investigate the changes over time in parental relationships with a gay son who comes out. Measures of the respondent?s perception of his parents on the six selected predictors were assessed. Also, a Likert-type scale was used to rate each respondent?s parent-son relationships before disclosure, immediately following disclosure, and presently. Rejection by the parent was defined as the parent-son relationship deteriorating after his coming out. A reported improvement or no change in the relationship was interpreted as parental acceptance (Cramer & Roach). Participants were 93 males with a mean age of 32.6. The participants were mostly well educated (72% had a college degree) and predominantly Anglo (94%). To obtain participants, contact was made with a variety of gay organizations and groups, and the procedure for obtaining participants for this study basically relied upon respondents picking up a questionnaire, filling it out, and mailing it to the authors in prepaid, preaddressed envelopes. Findings from this study indicated that most relationships between gay sons and their parents were strained following disclosure, and a period of turmoil usually ensued for most families. However, over time the gay son?s relationship with both his mother and father tended to improve, and in many cases these relationships became better than they were prior to disclosure, although one-third of the participants 42 felt disclosure had no long-term effect on their parental relationships (Cramer & Roach, 1988). In addition, results indicated that the mother?s level of education, location of childhood, and perceived attitudes on sex roles, religious orthodoxy, and authoritarianism could significantly predict the change in the relationship between her and her son. The father?s educational level, location of childhood, and his perceived attitude on religious orthodoxy significantly predicted the change in the relationship between him and his son. Surprisingly, the mother who was perceived as having more traditional sex-role attitudes and the father who was perceived as higher in religious orthodoxy were both reported to be actually more accepting of their sons following disclosure. Cramer and Roach (1988) suggested this may be for two possible reasons: Parents who are highly traditional may find strength in their conviction concerning the value of family unity to overcome a threat to the family?s cohesion that helps the family survive and remain a unit. Another possible interpretation may be that the son held certain expectations of how highly traditional mothers and orthodox fathers would react to disclosure. Sons of these parents may have felt more distant and ?hidden,? and when their fears were not realized, their now open relationships with their parents became less burdensome, resulting in a report of improvement. Results from Cramer and Roach (1988) illustrated that parental values and characteristics associated with homophobia were good predictors of the change in parent- son relationship following disclosure. One limitation of the study, however, is that parental acceptance was defined as reported improvement or no change in the relationship, regardless of what the relationship was like before disclosure. For example, 43 there was no clear assessment or description of the parent-child relationship prior to disclosure, and to claim that parental acceptance includes those instances when the relationship was negative and/or distant prior to disclosure and remained negative and/or distant following disclosure is questionable. In addition, parental acceptance is assessed through the perspective of the children, more specifically the sons, of the parents, and it is difficult to generalize results to the actual perceptions, reactions, and attitudes of the parents themselves. Finally, this study?s external validity is further limited because of its sole focus on gay males and, thus, results cannot be generalized to parents and their lesbian daughters. In another study looking at disclosure and the parent-child relationship from the perspective of the children, Floyd, Stein, Harter, Allison, & Nye (1999) examined separation-individuation during adolescence and young adulthood for gay, lesbian, and bisexual (GLB) youths. Floyd et al. (1999) tested several sets of hypotheses related to parental attitudes regarding sexual orientation, separation-individuation as a predictor of personal adjustment and identity consolidation, age and gender effects on separation- individuation, and effects of orientation on parent-youth relationships. For the purposes of the current review, only those hypotheses and results related to disclosure and the parent-child relationship (as opposed to the sexual orientation identity development of the youths) will be discussed. One hypothesis predicted that negative parental attitudes would be associated with estrangement in the form of high levels of autonomy paired with low levels of relatedness, along with less conflictual independence for the respondents. Another hypothesis examined whether the participants demonstrated age-related advances in separation-individuation from parents similar to other youths and young 44 adults, such that relatively older respondents would tend to display greater autonomous functioning, more positive relatedness, and greater conflictual independence with their parents. Floyd et al. also examined possible gender-related differences and explored questions regarding whether child and parent gender relate to the factors examined in this study. Finally, the authors examined open-ended responses to questions about participants? impressions of the impact of sexual orientation on their relationships with their parents, expecting that most youths would acknowledge stressful aspects for both themselves and for their parents, as well as positive impacts on their relationships. Participants included 72 adolescents and young adults (36 males, 36 females; 79% Caucasian), ages 16-27, who identified themselves as lesbian, gay, or bisexual. Participants were recruited in the upper midwest and southeast regions of the United States from youth support groups and campus organizations. Each participant completed a 2-hour interview, followed by a battery of questionnaires. A condensed version of the ?Young Adult-Parent Relationship Interview? (Frank, Avery, & Laman, 1988) was used to gather information relevant to autonomy and relatedness with each parent and consisted of 23 open-ended questions concerning contacts with the parent, sources of closeness and tenseness in the relationship, self-disclosure, support, disagreements, and impressions of the parents? strengths and weaknesses. During the interview, respondents rated the attitudes of each parent regarding the respondent?s sexual orientation on 5-point scales that ranged from 1 = ?very negative/completely unsupportive? to 5 = ?very positive/completely supportive.? Respondents also completed the ?conflictual independence? subscale of the Psychological Separation Inventory (Hoffman, 1984), which examines negative emotional and cognitive reactions to separation. 45 Overall, results indicated that, similar to other adolescents and young adults, the GLB youths in this study faced normative developmental challenges associated with renegotiating relationships with their parents, and that these challenges were made more difficult when parents had negative attitudes toward the youth?s sexual orientation. Specifically, more positive attitudes by the parents regarding the sexual orientation of their children were significantly correlated with greater relatedness with both mothers and fathers, and greater conflictual independence with both parents. However, perceptions of parental attitudes were not correlated with autonomy from either of the parents. Only a few age and gender effects on separation-individuation were discovered: Being older was associated with greater autonomy from mother and greater conflictual independence with father. Also, both the male and female respondents acknowledged greater positivity in their relationships with mothers than with fathers. More specifically, respondents received higher scores for relatedness with mother than father, and they also reported that their mothers had more positive attitudes regarding sexual orientation. Finally, a content analysis of the responses to open-ended questions about the impact of the youth?s sexual orientation on relationships with the parents revealed 9 categories of specific responses: no effect; brought us closer; causes tension, closes communication; creates distance, estrangement; has ruined relationship; worsens a bad relationship; youth empathic about struggle for parent; youth resentful of parent?s reaction; mixed effects, or don?t know. Nearly 1/4 and 1/3 of the respondents claimed that their sexual orientation had no impact on their relationships with mothers and with fathers, respectively. Only 4% and 9% claimed that their sexual orientation either totally destroyed or worsened an already bad relationship with their mothers and fathers, respectively (Floyd et al., 1999). 46 It is commendable that Floyd et al. (1999) broadened the scope of their investigation by including an examination of various individual differences and relationship variables for both gay sons and lesbian daughters through quantitative as well as qualitative methodologies. As the authors noted, the advantage of using an interview procedure over self-report measures to examine parent-child relationship variables is that it allows the respondent to speak freely about incidents, impressions, and reactions to the parents rather than merely judging the quality of those relationships. One disadvantage of the authors? choice of focus for the relationship measure, however, is the limited scope of definition the authors applied to the parent-child relationship. Although autonomy and relatedness are certainly important components of the parent-child relationship, defining the relationship solely in terms of separation/individuation fails to consider other components that have typically been explored in this type of relationship (e.g., communication, conflict, positive and negative affect, attitudes regarding parental or child role, cohesion). Moreover, it is difficult to generalize the results of this study given that results are based on the children?s perspectives both of the parent-child relationship as well as the parents? attitudes and feelings. Although children?s perspectives are equally as important and valid in understanding the impact of disclosure and the child?s sexual orientation on the parent-child relationship, children?s perceptions of their parent?s attitudes and feelings may be inaccurate or biased. Finally, it is clear from the literature on attitudes toward gay men and lesbians previously reviewed that attitudes are varied and complex. Typically, the measurement of attitudes is comprised of a variety of factors, such as heterosexism and gender role beliefs. Thus, Floyd et al.?s single-item assessment 47 of parent attitudes regarding sexual orientation appears limited, especially given that psychometric information was not presented. In a study attempting to incorporate the parental perspective, Ben-Ari (1995) investigated how family dynamics prior to the discovery of same-sex identity could be related to post-discovery experiences for both children and parents. The main research question in this study explored the various associations between attitudes, feelings, and background factors that manifest themselves prior to and following the discovery of a child?s same-sex identity. The study consisted of two samples: a sample of 32 gay and lesbian adults (60% males, mean age of 32.7) and a sample of 27 parents (70% mothers, mean age of 61.8). Sixteen of the parents? children were sons (60%) and eleven were daughters (40%). Six of the fathers interviewed had gay sons and two of the fathers had lesbian daughters. Ten of the mothers interviewed had gay sons and nine mothers had lesbian daughters. Ben-Ari reported that snowball was the main method of recruiting all of the participants, and the majority of the parents were recruited from support groups for parents of gays and lesbians in the San Francisco Bay Area. The gay and lesbian young adults in this study were not related to the sample of parents. Through an in-depth interview, parents were asked to describe both their children?s and their own experiences. Similarly, gay men and lesbians were asked to describe both their parents? and their own experiences. The participants were first asked a series of open-ended questions. They were then administered a demographic questionnaire and the Parental Reaction Scale, which specified the following reactions: shock, denial, shame, guilt, anger, rejection, acknowledgement, and acceptance. Participants were asked to rate the intensity of these reactions as they applied to four time 48 periods: the first week following discovery (T1), the first month following discovery (T2), the first six months following discovery (T3), and the time of the interview (T4). Gay and lesbian participants rated both mothers? and fathers? reactions during these four time periods. Ben-Ari (1995) first presented results of the study according to three distinct stages: pre-discovery experiences, thoughts, and feelings; the actual act of discovery; and post-discovery experiences, thoughts, and feelings. Results indicated that, from the perspectives of both parents and children, the reactions of parents to disclosure over time often did not reveal a typical pattern. Parents reported, however, that they primarily felt shock during the first week. Both children and parents thought that ?being honest, not to hide, not to live a lie? was the most common motive for coming out. Parents who felt their son or daughter had disclosed in order to be more honest about him or herself were significantly more likely later to acknowledge and accept their child?s same-sex identity. These same parents were also less likely to report experiencing shock, denial, and anger following the discovery. In addition, both parents and gay/lesbian participants who cited this motive for disclosure also more frequently considered that the parent-child relationship improved following discovery. ?Fear of rejection? also surfaced as a significant predictor of the experiences of both parents and gay/lesbian adults following discovery. For example, mothers who anticipated that their children feared rejection showed more denial, guilt, anger, shame, and rejection than mothers who believed that their child had not feared rejection. Both parents and children who considered ?fear of rejection? a primary inhibition to disclose were also less likely to perceive improvement in the parent-child relationship following discovery. Overall, however, a significant majority of the interviewed parents ? 84% of 49 the mothers and 63% of the fathers ? said that the relationship with the child improved following the discovery. In addition, participants from both samples thought that parental reactions to discovery of a child?s same-sex identity were the most important aspect of post-discovery family dynamics (Ben-Ari, 1995). Ben-Ari (1995) also found that parents? background and the ways they thought and felt about homosexuality were the two most significant factors in parents? experiences prior to and after disclosure. The gay and lesbian participants associated difficult post-discovery family dynamics with parents who had no previous experience with anything even remotely related to homosexuality. More specifically, gay and lesbian young adults who perceived their parents as not having friends from the gay community, not being involved in civil rights efforts, or not expressing social awareness or sensitivity, were more likely to show shame, denial, guilt, and anger on the Parental Reaction Scale. They also were less likely to show acknowledgment and acceptance across all three time phases. Ben-Ari discussed an important distinction between ?acknowledgement? and ?acceptance.? ?Acceptance? is more than just recognition of a fact; it includes an affective component. As Ben-Ari stated, ?one can acknowledge a fact without accepting it, but not vice versa?acknowledgment is a pre-condition for acceptance? (p. 107). Interestingly, the gay and lesbian participants in this study maintained that at best their parents acknowledged that they were gay, but had not reached full acceptance, while many parents at various stages of the process claimed actual acceptance. This discrepancy between parent and child perceptions, as well as how ?acceptance? and ?acknowledgment? are defined by individuals, needs to be further explored. 50 One major limitation of Ben-Ari (1995) is that there is no indication of participants? race/ethnicity or other demographic variables. Aside from age, it is unclear what other demographic factors may be related to the findings presented in this study. In addition, the retrospective nature of this study may limit the findings because participants? memories of their feelings and experiences may have been reconstructed over time, and may not be entirely accurate. This is a limitation of several studies that comprise the current literature exploring coming out/disclosure and parent-child relationships. There are several additional limitations that are reflected in this literature as a whole. Many of the empirical data that currently exist are based on youths? reports of parental reactions, and several of these retrospective reports are collected weeks to several years after the disclosure. More complicated questions, such as how the parent- child relationship history affects initial and subsequent parental reactions or how the disclosure subsequently changes family dynamics, have been ignored. In addition, research on the reaction a parent has to discovering the same-sex attractions of a child seldom has distinguished the initial response from the current status of the relationship. Responses are usually superficially categorized as ?accepting,? ?tolerant,? ?intolerant,? or ?rejecting.? Moreover, an important question is what is the process by which parents take advantage of the opportunity to move toward acceptance of a sexual-minority child? Finally, much of the current research fails to discriminate between mothers and fathers and their relationships with either gay sons or lesbians daughters. Thus, the gender of both the youth and the parent must be taken into account when describing any aspect of the coming-out-to-parents process. 51 Iasenza, Colucci, & Rothberg (1996) argued that often the parents? response to a child?s revelation of sexual orientation is more reflective of ongoing family patterns of interaction than of attitudes toward homosexuality per se. The current study sought to gain a comprehensive assessment of family and relationship variables as they relate to parents? reactions to their GLB child as well as to their subsequent involvement in and utilization of PFLAG?s support and education. Thus, this study assessed current overall family functioning more generally and the parent-child relationship functioning more specifically. Characteristics such as problem solving, communication, affective responsiveness, affective involvement, and general functioning were assessed using items based on scales from the McMaster Family Assessment Device (Epstein, Baldwin, & Bishop, 1983). In addition, global relationship quality, humor/closeness, tension, parental control/influence, and fights/arguments in the parent-child relationship were assessed using items developed by Aquilino (1999). Self-Help, Psychoeducation and Social Support: ?Turning Points? in Response to Disclosure of Same-Sex Identity Merighi and Grimes (2000) noted that much of the literature has focused on the consequences of coming out to family members, but relatively little has been documented that articulates the ?turning points? from confusion or disapproval to the eventual acceptance experienced by family members in response to initial disclosure. The authors further noted that it is important to explore the process that enables family members to ?take action? and change their perceptions of and response to gays and lesbians. In their interviews with 57 diverse gay males (ages 18 to 24), Merighi and Grimes found that an important step for many family members was to access resources in the community that help build knowledge and awareness of homosexuality. In fact, many families find it 52 helpful to become involved with Parents and Friends of Lesbians and Gays (PFLAG) during their coming out process (Goldfried & Goldfried, 2001; Matthews & Lease, 2000). Goldfried and Goldfried noted that the acceptance process takes time and corrective experiences, and many of these experiences can occur by having contact with other supportive parents ? as is the case with PFLAG. Before discussing more specifically the organization of PFLAG and some of the current literature on parents? experiences with psychoeducation and support related to having a lesbian or gay child, it may be useful to briefly address the nature and functioning of self-help groups more broadly, particularly those most comparable to groups such as PFLAG. In their identification of a typology for self-help groups, Schubert and Borkman (1991) utilize a useful definition that distinguishes self-help groups from other sources of help, such as family or peer networks, self-study, and therapy: self-governing groups whose members share a common?concern and give each other emotional support and material aid, charge either no fee or only a small fee for membership, and place high value on experiential knowledge in the belief that it provides special understanding of a situation. In addition to providing mutual support for their members, such groups may also be involved in information, education, material aid, and social advocacy in their communities. (U.S. Department of Health and Human Services, 1987, p. 5) In discussing their typology, Schubert and Borkman defined several dimensions along which different types of self-help groups could be differentiated, such as external dependence, sources of leadership, role of professionals internally, and kind of 53 knowledge used. One type, federated groups, closely resembles groups such as PFLAG and Alcoholics Anonymous. Such groups are associated with higher levels of their own self-help organization, often at the state, regional, and/or national levels. Thus, resources such as publicity and literature are made available, but the local units retain full control of decisions and functions. In addition, these groups tend to rely heavily upon experiential knowledge, such as personal interaction and sharing. Leaders of these groups could be any of the members who were comfortable with the format of the meetings and who are willing to accept responsibilities related to the group?s maintenance. There is rarely any kind of formal training for leaders, and although professionals might refer someone to such a group, they would rarely, if ever, interact with the group in their caregiver capacity (Schubert & Borkman). As noted in Kingree and Ruback (1994), self-help groups seem to promote well- being through multiple psychological processes, including encouraging participants to adopt more positive and adaptive perceptions of themselves and their problems. Some research has explored how certain perceptions and thoughts can be shaped by the ideologies or belief systems that are transmitted through self-help group meetings and literature (e.g., Maton, Leventhal, Madara, & Julien, 1989), as well as through informal conversations between members. Snowden and Lieberman (1994) described self-help groups as groups that create experiences such as the inculcation of hope and the development of understanding. These authors also described self-help groups as 1) ?cognitive restructuring systems? possessing ideologies about the source of difficulty and about the way individuals need to think about their dilemmas in order to be helped and 2) ?social linkage systems? where people form relationships and receive social support. 54 Lieberman and Snowden (1994) reported that services provided by self-help groups vary widely, from those that only offer support to those that provide multiple services as well as programs directed toward social or political change. There is virtually no current empirical data on self-help group use or demographic information on who participates in groups, but self-help group use for mental health problems appears to be a predominantly White, middle-class phenomenon (Lieberman & Snowden). PFLAG serves as both a self-help support group and an education and advocacy organization for parents, family, and friends of gay, lesbian, bisexual and transgender individuals as well as GLBT individuals themselves. With chapters in over 430 communities throughout the United States, PFLAG provides a resource for understanding the reactions and feelings that may develop when a family member comes out as lesbian or gay and a means for moving towards acceptance. Among the various activities associated with national PFLAG are its involvement in providing a clearinghouse for information about issues and publications related to GLBT issues, publishing informational brochures, providing speakers for public education, supporting local educational initiatives lobbying against discriminatory legislation, organizing interfaith dialogues, and responding to antigay media campaigns (Goldfried & Goldfried). More specifically, PFLAG chapters provide support through monthly meetings, education through speakers, publications, books, and resources, and advocacy through access to news and legislative updates as well as assistance with taking action in local communities. Many families feel helped by talking with others who have shared their experience (Matthews & Lease, 2000). Ben-Ari (1995) found that both parents and 55 lesbian and gay children viewed parental participation in support groups as an important contributing factor in their movement toward acknowledging the child?s homosexuality. Moreover, Goldfried and Goldfried (2001) discussed the notion that, just as GLB individuals undergo a coming-out process ranging from self-recognition as being GLB to integration and acceptance of a GLB identity, parents must develop their own identity as the parent of a GLB child. For parents to ?come out? themselves, they similarly need to recognize and accept the fact that their son or daughter is gay or lesbian. As indicated earlier, this is a process that requires time and corrective experiences as well as a total reorganization of one?s expectations and values. It also involves a socialization process, which is where involvement in PFLAG and other related organizations can be particularly relevant. Given the fact that little attention has been given to how families respond to disclosure over time and to how families integrate a gay or lesbian family member ? and his or her relationships ? once they have come to accept his or her same-sex identity, Beeler and DiProva (1999) sought to explore these questions through a narrative approach. Beeler and DiProva conducted exploratory, unstructured interviews with each family member within four families for a total of 16 interviews. Participation of three of these families was solicited from PFLAG meetings, and one family was volunteered by a graduate student with whom one of the authors was acquainted. All four families were white and middle class. In two of the families, the gay family member was an adult at the time of disclosure, and in the other two families, the gay family member was an adolescent or young adult. Three of the families included a gay male family member, and the remaining family included a lesbian. In one family, disclosure occurred within the last 56 year, in a second family within the last 3 years, and in two families within the last 10 years. Interview transcripts were reviewed to identify active efforts made by family members to accommodate learning that they have a gay or lesbian son, daughter, or other family member. Thus, the units of analysis were activities (defined by active present tense verbs), actions (defined by completed past tense verbs), and events (contexts that provided the opportunity for new experiences and/or actions). Beeler and DiProva (1999) observed several themes throughout the stories of most, if not all, participants. The authors noted that these themes did not occur in any sequence but were interwoven with the context of events within the participants? lives. The first theme was ?establishing rules for discussing homosexuality,? and the authors discussed that many of the relationships were at least partially defined by what can or cannot be discussed, what can or cannot be asked, and what degree of disapproval can or cannot be tolerated. Establishing these rules could be a subtle and ongoing process. The second theme was ?seeking information about homosexuality and the gay community from gay-positive sources,? and almost all of the participants reported learning more about homosexuality in one way or another (e.g., buying books, attending PFLAG meetings, directly asking the gay or lesbian family member). Another theme was ?exposure to gays and lesbians living ?gay and lesbian lives,?? and this exposure could be nonspecific to the gay/lesbian family member (e.g., through gay pride events, through PFLAG involvement) or specific to the gay/lesbian family member (e.g., meeting the family member?s friends and/or partner). A fourth theme was ?including gay and lesbian friends in the family,? and the authors pointed to the distinction between talking about an 57 individual?s gay and lesbian friends, meeting them, and developing a social relationship with them. A fifth theme was ?working through feelings of sadness, loss, and blame,? and the authors noted that the participants? accounts of their experience were only partially congruent with a grieving model. Although it was not pervasive, dominant, or even present in every case, there was evidence that dealing with negative feelings, typically immediately following disclosure, was an important process. The sixth theme was ?the family coming out,? a process by which family members have to go through their own process of ?coming out? to other family members, friends, coworkers, and associates. As Beeler and DiProva discussed, deciding whom, when, and how to tell is an important process that appears to occur over a lengthy period of time. The next theme was ?developing alternative visions of the future,? which means adjusting one?s expectations surrounding the future of the gay/lesbian individual and requiring parents and family to modify their vision of what constitutes a family. The eighth theme was ?stigma management,? and the issue of how to respond when confronted with homophobic comments occurred consistently in the interviews. The final theme was ?developing narrative coherency,? by which the authors meant that constructing a coherent ?story? surrounding the gay/lesbian family member and the impact of disclosure on the other family members? lives was important (e.g., by incorporating disclosure and subsequent events into the context of their history and current situation or by understanding the past in terms of their current knowledge about their gay/lesbian family member). Beeler and DiProva (1999) offered a rich analysis and discussion of several important themes that seem to characterize individuals? experiences with having a family 58 member who is gay or lesbian. As a result of this study, these and other similar families who are successful in meeting the challenge of contending with heterosexism and integrating a gay/lesbian family member offer the potential to understand what sort of activities, beliefs, attitudes, and adjustments are involved in such an accomplishment. Moreover, Beeler and DiProva add to the existing literature in a number of ways, one of which is in the observation that the themes discerned in the interviews occurred independent of the grieving process and may be seen in any and all stages of grieving. Thus, these results do not necessarily negate the notion that parents and family members may experience a sort of grieving and loss process after learning of their family member?s same-sex identity, but this process may be understood as part of a larger adjustment and reintegration process. Beeler and DiProva further noted that a family member?s ?grieving status? may influence the way he or she approaches any of these themes. One limitation of this study, however, is in its limited generalizability. It is difficult to generalize the results as a characterization of how families, generally, respond to disclosure. The small sample size, the highly selective nature of the sample, and the absence of families of color all contribute to limited generalizability. Beeler and DiProva noted that different cultural contexts would likely contribute uniquely to challenges faced by families of color with a gay/lesbian member. In addition, there is little differentiation in this study among the experiences of different family members (e.g., mother, father, sister, brother, grandparent), and it is unclear how different family members? responses vary. Moreover, it is impossible to utilize the results of this study in the context of families who are rejecting or who are finding disclosure exceptionally difficult to accept. 59 Beeler and DiProva, however, offer an important analysis of how these various themes are engaged on multiple levels: intrapsychic, interpersonal, familial, and social. A theoretical model such as this might allow future research to explore and understand how families approach the challenges subsequent to disclosure and provide insight into the interrelationship between the themes observed in this study. In another qualitative study, Herdt and Koff (2000) conducted interviews with approximately 50 parents of lesbians and gays. The parents ranged in age from their late thirties to their seventies, with an average age of sixty, and they were middle-class as well as working-class people. They were primarily White and from Protestant, Jewish, and Catholic families. In most cases, their sons and daughters had come out to them during young adulthood, and the parents were approximately equally represented by mothers and fathers. The length of time since their children had disclosed their homosexuality to them ranged from less than one year to more than ten years. Herdt and Koff located the parents primarily through the Chicago chapter of Parents, Families, and Friends of Lesbians and Gays (PFLAG) and the Horizons Community Services program for Gay, Lesbian, and Bisexual Teens. According to Herdt and Koff (2000), many of the parents in their study progressed from a state of devastation to a relationship of comfort and pride as a result of the process they termed ?integration,? meaning ?the rejoining of a missing element to the whole? (p. 4). In their interviews, Herdt and Koff came to discern a similar path for all of the families. Initially, families encountered a period of ?disintegration? characterized at various points by guilt and blame, secrecy, impaired relations with others, and shame. The stigma associated with homosexuality, and the concomitant fear associated with it, 60 often compelled the gay or lesbian child and his or her family to separate or to deny any parts of the child?s life and personality that could reveal a gay or lesbian identity. It also compelled parents to reconsider a whole range of attitudes and beliefs. Some families moved on to a period of ambivalence involving continued discomfort but with a small degree of hope, often informed by successful disclosure to others. Finally, a number of families entered a state of integration in which the relationships between family members appeared enhanced (e.g., through improved communication or enhanced ability to respond to problems). Through the interviews, Herdt and Koff (2000) documented eight indicators of integration: positive changes in relationships between the target child and the parent; expressions of appreciation of positive contribution child?s coming out has made to parents? life; disclosure to others; inclusion; positive effect on family; acknowledgment of child?s improved state; involvement in PFLAG or gay community; and capacity to project into the future. Thus, integration also appears to be characterized by more congruence between families? public behavior and their private thoughts, appreciation of the child?s sexual orientation, inclusion of their son?s or daughter?s dating partners or friends, involvement with the gay/lesbian community or other parents of gays or lesbians, and the ability to project a future and major life events for their gay or lesbian child. Herdt and Koff (2000) also discussed various factors that seemed to affect a family?s progress along the disintegration-integration path. In addition to observing previously identified elements that contribute to a family?s resilience, such as the family?s capacity to empathize with others and engage in social activism as well as the ability to find meaning, maintain a balance between stability and change, communicate clearly and 61 effectively, and engage in collaborative problem-solving (Walsh, 1998), Herdt and Koff identified other aspects of family life that seemed to enhance or inhibit a healthy response. The authors found that how parents responded to previous crises typically predicted how they would deal with their child?s coming out. Several of the parents had some previous exposure to lesbian or gay people, and this also affected their initial responses. The authors also discussed what they termed ?extra steps? that the parents who reached the integration stage often pursued: purchasing books, reading articles, talking to supportive and knowledgeable therapists and clergy, attending meetings of PFLAG, befriending other parents of gays and lesbians, and becoming familiar with local lesbian and gay communities. As Herdt and Koff noted, the parents learned to ?let go? in various ways: to let go of the myth that only through heterosexual relationships can families and individuals find fulfillment and that having gay or lesbian children can lead only to loss, alienation, and shame; to let go of the children themselves, so that those children can resume their developmental path; and to let go of the parental role enough to determine freely their own next step in life, having successfully launched their children into responsible adulthood. Oswald (2002) studied the resourcefulness that facilitates successful relationships between people of diverse sexual orientations and discussed two processes, intentionality and redefinition, that serve as forms of resilience that enable family members to create and strengthen their family networks. According to Oswald, intentionality refers to behavioral strategies that legitimize and support relationships, including choosing kin (e.g., creating family relationships out of friendships, such as those formed in support group settings), managing disclosure (e.g., by bringing people that are gay-affirming 62 closer together while creating distance from those who are more hostile), building community (e.g., actively building and participating in supportive community resources, which may be located within the gay community, or they may be part of a particular ethnic or religious community), ritualizing (e.g., integrating gay and lesbian loved ones into family of origin rituals, such as weddings, in order to maintain a sense of group cohesion and identity), and legalizing (e.g., using creative strategies to legalize relationships, enabling network members and outsiders to recognize relationships as legitimate). Oswald (2002) defined redefinition as meaning making strategies that create linguistic and symbolic structures to affirm the existence of gay and lesbian people and their relationships. Redefinition processes include politicizing (e.g., actively considering how living within a heterosexist social context shapes relationships, both family and non- family relationships, thereby developing deeper understandings of heterosexism and possibly becoming activists for the rights of sexual minorities), naming (e.g., attaching familial meanings to otherwise unmarked identities, events, and relationships, thereby acknowledging and legitimating their existence), integrating gayness (the melding of homosexuality with other aspects of identity, especially ethnicity and religion), and envisioning family (the process of redefining family to be an ongoing construction that affirms human differences, rather than an inevitable set of biological relationships). Thus, Oswald incorporates all of these processes into a resilience approach that attends to both behavioral strategies and the ongoing construction of meaning within families. This approach identifies families as actively engaged within their cultures and communities and examines the ways in which active engagement facilitates social support. 63 Psychoeducation and Social Support Much of the research on disclosure/coming out and parental reactions, and more specifically the preceding reviewed literature that addresses more of the process by which individuals learn to understand, accept, and integrate their lesbian or gay family members, points to the potential importance of psychoeducation and social support resources in facilitating more positive, productive, and affirming responses to the experience of having a lesbian or gay child. Although there is no literature that discusses more specifically the use and efficacy of psychoeducational programming with families of lesbian and gay children, Schreier and Werden (2000) provide a comprehensive overview of psychoeducational programming development and efficacy in the context of exploring alternative forms of service delivery designed to reach individuals who are not currently using counseling services, such as lesbian and gay individuals. Schreier and Werden defined programming as ?any proactive, remedial, or early-intervention outreach effort that is cognitively oriented and time limited and is designed to enhance intrapsychic change, modify the interpersonal-social context, and create systemwide enhancement of the mental health of a targeted population? (p. 361). The authors further reported that efficacy research on programming designed to address issues faced by lesbian and gay individuals is scarce. It is clear, moreover, that efficacy research on programming designed to address issues faced by family members and parents of lesbian and gay individuals is nonexistent. Schreier and Werden (2000) reported that, in a meta-analytic review of 156 psychological, behavioral, and educational interventions, Lipsey and Wilson (1993) calculated a grand mean treatment effect of 0.47 (SD = 0.28). Grand mean scores for 64 analyses focusing on psychoeducational treatments and primary preventions ranged from 0.30 to 1.51, supporting the contention that programming and prevention efforts are effective service delivery alternatives. Although these studies did not review interventions and programs for lesbian and gay people and their families, the positive results reported can be extrapolated to encourage more research and the development of programming to address the needs and concerns of families who are struggling to understand and accept their lesbian or gay member. Much of the literature that specifically discusses the use of and efficacy of family psychoeducation and family interventions focuses on the families and support networks of individuals with chronic physical and mental illnesses. For example, Dixon, Adams, and Lucksted (2000) reviewed several studies on family psychoeducation and interventions for families of persons with schizophrenia. The authors found that the data supporting the efficacy of family psychoeducation were compelling. They further suggested that assessment of the appropriateness of family psychoeducation for a particular patient and family should consider 1) the interest of the family and patient; 2) the extent and quality of family and patient involvement; 3) the presence of patient outcomes that clinicians, family members, and patients can identify as goals; and 4) whether the patient and family would choose family psychoeducation instead of alternatives available in the agency to achieve outcomes identified. Schreier and Werden (2000) offered several suggestions for elements that should be included in programming targeting heterosexual as well as GLB communities, including nonheterosexist language (in discussions and examples), statements of inclusion (inclusion of race, gender, age, social class, religious affiliation, etc.), bibliographies, 65 advocacy-related information (related to city, state, federal, and institutional ordinances and laws concerning the civil rights of GLB people), displays of pride, movies, resources lists (including activities, support groups, social events, referrals, etc.), and attention to personal attitudes and beliefs. Schreier and Werden further suggested that programming related to GLB issues often is geared toward presenting information or engaging participants at a cognitive level (e.g., accessing beliefs and attitudes), but emotions affect the attitude change process by influencing what information is attended to, encoded, and retrieved. Thus, programming should go beyond the dissemination of information, exercises that are cognitively based, or exposure to GLB people. Accessing emotionality might involve (a) eliciting feelings of self-dissatisfaction concerning a given attitude or belief, (b) appealing to the commonality of emotional experiences and of natural curiosity, (c) presenting information that suggests that most positive emotional experiences result from adopting new attitudes or beliefs, and (d) demonstrating the emotional dysfunctionality of targeted attitudes or beliefs (Schreier & Werden). Schreier and Werden (2000) further suggested that measures of programming efficacy for psychoeducational programming around these particular issues and populations should achieve two primary goals. The first is validating the extent to which intended outcomes were obtained, and the second is identifying what elements of the program contributed to the achievement of intended outcomes. As noted above, PFLAG serves as both a self-help support group and an education and advocacy organization. The current study sought to explore parents? involvement in, experience of, and satisfaction with PFLAG resources and activities. Items assessed each participant?s level of involvement in various aspects of PFLAG, the participant?s satisfaction with her or his 66 PFLAG experiences, and how the participant?s involvement in PFLAG has influenced family and parent-child relationship functioning. Overall Summary and Statement of the Problem The current study was exploratory in nature and utilized a survey methodology to examine the relationships among heterosexual parents? gender role attitudes, religious orientation, heterosexist beliefs, involvement in PFLAG, and relationships with their gay or lesbian child. Several studies have explored associations among heterosexist attitudes, gender role attitudes, and religious orientation, and some research has begun to explore these factors in the context of parental reactions to a child?s coming out as gay or lesbian. More specifically, research has demonstrated that people?s attitudes toward homosexuality seem to have some foundation in a broader belief system about women, men, and their appropriate roles (Kite, 1994). Research consistently has found gender differences in attitudes toward homosexuality as well as a correlation between negative attitudes and people?s adherence to traditional values concerning gender and family. Higher levels of homophobia or heterosexism appear to be related to more traditional gender role beliefs (e.g., Herek, 1988; Kerns & Fine, 1994). In addition, a relationship between religious orientation and attitudes of intolerance has been identified consistently by researchers, particularly with regard to homophobic or heterosexist attitudes. In general, low scores on particular types of religious orientation indicate more tolerant attitudes toward same-sex orientations. More specifically, taking into account fundamentalist beliefs, intrinsic religion (religion that is the orienting center of one?s life and motivation) is directly related to discrimination against GLB individuals (e.g., McFarland, 1989; Fulton et al., 1999). Extrinsic religion 67 (religion used for ulterior motives such as security, comfort, status, or social support) has been found to be unrelated to discrimination against same-sex orientations; however, there are mixed results with regard to an Extrinsic-Social orientation. Kirkpatrick (1993) found that Extrinsic-Social religious orientation predicted less tolerant attitudes toward same-sex orientations. A third construct within religious orientation called quest (the degree to which religion involves an openness to change and a questioning approach to religion) has been found to be inversely related to discriminatory or prejudicial attitudes (Batson et al., 1986), specifically with regard to attitudes toward same-sex orientations (McFarland). One purpose of the current study was to explore these attitudinal variables as they relate to parents? responses to and relationships with their lesbian or gay children. Although the literature is scant, some research has demonstrated that parents with higher levels of homophobia have had knowledge of their child?s gay or lesbian sexuality for a relatively shorter period of time and endorse more traditional sex role stereotypes than relatively non-homophobic parents (Holtzen & Agresti, 1990). Cramer and Roach (1988) showed that parents? religious orthodoxy, in addition to other variables such as attitudes about sex roles, level of education, location of childhood, and authoritarianism, could significantly predict changes in their relationships with their gay sons following disclosure. In addition, many writers have proposed paradigms by which parents evolve through a series of stages from initial shock to eventual acceptance of their child, and the spectrum of parental reactions has been characterized as similar to those experienced by individuals undergoing grief and mourning (e.g., DeVine, 1984; Robinson et al., 1989). 68 Empirical research seldom has tested the validity of grieving models against the real-life reactions parents have on learning about the sexual-minority status of their child. In one summary of research on the family relationships of lesbians and gay men, Patterson (2000) reported that the most common initial parental reactions to disclosure are negative. Negative reactions are likely to be more pronounced among older parents, those with less education, and those whose parent-child relationships were troubled before the disclosure. Patterson noted that although the parent-child relationship often suffers difficulties after disclosure, the relationship most often improves again over time as families assimilate the new information into existing images of the lesbian or gay child. Moreover, the best predictor of post-disclosure relationships between lesbian and gay young adults and their parents is the quality of their relationships before the disclosure (Patterson). Other factors that have been proposed to affect parental reactions include family culture, the child?s age when the discovery occurs, race/ethnicity, parent and child gender, although none of these variables have been pursued systematically (Matthews & Lease, 2000; Savin-Williams, 2001), particularly from the perspective of the parent. In addition, much of the literature has focused on the consequences of coming out to family members, but relatively little has been documented that articulates the process by which parents learn to understand, accept, and integrate their lesbian or gay children. As noted in Broido (1999), research has focused more on the demographic correlates of relatively positive (or negative) attitudes than on the development of positive attitudes and behaviors. Another purpose of the current study was to explore parents? attitudes and relationships with their children (and families) in the context of support group 69 experiences. Some qualitative studies have indicated several important steps that family members, and parents in particular, take in changing their perceptions of and responses to their lesbian or gay family member or child. Some of these steps include accessing resources in the community that help build knowledge and awareness of homosexuality, talking with others who have shared their experience, becoming involved in the lesbian and gay community through social activities and/or advocacy, ?coming out? to other family members, friends, coworkers, and associates, adjusting their expectations surrounding the future of their lesbian or gay child, and modifying their vision of what constitutes a family (Beeler & DiProva, 1999; Herdt & Koff, 2000; Oswald, 2002). Much of the research on disclosure/coming out and parental reactions, and more specifically the literature that addresses more of the process by which individuals learn to understand, accept, and integrate their lesbian or gay family members, points to the potential importance of psychoeducation and social support resources in aiding parents to develop more positive and affirmative responses to their lesbian or gay child. Although efficacy research on programming designed to address issues faced by family members and parents of lesbian and gay individuals is nonexistent, research has indicated that programming and prevention efforts are effective service delivery alternatives (Lipsey & Wilson, 1993). Schreier and Werden (2000) offered several suggestions for elements that should be included in programming targeting heterosexual as well as GLB communities, such as disseminating information, offering exercises that are cognitively based, exposing individuals to the GLB community, and accessing emotionality through attention to personal attitudes and beliefs. 70 In order to begin to understand how people, and parents more specifically, experience the transition to becoming more accepting and supportive of their lesbian and gay family members, the current study sought to explore how parents? attitudes and relationships with their lesbian and gay children are related to their participation in a support group. This study specifically assessed a set of attitudinal variables (gender role attitudes, religious orientation, and heterosexist beliefs) in relation to relationship variables (current family and parent-child relationship functioning and changes in relationship functioning following disclosure and participation in a support group). These attitudinal variables and relationship variables then were explored in relation to a set of support group variables (overall involvement in a support group, satisfaction with involvement in a support group, and involvement in support, education, and advocacy activities). The overall purpose of this study, then, was to further understand the development of positive attitudes and relationships of parents toward their LGB children and what experiences might facilitate this process. Research Questions 1, 2, and 3 explored correlations among attitudinal variables (i.e., gender role attitudes, intrinsic/extrinsic religious orientation, quest religious orientation, and heterosexist attitudes), relationship variables (i.e., parent-child relationship functioning, family functioning, and parents? perception of changes in parent-child relationship functioning since their child?s disclosure), and support group variables (i.e., overall involvement in PFLAG, level of involvement in supportive, educational, and advocacy activities, and change in family/parent-child relationship functioning following involvement in PFLAG). Research Questions 4, 5, and 6 utilized 71 regression analyses to predict relationships among the attitudinal variables, support group variables, and family/parent-child relationship variables. Research Question 1: Are attitudinal variables (gender role attitudes, religious orientation, and heterosexist attitudes) related to relationship variables (parent-child relationship functioning, family functioning, and parents? perception of changes in parent-child relationship functioning since their child?s disclosure)? Research Question 2: Are parents? perception of changes in parent-child relationship functioning since their child?s disclosure related to parents? perception of changes in family and parent-child relationship functioning following their involvement in PFLAG? Research Question 3: Are attitudinal variables (gender role attitudes, religious orientation, and heterosexist attitudes) related to support group variables (parents? overall involvement in PFLAG and parents? level of involvement in supportive, educational, and advocacy activities)? Research Question 4: Do parents? involvement in support, education, and advocacy activities differentially influence heterosexist attitudes? Similarly, do parents? involvement in support, education, and advocacy activities differentially influence gender role attitudes? Research Question 5: Do attitudinal variables (gender role attitudes, religious orientation, and heterosexist attitudes) and support group variables (overall involvement and involvement in support, education, and advocacy) differentially influence parent- child relationship functioning? Do attitudinal variables and support group variables differentially influence family functioning? 72 Research Question 6: Do support group experiences (overall involvement and involvement in support, education, and advocacy) differentially influence parents? perception of changes in family and parent-child relationship functioning following their involvement in PFLAG? Do support group experiences differentially influence parents? perception of changes in parent-child relationship functioning since their child?s disclosure? 73 Chapter 3 Method Participants Utilizing recruitment procedures similar to those successfully used by Holtzen and Agresti (1990), participants were contacted through chapters of Parents and Friends of Lesbians and Gays (PFLAG) in 13 states and Washington, DC. Through a consultative relationship already established with the national PFLAG office in Washington, DC, a contact person was identified for each PFLAG chapter and sent information describing the study and materials for participation. These individuals generally were chapter leaders and were contacted first via email (see Appendix A) and second by a follow-up phone call from the primary investigator. Chapter leaders were then sent as many questionnaire packets as they requested, generally based on their membership and next meeting time. Chapter leaders were asked to distribute questionnaire packets to PFLAG parents who identified as exclusively heterosexual and who had one or more children who are 16 years of age or older and who have identified themselves as gay, lesbian, or same-sex oriented. In the cover letter the investigators acknowledged that there is a wide variation in sexual identity and sexual orientation and requested participation from parents with children who specifically have declared themselves as gay, lesbian, or attracted to persons of the same sex. A total of 382 packets were mailed out to chapter leaders, and 190 questionnaires were returned. Of those, 20 were blank, 2 were completed by parents who identified as lesbian, and 1 was completed by a parent who did not belong to a formally organized PFLAG chapter. Thus, the final sample for the current study consisted of 167 individuals. 74 It should be noted that hroughout data collection it became apparent that there was a significant underrepresentation of racial/ethnic minorities in the sample. Thus, during data collection there was a specific effort to recruit racial/ethnic minorities in order to increase generalizability and have a more inclusive sample. The primary investigator sought participation from members and chapter leaders in specific PFLAG chapters that were suggested to have a more culturally diverse membership. Participants in the current study were 110 females and 52 males who identified as heterosexual parents with one or more same-sex oriented children (participants who identified as lesbian or gay were excluded). Demographic information about the sample is provided in Table 1. ------------------------------------------ Insert Table 1 here ------------------------------------------ Participants? age ranged from 40 to over 80 years of age. In terms of race/ethnicity, the sample identified predominantly as Caucasian/Euroamerican (n = 157) but included some participants who identified as Asian American/Pacific Islander (n = 3), African American /Black (n = 2), Native American/American Indian (n = 2), Hispanic (n = 1), and Other (n = 1). Individuals participated from a wide variety of geographic locations within the United States, including Florida (17%), New York (14%), North Carolina (12%), Illinois (10%), Virginia (9%), Pennsylvania (7%), New Jersey (7%), Connecticut (7%), Texas (6%), Rhode Island (5%), Maryland (4%), California (1%), Oklahoma (1%), and Washington D.C. (1%). Most were from suburban settings (63%) rather than urban (26%) and rural (11%) settings. Individuals were highly educated (39% had a graduate or professional degree and 38% had an undergraduate degree) and were equally employed 75 Table 1 Demographic Characteristics of Participants (N = 167) Characteristic n % Gender Female 110 68 Male 52 32 Race Caucasian 157 94 Asian 3 2 Black 2 1 Native American 2 1 Hispanic 1 1 Other 1 1 Age 40-50 11 7 51-60 64 40 61-70 50 32 71-80 23 15 Over 80 10 6 Highest Education Level Less than High School 1 1 High School 9 5 Technical/Vocational 4 2 Some College 25 15 Undergraduate Degree (e.g., BA, AA) 63 38 Graduate or Professional Degree 65 39 Employment Status Employed 75 45 Retired 75 45 76 Unemployed 15 10 Work Status Full Time 57 77 Part Time 17 23 Marital Status Married 142 85 Divorced 15 9 Widowed 8 5 Single 1 1 Household Income 0-9,999 1 1 10,000-19,999 2 1 20,000-29,999 8 5 30,000-39,999 14 9 40,000-49,999 15 10 50,000-59,999 18 11 60,000-69,999 14 9 70,000-79,999 15 10 80,000-89,999 9 6 90,000 and over 60 38 Location Suburban 104 63 Urban 43 26 Rural 18 11 State of Residence Florida 27 17 New York 22 14 North Carolina 19 12 Illinois 16 10 Virginia 14 9 Pennsylvania 12 7 77 New Jersey 12 7 Connecticut 11 7 Texas 9 6 Rhode Island 8 5 Maryland 7 4 California 2 1 Oklahoma 2 1 Washington D.C. 1 1 Religious Affiliation Protestant 52 31 Jewish 32 19 Catholic 24 15 Agnostic/Atheist 8 5 Unitarian 9 5 Buddhist 1 1 Christian Scientist 1 1 NA 5 3 Other 34 20 Frequency of attendance at place of worship Not at all 26 16 More than once per week 15 9 Once per week 53 32 Several times per month 15 9 Once per month 6 4 Several times per year 27 17 Once every 1-3 years 10 6 Other 11 7 Do you have more than one child who has come out to you? Yes 23 14 No 142 86 78 Do you have other family members who identify as LGBT? No 96 67 Yes 47 33 How long have you been participating in PFLAG? Less than 1 month 1 1 1-6 months 6 4 7-12 months 6 4 1-2 years 9 5 2-3 years 18 11 3-5 years 29 18 5-10 years 42 25 More than 10 years 55 33 How often do you attend PFLAG meetings? More than once per month 15 9 Once per month 105 64 Once every two months 8 5 Once every three months 10 6 Once or twice per year 14 9 Once every year or two 2 1 Other 9 6 79 (45%) or retired (45%) while some were unemployed (10%). Of those who were employed, 77% indicated that they work full time and 23% indicated that they work part time. The majority of participants were married (85%) while some were divorced (9%), widowed (5%), and single (1%). Participants? household income was predominantly $90,000 and over (38%), although 49% of participants had indicated a household income that fell between $30,000 and $79,999. Participants varied in religious affiliation including Protestant (31%), Jewish (19%), Catholic (15%), Agnostic/Atheist (5%), Unitarian (5%), Buddhist (1%), Christian Scientist (1%), NA (3%), and Other (20%). Nine percent of participants indicated that they attend a place of worship more than once per week while 32% indicated once per week, 9% indicated several times per month, 4% indicated once per month, 17% indicated several times per year, 6% indicated once every 1-3 years, and 16% indicated not at all. In terms of PFLAG participation, most participants had been participating in PFLAG for several years: more than 10 years (33%), 5-10 years (25%), 3-5 years (18%), 2-3 years (11%), 1-2 years (5%), 7-12 months (4%), 1-6 months (4%), and less than 1 month (1%). The majority of participants indicated that they attend PFLAG meetings once per month (64%), while other participants indicated that they attend meetings more than once per month (9%), once every two months (5%), once every three months (6%), once or twice per year (9%), once every year or two (1%), and other (6%). Some participants indicated that they have more than one child who has come out to them (14%), and some indicated that they have other family members who identify as LGBT (33%). Parents in this study also indicated demographic characteristics of their children. Demographic information about participants? children is provided in Table 2. 80 ------------------------------------------ Insert Table 2 here ------------------------------------------ For the first child that had come out to participants, 77% were male and 23% were female. In terms of sexual orientation, participants indicated that their children identified as gay (76%), lesbian (20%), bisexual (2%), and other (2%). Participants? children varied in age, with 7% being 16-20, 32% being 21-30, 34% being 31-40, 24% being 41-50, and 3% being over 50. The majority of parents indicated that their child had been out to them for more than three years (89%), and the majority also indicated that their child did not currently live at home (89%). Twenty-two parents had a second child who had come out to them. For the second child that had come out to participants, 46% were male and 54% were female. In terms of sexual orientation, participants indicated that their children identified as gay (46%), lesbian (36%), bisexual (9%), and other (9%). The children varied in age, with 4% being16-20, 23% being 21-30, 36% being 31-40, 23% being 41-50, and 14% being over 50. Again, the majority of parents indicated that their child had been out to them for more than three years (86%), and the majority indicated that their child did not currently live at home (95%). Instruments Demographic Questionnaire (see Appendix B) The following demographic information was obtained from a demographic questionnaire developed for this study: age, gender, race/ethnicity, level of education, employment status, annual income, occupation, marital/spousal status, geographic location, religious affiliation, sexual orientation, and other GLBT-identified family 81 Table 2 Demographic Characteristics of Children Characteristic n % Child 1 (n = 167) Gender Male 129 77 Female 38 23 Age 16-20 11 7 21-30 54 32 31-40 55 34 41-50 39 24 Over 50 6 3 Sexual Orientation Gay 127 76 Lesbian 33 20 Bisexual 4 2 Other 3 2 Length of time since disclosure 4-9 months 2 1 More than 1 year 5 3 More than 2 years 11 7 More than 3 years 149 89 Does your child currently live at home? No 141 89 Yes 17 11 Child 2 (n = 22) Gender Male 10 46 Female 12 54 82 Age 16-20 1 4 21-30 5 23 31-40 8 36 41-50 5 23 Over 50 3 14 Sexual Orientation Gay 10 46 Lesbian 8 36 Bisexual 2 9 Other 2 9 Length of time since disclosure 4-9 months 2 9 More than 2 years 1 5 More than 3 years 19 86 Does your child currently live at home? No 19 95 Yes 1 5 83 members. For each child who has come out (if more than one), participants were asked to indicate the child?s gender, age, sexual identity, and length of time since the child?s disclosure of her/his same-sex identity, as well as whether or not the child currently lives at home. PFLAG Involvement (see Appendix C) Information related to parents? PFLAG chapter and involvement in PFLAG was assessed through a series of items developed by the principal investigator. These items assessed the participant?s level of overall involvement in PFLAG (i.e., attendance and length of time in the chapter), the participant?s level of participation in supportive (e.g., support group discussion, social events), educational (e.g., speaker, reading information), and advocacy (e.g., participating in marches, organizing talks) activities, the chapter?s level of emphasis on supportive, educational, and advocacy activities, and the participant?s level of satisfaction with the various resources offered by her/his chapter. This section also included open-ended questions regarding what resource/activity/publication had been most and least helpful to the participant, what the participant would like to see PFLAG do to better support parents, and what led the participant to involvement in PFLAG. It should be noted that, for the purposes of the current study, data from these open-ended questions were collected and analyzed but will only be reported to PFLAG. Construction of all items for the PFLAG Involvement section was based on a review of literature distributed by the national PFLAG office as well as individual PFLAG chapters. Items also were constructed in consultation with several chapter leaders for PFLAG chapters in Florida, New York, and Rhode Island. Scores for participants? level of overall involvement in PFLAG were calculated by 84 weighting the duration of their participation in PFLAG with their frequency of attendance at PFLAG meetings/activities. Thus, scores were calculated first by converting both duration and attendance into years, and then by multiplying the participant?s response for length of time in his/her chapter and the participant?s response for level of attendance. Overall involvement scores ranged from 0 (no involvement) to 240 (a great deal of involvement). Mean scores for participants? level of participation in support, education, and advocacy each ranged from 1 (no participation) to 7 (a great deal of participation). Mean scores for participants? level of satisfaction with PFLAG resources ranged from 1 (very dissatisfied) to 7 (very satisfied). Heterosexist Beliefs (see Appendix D) Heterosexist attitudes were assessed using a newly developed measure called the Heterosexism Scale (Park, in progress). This measure was developed with the goal of measuring attitudes toward GLB individuals within the context of heterosexism. In other words, the goal was to measure less obvious forms of bias and prejudice toward GLB individuals and tap into the ambivalence exhibited even by individuals who consider themselves to have positive or affirming attitudes towards GLB individuals. Confirmatory factor analyses indicated a model that included two factors: superiority of heterosexuality and tolerance/acceptance of non-heterosexuality. The final Heterosexism Scale consists of 17 items and was rated on a 5-point Likert scale (1=strongly disagree to 5=strongly agree). Mean scores for the scale ranged from 1 to 5, with higher scores indicating more intolerant or heterosexist attitudes. Sample items include ?Positive aspects of various sexual orientations should be included in public education;? ?Heterosexual couples make better candidates for parents than do same-sex couples for 85 adoption;? and ?I would not think less of my co-worker if I found out that he or she was a lesbian, gay, or bisexual individual.? For the purposes of the current study, one item was eliminated from the Heterosexism Scale (?My relationship with my son or daughter would remain the same even if I found out that he or she was romantically involved with a person of the same sex?) given the nature of the study?s sample. Park (in progress) conducted five studies designed to establish the psychometric properties of the Heterosexism Scale, and, collectively, the results of the five studies established that the Heterosexism Scale is both reliable and valid. In a sample of 121 undergraduate and graduate students in a large northeastern university (94.2% was younger than 26 years of age; 74.4% female; 83.5% Caucasian; 96.7% heterosexual; and 81% Christian), coefficient alpha of the total scale was .95. In addition, most of the items of the Heterosexism Scale did not correlate significantly with a measure of social desirability, and those that did accounted for a small portion of the variance. In a separate sample of participants recruited from undergraduate and graduate classes at a large northeastern university (N = 304; 55.6% between 21 and 25 years of age; mean age of 22.2 years [SD = 4.66]; 72.4% female; 88.2% Caucasian; 97% heterosexual; 70.1% Christian; 42.8% college seniors; and 76% who reported some interactions with GLB individuals), construct validity was established by examining the relationship between heterosexist attitudes and theoretically relevant constructs documented in the literature (i.e., attitudes towards ethnic minorities, attitudes towards women, authoritarian attitudes). As predicted, total scores of the Heterosexism Scale were associated with biases towards ethnic minorities and women and authoritarian attitudes. Also, the Heterosexism Scale as well as the scale?s two factors were significantly correlated with 86 the Modern Homophobia Scale (MHS; Raja & Stokes, 1998) and its subscales (MHS-L and MHS-G). Thus, a higher level of heterosexist attitudes was significantly correlated with a higher level of homophobic attitudes. Gender Role Attitudes (see Appendix E) Attitudes toward gender roles and stereotypes were measured, in part, by the Sex Role Egalitarianism Scale-Short Form (SRES; King & King, 1997), a scale developed to measure attitudes toward the equality of women and men in five domains: marital roles (addressing beliefs about men and women in their spousal roles), parental roles (addressing beliefs about maternal and paternal roles), employment roles (addressing beliefs about men and women in their workplace roles), socio-interpersonal-heterosexual roles (addressing beliefs about men and women in a variety of social relationships), and educational roles (addressing beliefs about educational alternatives available to men and women). Sex-role egalitarianism was defined as ?an attitude that causes one to respond to another individual independently of the other individual?s sex. One who possesses this attitude believes that the sex of an individual should not influence the perception of an individual?s abilities or the determination of an individual?s rights, obligations, and opportunities? (Beere, King, Beere, & King, 1984, p. 564). Correlational and factor analytic evidence across several studies indicated a single dimension of sex-role egalitarianism in the SRES. Items are scored on a 5-point Likert-type scale, ranging from 1 (strongly agree) to 5 (strongly disagree). Participants? mean scores for the instrument ranged from 1 to 5, with higher scores indicating more egalitarian attitudes (King & King, 1993). Sample items are ?Keeping track of a child?s activities should be mostly the 87 mother?s task,? ?When both husband and wife work outside the home, housework should be equally shared,? and ?An applicant?s sex should be important in job screening.? There are four forms of the SRES, two full forms and two short forms. The current study utilized the SRES-Short Form KK, which contains 25 items selected from those appearing on its counterpart full form, with an equal representation of five items for each of the five SRES domains. Correlations between the SRES full forms and short forms have ranged from .83 to .95, thus indicating that the abbreviated versions appear to satisfactorily represent that which is measured by their full forms (King & King, 1993). Results from various empirical studies support the psychometric integrity of the SRES. In one study of college faculty, administrative staff, and students at a large Midwestern university (N = 98), an internal consistency reliability estimate of .95 was found for the SRES-Short Form KK (Billingham & King, 1991). Another study of several employee groups also in a large Midwestern university (N = 440) reported an alpha coefficient of .92 for the SRES-Short Form KK (Mattimore, King, & King, 1991). In terms of convergent and discriminant validity evidence, several studies have confirmed expected relationships between scores on the SRES and scores on other measures that attempt to detect individual differences along a traditional/nontraditional sex-role attitudinal dimension (King & King, 1993). For example, King, Beere, King, and Beere (1984) computed a correlation of .81 between the widely-used Attitudes Toward Women Scale (AWS; Spence & Helmreich, 1972) and SRES total scores for a sample of 479 college students, with AWS-SRES domain score correlations ranging from .69 to .78. As reported in King and King (1993), Honeck?s (1981) study of sex-role attitudes of both college and high school students also yielded very similar AWS-SRES correlational 88 values. Finally, support for discriminant validity comes from research efforts to determine the relationships between SRES scores and scores on self-report indices of gender traits assessed by the Bem Sex Role Inventory (BSRI; Bem, 1974). One study by King and King (1990c) correlated a sample of students? SRES scores with androgyny scores computed both as the simple Masculinity-Femininity difference and as the continuous androgyny score developed by Kalin (1979). The four resulting correlation coefficients were quite low (r = .07-.15) and nonsignificant. Gender role attitudes also were measured by eight items related to sexual initiative and casual sex. These eight items are from two subscales of the Gender Attitudes Inventory (GAI; Ashmore, Del Boca, & Bilder, 1995), a structured inventory that assesses attitudes toward a broad range of gender topics relative to females, with parallel items regarding attitudes toward male roles constructed by the primary investigator (see Appendix C). The subscales used in the current study include Female Casual Sex and Female Sexual Initiative. From the perspective that gender attitudes are not unidimensional, Ashmore et al. articulated an intergroup relations model in which gender attitudes or ideologies operate at several different, though interconnected, levels. These levels pertain to gender-stereotypic beliefs about (a) individuals, (b) male-female relations, (c) culturally defined roles, and (d) societal issues. The eight items utilized in this study were rated on a 7-point Likert scale (1 = strongly disagree to 7 = strongly agree). For these eight items, mean scores could range from 1 to 7, with higher scores indicating more traditional attitudes. Sample items included: ?The initiative in asking for a date should come either from the man or the woman? (GAI) and ?I would have no 89 respect for a man who engages in sexual relationships without any emotional involvement? (author-generated). In a study of 126 introductory psychology students (86 females, 40 males; 69% White, 5% Black, 5% Hispanic, 13% Asian-American, and 7% ?other?), Ashmore et al. (1995) sought to validate the GAI using several existing standardized measures (i.e., global sex-role ideology, liberalism vs. conservatism, and self-deception and impression management in self-reports). Also, connections between respondents? attitudes on specific topics as assessed by the GAI scales and parallel measures tapping attitudes with similar contents were tested. Alpha values for both the Female Sexual Initiative subscale and the Female Casual Sex subscale ranged from .75 to .86, and Ashmore et al. reported that, in general, sex differences in alphas (including test-retest correlations) were modest. Although males reported more traditional attitudes than their female counterparts on 10 of the 14 subscales, the scale for which male attitudes were most modern relative to females? views was Female Casual Sex. Males also exhibited less traditional attitudes for the Female Sexual Initiative subscale, although the sex differences were not large or statistically significant. In terms of validity, results indicated that, as hypothesized, the GAI scales were consistently correlated with the Attitudes Toward Women Scale. In addition, the GAI scales showed moderate positive correlations with conservatism (further indicating convergent validity) and small correlations with a measure of social desirability responding (indicating discriminant validity). Finally, the pattern of covariation among the GAI scales and questionnaires about various social attitudes and issues further supported the convergent and discriminant validity of the GAI. Other studies have found 90 satisfactory psychometric properties for the GAI. In another study, Jackson, Fleury, and Lewandowski (1996) reported adequate reliability among the GAI subscales (U = .70 to .83) in a study of 362 undergraduates? (247 females, 115 males; 92% White/Caucasian; approximately 21 years old) definitions of feminism, support for feminism, and gender attitude correlates of support. Jackson et al. also reported that gender attitudes (i.e., more egalitarian attitudes) were related to support for feminism in the expected directions. Religious Orientation (see Appendix F) Religious orientation was measured, in part, by the Age-Universal I/E Scale- Revised (Maltby & Lewis, 1996). The revised Age-Universal I/E Scale (Maltby & Lewis, 1996) was adapted from Gorsuch and Venable?s (1983) revision of Allport and Ross? (1967) Religious Orientation Scale. It is a 19-item measure with two separate scales including 8 intrinsic items and 11 extrinsic items. Items are scored on a 3-point scale: ?yes? (3), ?not certain? (2), and ?no? (1). It is used to measure an intrinsic orientation to religion (using religion as the orienting center of one?s life and motivation orientations toward religion) and an extrinsic orientation to religion (using religion for personal motives such as security, comfort, status, or social support). Typically, scores are calculated by summing across items. Thus, scores on the Intrinsic Religion Scale range from 7 to 21, with higher scores indicating a stronger intrinsic orientation to religion. Scores on the Extrinsic Religion Scale range from 12 to 36, with higher scores indicating a stronger extrinsic orientation to religion (Maltby & Lewis). For the purposes of standardization and because the items are treated as continuous variables, scores range from 1 (indicating less intrinsic or extrinsic orientation) to 3 (indicating a stronger intrinsic or extrinsic orientation). A sample intrinsic item is ?I try to live all my life 91 according to my religious beliefs,? and a sample extrinsic item is ?I go to my place of worship mostly to spend time with my friends.? For the current study, minor modifications were made to the wording of some items in an effort to extend the relevance of the scale to a broader range of religious backgrounds. Thus, the word ?church? was changed to ?my place of worship? and item 11 was changed from ?I would rather join a Bible study group than a church social group? to ?I would rather join a group that studies the sacred book(s) of my religion than attend a social group in my place of worship.? The Age-Universal Intrinsic/Extrinsic Orientation Scale is noted to be the ?instrument of choice in this area of research? (Hall, Tisdale, & Brokaw, 1994, p. 396). Historically, however, the original scale lacked validity when administered to non- religious participants (Maltby & Lewis, 1996). Maltby and Lewis revised the I/E scales to make them applicable for all possible respondents, religious and non-religious. The authors altered the directions and removed the agree-disagree response format. Maltby and Lewis conducted a study to explore the psychometric properties of the I/E scales after revisions, and the authors utilized six samples of adults: one community sample in North Carolina (N = 156; 70 males, 86 females); one university student sample in Ohio (N = 144; 43 males, 101 females); one sample of high school students in England (N = 135; 75 males, 60 females); one adult sample in England (N = 149; 65 males, 84 females); one adult sample in Northern Ireland (N = 189; 74 males, 115 females); and one adult sample in the Republic of Ireland (N = 167; 70 males, 97 females). Participants in all six samples were predominantly Protestant and Catholic. Alpha coefficients for the Intrinsic and Extrinsic scales ranged from .87-.91 and .82-.90, respectively. To demonstrate the 92 construct validity of the original instrument, intrinsicness has been found to correlate positively with an internal locus of control (Kahoe, 1974) and conceptualization of God as an important causal agent (Watson, Morris, & Hood, 1990). Positive correlates of extrinsicness include perceived powerlessness (Minton & Spilka, 1976), trait anxiety (Baker & Gorsuch, 1982), and depression (Genia & Shaw, 1991). Religious orientation also was measured by the Age-Universal Quest Scale-Revised (Maltby & Day, 1998). The revised Age Universal Quest Scale was adapted from Batson and Schreonrade?s (1991) Quest Scale measuring ?an open-ended, responsive dialogue with existential questions raised by the contradictions and tragedies of life? (Batson et al., 1993, p. 169). Quest orientation includes three factors. First is the ability to explore existential questions without reducing their complexity. Second is the perception of self- criticism and religious doubt as positive. Third is a tentativeness or openness regarding religious beliefs (Batson et al.). The original Age Universal Quest Scale consists of 12 items organized into three 4-item subscales (Complexity, Doubt, and Tentativeness). Each item is scored on a 9-point scale ranging from ?strongly disagree? (1) to ?strongly agree? (9). Like the revised version of the Age Universal I/E scale, the revised version of the Quest scale has been shown to be more effective than the original version of the scale among samples that contain non-religious persons. Two changes were made for the revised version. First, as for the I/E scale, Maltby and Day changed the instructions. Second, responses were scored on a 3-point scale that reads: ?yes? (3), ?not certain? (2), and ?no? (1). Possible scores on the Quest scale range from 12 to 36 with higher scores indicating a stronger orientation to quest. In a sample of 481 undergraduate students (252 males, 229 females), internal consistency for the amended scale was found to be adequate 93 (U = .79). Also, a scree test showed greater support for the three component structure in the amended version (Complexity, Doubt, and Tentativeness) than was seen in the original version (Maltby & Day, 1998). In terms of the scale?s validity, Quest has been found to have low correlations with both extrinsic and intrinsic religion (Batson & Schoenrade). Quest has been negatively correlated with religious orthodoxy (Batson & Ventis, 1982) and positively associated with, but not the same as, religious conflict and anxiety (Kojentin, 1988). For this study, the total score of the instrument was used and subscales were not analyzed separately. Family Functioning and the Parent-Child Relationship (see Appendix G) General family functioning and the parent-child relationship were assessed along two dimensions. The first dimension was the parent?s perception of current family and parent- child relationship functioning. The second dimension was the parent?s perception of changes in family and parent-child relationship functioning related to the child?s disclosure and the parent?s involvement in PFLAG. The parent?s perception of current family functioning was assessed, in part, by the General Functioning subscale of the McMaster Family Assessment Device (FAD; Epstein, Baldwin, & Bishop, 1983). Overall, the FAD assesses six dimensions of family functioning: Problem Solving (the family?s ability to resolve issues which threaten the integrity and functional capacity of the family); Communication (the exchange of information among family members); Roles (established patterns of behavior for handling family functions); Affective Responsiveness (the extent to which individual family members are able to experience appropriate affect over a range of stimuli); Affective Involvement (the extent to which family members are interested in and place value on each other?s activities and concerns); 94 Behavior Control (the way in which a family expresses and maintains standards for the behavior of its members); and General Functioning (the overall health/pathology of the family). For the General Functioning subscale, 12 items are rated on a 4-point Likert scale ranging from strongly disagree (1) to strongly agree (4). Higher scores reflect unhealthy functioning, as indicated by stronger endorsement of difficulty communicating, supporting each other, expressing feelings, decision-making, and getting along. In a sample of 503 individuals (294 individuals from 112 families and 209 students in an introductory psychology course), Epstein et al. reported an alpha coefficient of .92 for the General Functioning subscale. In another study with a sample of 45 hospital employees (67% female, mean age of 32.9, equally split between father, mother and child roles), Miller, Epstein, Bishop, and Keitner (1985) reported a test-retest reliability of .71 for the General Functioning subscale. Concurrent validity for the General Functioning subscale was observed when correlations between the subscale and two other well-known family assessment measures corresponded closely with theoretical predications and expectations (Miller et al., 1985). Participants? level of satisfaction with current family functioning was also assessed through five items developed for the present study. For this section participants were instructed to rate their level of satisfaction on a scale of 1 (extremely dissatisfied) to 5 (extremely satisfied) in the broad areas of communication, level of affection, level of involvement, amount of support, and problem solving. Mean scores for satisfaction with current family functioning ranged from 1 to 5, with higher scores indicating higher satisfaction with family. Current parent-child relationship functioning was assessed using items developed by Aquilino (1995) for a study comparing parents? and young adult children?s reports of 95 relationship quality. For the current study, five broad items measuring humor/closeness, tension, level of involvement and interest in child?s life, fights/arguments, and overall relationship were selected from a number of other items which more specifically assessed these areas of relationship functioning. The first three items are scored on a 5-point Likert scale, ranging from 1 (strongly disagree) to 5 (strongly agree). Consistent with the original scoring formats, the fourth item is scored on a 4-point scale, ranging from 1 (not at all) to 4 (very frequently), and the fifth item assessing overall relationship quality is scored on a 10-point scale, ranging from 1 (really bad) to 10 (absolutely perfect). For all five items, after reverse coding, higher ratings on each item indicated more positive relationship quality. No reliability information is available for these items given that they were selected from a number of other items which more specifically assessed humor/closeness, tension, level of involvement, and fights/arguments. The current study attempted to address possible ethnic and cultural differences among families by utilizing broad assessments of family and relationship functioning (based on previously established measures and items as described here) and instructing participants to respond to each item according to how their particular family or parent- child relationship operates. For example, Aquilino?s (1995) Control-Conflict subscale (which includes overall level of involvement and fights/arguments) includes more specific questions related to parental disapproval of his/her child?s academic performance, dating partner, and occupational choice. As noted above, the current study assessed more broadly the constructs of humor/closeness, tension, level of involvement, and fights/arguments with select items from Aquilino (1995) as well as broad notions of 96 communication, level of affection, level of involvement and interest, amount of support, and problem solving as assessed in Epstein et al.?s (1983) General Functioning subscale. The final three sections of the family and parent-child relationship functioning questionnaire were developed by the principal investigator for the present study. The first section included five items assessing parents? perception of change in their parent-child relationship following their child?s disclosure of his/her same-sex orientation. Responses range from ?has changed for the worse? (1) to ?has not changed at all? (3) to ?has changed for the better? (5). Thus, low scores indicate negative change and high scores indicate positive change. The next section included five items assessing parents? perception of change in their general family functioning following their involvement in PFLAG, and the final section included five items assessing parents? perception of change in their parent-child relationship following their involvement in PFLAG. For both sections, responses range from ?no influence? (1) to ?very strong influence? (5). Thus, higher scores indicate a stronger relationship between PFLAG involvement and family or parent-child relationship quality. Procedures A packet containing a cover letter (see Appendix H), the demographic questionnaire, the instruments, and a self-addressed, stamped envelope was mailed to each person agreeing to take part in the study. To minimize possible response bias, instruments were randomly ordered in the packets. After completing the demographic questionnaire, participants were asked to complete the remainder of the questionnaire packet based on their experiences with any one of their lesbian or gay children if they had more than one child who had come out. The participants were instructed to return the 97 completed questionnaires, along with the informed consent form, back to the primary researcher in the self-addressed stamped envelope. Participants also were instructed not to complete one set of questionnaires for both parents within the same family. Instead, for those participants who did come from the same family, the cover letter instructed participants to indicate a common ID number on their individual return envelopes (i.e., the last four digits of their telephone number). In addition, all participants were informed that responses are completely anonymous. No identifying information (e.g., participant name, address, phone number) was collected, and there was no way for responses to be connected to the participants. 98 Chapter 4 Results Several research questions were analyzed to determine relationships among attitudinal variables, family and parent-child relationship variables, and support group variables. Prior to analyses, the main variables in the study were examined for skewness and kurtosis in order to assess the distribution of the data. Skewness and kurtosis values for all of the variables were deemed acceptable and within the guidelines suggested for data screening (Lomax, 1998), except for general family functioning (kurtosis = 3.3) and parent-child relationship functioning (skewness = -2.3, kurtosis = 6.1). Thus, the distributions for these variables are significantly skewed and asymmetrical, and results should be interpreted with caution as the assumption of normality is not met for these variables. In addition, it should be noted that bivariate correlations among several of the predictor variables in the regression analyses conducted for this study were significant. Thus, judgments about the relative importance of each predictor for the regression models should be made with caution. Due to the exploratory nature of this study, a pilot study and several preliminary analyses were conducted first. Results from the pilot study, which explored the psychometric properties of two gender role attitude measures, will be discussed first. Pilot Study In an effort to choose the most appropriate gender role attitude measure from among two forms of the Sex Role Egalitarianism Scale-Short Form (SRES; King & King, 1997) and various subscales of the Gender Attitudes Inventory (GAI; Ashmore, Del Boca, & Bilder, 1995), a pilot study was conducted prior to the main study. The purpose of the 99 pilot study was twofold: (a) to compare the reliabilities between the two SRES short forms, and (b) to compare the reliabilities and means between the SRES and the GAI. These two instruments were compared not only for reliability and mean differences but also for participants? comments, reactions, and feedback after completing the instruments. Participants in the pilot study were 10 mothers and 6 fathers, and all identified as Caucasian/Euroamerican except one participant who identified as Hispanic. Participants were recruited through the Washington, DC PFLAG chapter and were mailed questionnaire packets as described above. Half of the participants (n = 9) were given Questionnaire A, which included all of the measures in the current study, in addition to Form B of the SRES, four subscales of the GAI (i.e., attitudes toward traditional stereotypes, family roles, differential work roles, and female political leadership), and eight items related to sexual initiative and casual sex, some of which come directly from GAI subscales and some of which are parallel items regarding attitudes toward male roles constructed by the primary investigator. The other half of the participants (n = 7) were given Questionnaire B, which also included all of the measures in the current study, in addition to Form K of the SRES and all of the same subscales and items from the GAI as in Questionnaire A. Participants in the pilot study had high scores on both the SRES Form K and SRES Form B. Participants also had high scores on the GAI and moderate scores on the GAI casual sex and dating subscale. Thus, participants had stronger egalitarian than non- egalitarian gender role attitudes according to the SRES, more nontraditional than traditional gender role attitudes according to the GAI, and neutral attitudes toward women?s and men?s roles in sex and dating, as indicated by their predominant 100 endorsement of neither disagreeing nor agreeing with statements about women?s and men?s roles in sex and dating. Results also indicated that all measures showed adequate internal consistency, with the SRES Form B having a higher reliability (U = .87) than Form K (U = .80), and the full scale GAI having a slightly higher reliability than the SRES (U = .89). T-tests revealed that mean scores on the SRES Form K and Form B were not statistically different, and mean scores on the SRES and GAI were significantly different from each other (t = -5.17, p = .002). T-tests also revealed that the SRES and GAI without the sex and dating items were significantly different from each other (t = - 6.358, p = .001), and the SRES and GAI sex and dating subscale were also significantly different from each other (t = -6.134, p = .001). The SRES Form K was chosen as a measure of gender role attitudes for the primary study for several reasons. First, the SRES is significantly shorter in length than the GAI and, thus, minimized the total amount of time participants took to complete the questionnaire. Second, the SRES, compared to the GAI, is a more well-established measure that has been widely used and validated on diverse samples, including students, adults, and various groups. Finally, both forms of the SRES were not statistically different from each other, and the SRES Form K showed adequate reliability. Although the SRES Form B and the GAI showed higher reliabilities than the Form K, the majority of respondents? written feedback regarding their experience taking all of the measures indicated that Form K was easier to complete, easier to understand, and appeared conceptually clearer. Thus, the SRES Form K seemed more applicable to this population in terms of the various parental and relationship role items included in it, and, subsequently, showed adequate content and face validity in addition to its reliability. 101 Preliminary Analyses Table 3 reflects the means, standard deviations, and internal consistency estimates for the primary variables in the study. ------------------------------------------ Insert Table 3 here ------------------------------------------ The internal consistency reliabilities of the scales measuring general family functioning, satisfaction with family functioning, parent-child relationship functioning, change in the parent-child relationship after disclosure, influence of PFLAG involvement on the family, influence of PFLAG involvement on the parent-child relationship, heterosexist attitudes, gender role attitudes, attitudes toward sex and dating, intrinsic religious orientation, extrinsic religious orientation, extrinsic-social religious orientation, and quest orientation were estimated by calculating Cronbach coefficients. All of the measures showed adequate internal consistency except for the Extrinsic Religion subscale (alpha = .53). Upon examination of items and reliability estimates for the parent-child relationship scale, one item was eliminated from the scale. As reflected in Table 3, on the whole, participants tended to view their current family functioning as healthy and reported high levels of satisfaction. Participants also viewed their current parent-child relationship as positive, and they reported positive changes in their relationships with their children following their child?s disclosure of same-sex orientation. Participants indicated that their involvement in PFLAG had had minor to some influence on both their family and parent-child relationships. In terms of involvement, participants reported that they had engaged in a moderate to a great deal of participation in supportive activities, a moderate to a great deal of participation in 102 Table 3 Means, Standard Deviations, and Reliability Coefficients for Study Variables Variable n M SD Range U Perceptions of Family Functioning 163 1.54 .50 1-4 .89 Satisfaction with Family Functioning 163 4.13 .74 1-5 .92 Parent-Child Relationship 167 4.54 .54 2-5 .79 Change after Disclosure 164 3.92 .75 1-5 .83 Influence of PFLAG Involvement 156 2.86 1.15 1-5 .93 on Family Influence of PFLAG Involvement 158 2.86 1.25 1-5 .92 on Parent-Child Relationship Heterosexist Attitudes 165 1.30 .33 1-3 .71 Gender Role Attitudes 167 4.66 .34 3-5 .85 Sex/Dating Attitudes 165 4.53 1.02 2-7 .77 Intrinsic Religious Orientation 167 2.24 .57 1-3 .80 Extrinsic Religious Orientation 166 1.63 .33 1-3 .53 Quest Religious Orientation 164 2.13 .45 1-3 .79 Religious Importance 167 3.43 1.33 1-5 Religious Conservatism 167 3.49 .68 1-4 Involvement in Support 164 5.80 1.46 1-7 Involvement in Education 166 5.57 1.43 1-7 Involvement in Advocacy 166 4.60 2.02 1-7 Overall Involvement 162 69.73 60.65 0-240 Chapter Emphasis on Support 162 6.56 .79 3-7 Chapter Emphasis on Education 162 5.98 1.10 2-7 Chapter Emphasis on Advocacy 161 5.73 1.30 2-7 Satisfaction with Support 161 5.91 1.25 1-7 Satisfaction with Education 158 5.53 1.28 1-7 Satisfaction with Advocacy 158 5.22 1.55 1-7 103 educational activities, and a moderate amount of participation in advocacy activities. On the whole, participants tended to be moderately involved, indicating that most participants had been involved in PFLAG for several years and they tended to attend PFLAG meetings approximately once per month. In terms of attitudes, participants tended to endorse more tolerant and less heterosexist attitudes, more egalitarian gender role attitudes, and neutral (neither traditional nor liberal) attitudes toward sex and dating between men and women. Participants viewed religion as moderately to very important in their life, and they tended to view themselves as more religiously liberal than conservative. They also tended to view intrinsically religious statements as more applicable to themselves than extrinsically religious statements. Participants appeared relatively uncertain about their quest orientation to religion. Group differences. Chi-square tests, t-tests, and ANOVA were used to determine whether there were sample subgroup differences that might represent confounding variables that would need to be included as covariates in subsequent analyses. Chi-square tests among various demographic characteristics revealed no significant differences between mothers and fathers in this study?s sample. Independent-samples t-tests among the primary variables of the study (i.e., overall involvement in PFLAG, involvement in support, involvement in education, involvement in advocacy, family functioning, parent- child relationship functioning, parents? perception of changes in parent-child relationship functioning since their child?s disclosure, parents? perception of changes in family and parent-child relationship functioning following their involvement in PFLAG, gender role attitudes, religious orientation, and heterosexist attitudes) revealed significant differences 104 between mothers and fathers in gender role attitudes (t=3.36, p=.00) and parents? perception of changes in parent-child relationship functioning since their child?s disclosure (t=2.80, p=.01). Mothers had more egalitarian gender role attitudes and perceived more positive change in their relationships with their children following their child?s disclosure of her/his same-sex orientation than did fathers. T-tests also revealed significant differences between parents with sons and parents with daughters in gender role attitudes (t=2.06, p=.04) and parents? perception of changes in family functioning following their involvement in PFLAG (t=-2.16, p=.03). Parents with daughters had more egalitarian gender role attitudes and indicated that their involvement in PFLAG had had less of an influence on their family relationships than did parents with sons. In addition, t-tests revealed a significant difference between parents who had more than one child who had come out to them and parents who did not in parents? perception of changes in parent-child relationship functioning since their child?s disclosure (t=2.08, p=.04). Parents with more than one child who had come out to them perceived more positive change in their relationships with their children following their children?s disclosure than did parents with only one child who is out. Finally, a oneway ANOVA revealed significant differences among varying levels of time since the child?s disclosure in parents? perception of changes in parent-child relationship functioning since their child?s disclosure (F[5,158] = 2.31, p = .05) and parents? perception of changes in both family functioning (F[5,150] = 3.09, p = .01) and parent-child relationship functioning (F[5,152] = 2.25, p = .05) following their involvement in PFLAG. Thus, the longer parents have known about their children?s same-sex sexual orientation, the more they have perceived positive change in their 105 relationships with their children following their children?s disclosure and the more they have perceived that their involvement in PFLAG has influenced their family and parent- child relationships. It should be noted that all of the t-tests reported here are significant at the .05 level. In considering confounding variables for subsequent primary analyses, and also to account for practical differences between the actual mean scores, only those variables on which differences were significant at the .01 level were considered (e.g., gender role attitudes), unless otherwise specified. Primary Analyses Correlational analyses. Correlations among all of the primary variables were examined to provide an overview of relationships (Table 4) as well as to test several of the study?s research questions. ------------------------------------------ Insert Table 4 here ------------------------------------------ Research question 1 explored relationships among the attitudinal variables (gender role attitudes, religious orientation, and heterosexist attitudes) and the relationship variables (current family functioning, current parent-child relationship functioning, and parents? perception of changes in parent-child relationship functioning since their child?s disclosure). As can be seen in Table 4, results indicated significant inverse relationships between heterosexist attitudes and parent-child relationship functioning (r = -.21, p = .01) and between heterosexist attitudes and parents? perception of changes in parent-child relationship functioning since their child?s disclosure (r = -.25, p = .01). Thus, the more parents endorsed heterosexist attitudes, the less positively they viewed their current parent-child relationship functioning and the less they reported positive changes in their Table 4 Correlations Among Main Study Variables Measures 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 1. Family ----- 2. P-C Rel -.51** ----- 3. Disclosure -.05 .28** ----- 4. Fam. Infl .10 -.02 .35** ----- 5. P-C Infl .14 .03 .43** .82** ----- 6. HA .13 -.21** -.25** -.03 .01 ----- 7. GRA -.05 -.02 .17 -.12 .03 -.44** ----- 8. Sex Att .04 -.08 -.12 -.14 -.12 -.11 .23** ----- 9. IR .09 .10 .06 .03 .07 .09 -.06 -.28** ----- 10. ER -.03 .07 .00 .11 .12 .00 .03 .05 -.23** ----- 11. Quest .05 -.08 .06 .06 .13 -.14 .09 -.03 .16 .10 ----- 12. PI Supp -.09 .13 .08 .19 .13 -.28** .09 -.04 -.07 .18 .16 ----- 13. PI Educ -.09 .13 .11 .22** .23** -.26** .15 -.05 .07 .07 .20 .72** ----- 14. PI Advo -.05 .12 .34** .24** .24** -.40** .15 .05 -.10 .06 .16 .52** .58** ----- 15. Overall .06 .01 .15 .38** .32** -.21** -.05 -.09 .04 .09 .06 .44** .40** .50** ----- Note. Family = Family Functioning; P-C Rel = Parent-Child Relationship; Disclosure = Perception of Change after Disclosure; Fam. Infl = Influence of PFLAG Involvement on Family; P-C Infl = Influence of PFLAG Involvement on Parent-Child Relationship; HA = Heterosexist Attitudes; GRA = Gender Role Attitudes; Sex Att = Sex/Dating Attitudes; IR = Intrinsic Religious Orientation; ER = Extrinsic Religious Orientation; Quest = Quest Religious Orientation; PI Supp = Involvement in Support; PI Educ = Involvement in Education; PI Advo = Involvement in Advocacy; Overall = Overall Involvement. **p < .01. 107 relationships with their children since disclosure. No other relationships were found among gender role attitudes, religious orientation, and relationship functioning. Research question 2 explored the relationships among parents? perception of changes in parent-child relationship functioning since their child?s disclosure and parents? perception of changes in family and parent-child relationship functioning since beginning their involvement in PFLAG. As can be seen in Table 4, results indicated a significant positive relationship between parents? perception of change in parent-child relationship functioning since their child?s disclosure and their perception of change in family functioning since beginning their involvement in PFLAG (r = .35, p = .01). There was also a significant positive relationship between parents? perception of change in parent- child relationship functioning since disclosure and their perception of change in parent- child relationship functioning since beginning their involvement in PFLAG (r = .43, p = .01). Thus, the more parents reported positive changes in their relationships with their children following their child?s disclosure, the more they perceived that their involvement in PFLAG had influenced their family and parent-child relationships. Research Question 3 explored the relationships among the attitudinal variables (gender role attitudes, religious orientation, and heterosexist attitudes) and the support group variables (parents? overall involvement in PFLAG and level of involvement in supportive, educational, and advocacy activities). As can be seen in Table 4, results indicated significant inverse relationships between heterosexist attitudes and overall involvement (r = -.21, p = .01), between heterosexist attitudes and involvement in supportive activities (r = -.28, p = .01), between heterosexist attitudes and involvement in educational activities (r = -.26, p = .01), and between heterosexist attitudes and 108 involvement in advocacy activities (r = -.40, p = .01). Thus, the more participants were involved in PFLAG overall as well as in support, education, and advocacy activities within PFLAG, the less they endorsed heterosexist attitudes. There were no other significant relationships among the attitudinal variables and support group variables. Regression analyses. Research Question 4 explored whether parents? involvement in support, education, and advocacy predicted heterosexist attitudes. A simultaneous regression analysis (Table 5) was run with heterosexist attitudes as the criterion variable and involvement in support, involvement in education, and involvement in advocacy as the predictor variables. ------------------------------------------ Insert Table 5 here ------------------------------------------ The results of this analysis indicated that the linear combination of the three support group variables was significantly related to heterosexist attitudes, F(3, 158) = 11.10, p < .01. The multiple correlation coefficient was .42, indicating that approximately 17% of the variance in heterosexist attitudes can be accounted for by the linear combination of the support group variables (adjusted R 2 = .17). The X values further indicated that participation in advocacy was found to be a significant unique predictor of heterosexist attitudes. These results suggest that parents with higher levels of participation in advocacy activities are least likely to endorse heterosexist attitudes. Research Question 4 also sought to explore whether parents? involvement in support, education, and advocacy differentially predicted gender role attitudes. Preliminary t-tests had indicated that there were significant differences between mothers and fathers and between parents with sons and parents with daughters in gender role attitudes. Thus, a 109 Table 5 Simultaneous Regression Analysis for Involvement in Support, Involvement in Education, and Involvement in Advocacy Predicting Heterosexist Attitudes (N = 162) Variable B SEB X Involvement in Support -.02 .02 -.11 Involvement in Education .00 .03 .02 Involvement in Advocacy -.06** .02 -.37** Note. R 2 = .17 (p < .01). **p < .01. 110 hierarchical regression analysis (see Table 6) was run with gender role attitudes as the criterion variable, parent gender and child gender as covariates in the first block of the regression analysis, and involvement in support, involvement in education, and involvement in advocacy as predictor variables in the second block. ------------------------------------------ Insert Table 6 here ------------------------------------------ Results indicated that parent gender and child gender were significantly related to gender role attitudes, F(2, 156) = 8.37, p < .01. The multiple correlation coefficient was .31, indicating that approximately 10% of the variance in gender role attitudes can be accounted for by a combination of parent gender and child gender (adjusted R 2 = .10). The X values further indicated that parent gender was found to be a significant unique predictor of gender role attitudes. Results indicated that the three support group variables did not account for a significant proportion of the variance in gender role attitudes above and beyond parent gender and child gender. These results suggest that mothers as well as parents with daughters are more likely to have egalitarian gender role attitudes. Research Question 5 sought to explore whether the attitudinal variables as a group (gender role attitudes, intrinsic religious orientation, extrinsic religious orientation, quest orientation, and heterosexist attitudes) and the support group variables as a group (overall involvement, involvement in support, involvement in education, and involvement in advocacy) differentially predicted parent-child relationship functioning. A hierarchical regression analysis (see Table 7) was run with parent-child relationship functioning as the criterion variable, the five attitudinal variables as predictor variables in the first block, and the four support group variables as predictor variables in the second block. 111 Table 6 Hierarchical Regression Analysis for Involvement in Support, Involvement in Education, and Involvement in Advocacy Predicting Gender Role Attitudes Controlling for Parent Gender and Child Gender (N = 159) Variable B SEB X R 2 YR 2 Step 1 .10** Parent Gender .19** .05 .27** Child Gender .12 .06 .15 Step 2 .12 .02 Involvement in Support -.02 .03 -.09 Involvement in Education .02 .03 .09 Involvement in Advocacy .02 .02 .13 Note. **p < .01. 112 ------------------------------------------ Insert Table 7 here ------------------------------------------ The results of this analysis indicated that the linear combination of attitudinal variables was significantly related to parent-child relationship functioning, F(5, 149) = 4.02, p < .01. The multiple correlation coefficient was .35, indicating that approximately 12% of the variance in parent-child relationship functioning can be accounted for by the linear combination of the attitudinal variables (adjusted R 2 = .12). X values indicated that heterosexist attitudes was found to be a unique significant predictor of parent-child relationship functioning. Thus, the less heterosexist bias parents endorsed, the more likely they perceived their relationships with their children positively. Results further indicated that the support group variables did not contribute a significant amount of variance over and above the attitudinal variables. Research Question 5 also sought to test whether the attitudinal variables as a group (gender role attitudes, intrinsic religious orientation, extrinsic religious orientation, quest orientation, and heterosexist attitudes) and the support group variables as a group (overall involvement, involvement in support, involvement in education, and involvement in advocacy) predicted family functioning. A hierarchical regression analysis was run with family functioning as the criterion variable, the five attitudinal variables as predictor variables in the first block, and the four support group variables as predictor variables in the second block. As can be seen in Table 8, the results of this analysis did not indicate that any linear combination of predictor variables was significantly related to family functioning, and none of the attitudinal variables or support group variables contributed a significant amount of variance to family functioning. 113 Table 7 Hierarchical Regression Analysis for Attitude Variables and Support Group Variables Predicting Parent-Child Relationship Functioning (N = 155) Variable B SEB X R 2 YR 2 Step 1 .12** Gender Role Attitudes -.27 .14 -.17 Heterosexist Attitudes -.55** .15 -.32** Intrinsic Orientation .16 .08 .17 Extrinsic Orientation .27 .13 .16 Quest Orientation -.17 .10 -.14 Step 2 .14 .02 Overall Involvement -.00 .00 -.15 Involvement in Support .03 .05 .09 Involvement in Education .01 .05 .03 Involvement in Advocacy .01 .03 .05 Note. **p < .01 114 ------------------------------------------ Insert Table 8 here ------------------------------------------ Finally, Research Question 6 sought to explore whether parents? overall involvement and involvement in support, education, and advocacy differentially predicted their perception of changes in family functioning following their involvement in PFLAG. Preliminary t-tests had indicated that there were significant differences between parents with sons and parents with daughters and among varying levels of time since the child?s disclosure in parents? perception of changes in family functioning following their involvement in PFLAG. Thus, a hierarchical regression analysis (see Table 9) was run with perception of changes in family functioning as the criterion variable, child gender and time since disclosure as covariates in the first block, and overall involvement, involvement in support, involvement in education, and involvement in advocacy as predictor variables in the second block. ------------------------------------------ Insert Table 9 here ------------------------------------------ As can be seen in Table 9, results indicated that child gender and time since disclosure were significantly related to parents? perception of changes, F(2, 146) = 8.41, p < .01. The multiple correlation coefficient was .32, indicating that approximately 10% of the variance in parents? perception of changes can be accounted for by a combination of child gender and time since disclosure (adjusted R 2 = .10). These results suggest that parents with sons and parents who have known about their child?s same-sex orientation for longer periods of time are more likely to indicate that their involvement in PFLAG has 115 Table 8 Hierarchical Regression Analysis for Attitude Variables and Support Group Variables Predicting Family Functioning (N = 151) Variable B SEB X R 2 YR 2 Step 1 .02 Gender Role Attitudes .03 .14 .02 Heterosexist Attitudes .15 .15 .09 Intrinsic Orientation .05 .08 .06 Extrinsic Orientation -.05 .13 -.03 Quest Orientation .10 .10 .09 Step 2 .04 .02 Overall Involvement .00 .00 .13 Involvement in Support -.02 .05 -.07 Involvement in Education -.04 .05 -.10 Involvement in Advocacy .09 .03 .04 116 Table 9 Hierarchical Regression Analysis for Overall Involvement, Involvement in Support, Involvement in Education, and Involvement in Advocacy Predicting Perception of Changes in Family Functioning Controlling for Child Gender and Time Since Disclosure (N = 149) Variable B SEB X R 2 YR 2 Step 1 .10** Child Gender -.41 .21 -.16 Time Since Disclosure .33** .10 .26** Step 2 .22** .12** Overall Involvement .00** .00 .29** Involvement in Support -.03 .09 -.04 Involvement in Education .08 .09 .10 Involvement in Advocacy .04 .06 .07 Note. **p < .01. 117 had an influence on changes in their family functioning. The X values further indicated that time since disclosure was found to be a unique significant predictor of parents? perception of changes. Results further indicated that the four support group variables accounted for a significant proportion of the variance in parents? perception of changes above and beyond child gender and time since disclosure, F(2, 142) = 6.56, p < .01. The multiple correlation coefficient for this model was .47, indicating that approximately 22% of the variance in parents? perception of changes can be accounted for by a combination of all predictors plus the covariates (adjusted R 2 = .22). Approximately 12% of the variance in parents? perception of changes can be accounted for by the four support group variables alone (YR 2 = .12). Furthermore, X values indicated that overall involvement was found to be a unique significant predictor of parents? perception of changes. These results suggest that the more parents are involved in support activities, in education activities, in advocacy activities, and especially overall, the more they perceive changes in family functioning following their involvement in PFLAG. Research Question 6 sought to test whether parents? overall involvement and involvement in support, education, and advocacy predicted their perception of changes in parent-child relationship functioning following their involvement in PFLAG. A preliminary ANOVA test had indicated that there were significant differences among varying levels of time since the child?s disclosure in parents? perception of changes in parent-child relationship functioning following their involvement in PFLAG. Thus, a hierarchical regression analysis (see Table 10) was run with perception of changes in parent-child relationship functioning as the criterion variable, time since disclosure as a 118 covariate in the first block, and overall involvement, involvement in support, involvement in education, and involvement in advocacy as predictor variables in the second block. ------------------------------------------ Insert Table 10 here ------------------------------------------ Results indicated that time since disclosure was significantly related to parents? perception of changes, F(1, 149) = 6.85, p = .01. The multiple correlation coefficient was .21, indicating that approximately 4% of the variance in parents? perception of changes can be accounted for by time since disclosure (adjusted R 2 = .04). These results suggest that parents who have known about their child?s same-sex orientation for longer periods of time are more likely to indicate that their involvement in PFLAG has had an influence on changes in their parent-child relationship functioning. Results further indicated that the four support group variables accounted for a significant proportion of the variance in parents? perception of changes above and beyond time since disclosure, F(5, 145) = 4.24, p < .01. The multiple correlation coefficient for this model was .36, indicating that approximately 13% of the variance in parents? perception of changes can be accounted for by a combination of all predictors plus the covariate (adjusted R 2 = .13). Approximately 9% of the variance in parents? perception of changes can be accounted for by the four support group variables alone (YR 2 = .09). These results suggest that the more parents are involved in support activities, in education activities, in advocacy activities, and overall, the more they perceive changes in parent- child relationship functioning following their involvement in PFLAG. Research Question 6 lastly sought to explore whether parents? overall involvement and involvement in support, education, and advocacy differentially 119 Table 10 Hierarchical Regression Analysis for Overall Involvement, Involvement in Support, Involvement in Education, and Involvement in Advocacy Predicting Perception of Changes in Parent-Child Relationship Functioning Controlling for Time Since Disclosure (N = 151) Variable B SEB X R 2 YR 2 Step 1 .04** Time Since Disclosure .29** .11 .21** Step 2 .13** .09** Overall Involvement .00 .00 .23 Involvement in Support -.13 .10 -.15 Involvement in Education .14 .10 .16 Involvement in Advocacy .05 .06 .09 Note. **p < .01. 120 predicted their perception of changes in parent-child relationship functioning since their child?s disclosure of her/his same-sex orientation. Preliminary t-tests had indicated that there were significant differences between mothers and fathers, between parents who had more than one child who has come out to them and parents who did not, and among varying levels of time since their child?s disclosure in parents? perception of changes in parent-child relationship functioning since their child?s disclosure. Thus, a hierarchical regression analysis (see Table 11) was run with perception of changes in parent-child relationship functioning since disclosure as the criterion variable, parent gender, number of out children, and time since disclosure as covariates in the first block, and overall involvement, involvement in support, involvement in education, and involvement in advocacy as predictor variables in the second block. ------------------------------------------ Insert Table 11 here ------------------------------------------ As can be seen in Table 11, results indicated that the three covariates together were significantly related to parents? perception of changes, F(3, 147) = 5.36, p < .01. The multiple correlation coefficient was .31, indicating that approximately 10% of the variance in parents? perception of changes can be accounted for by a combination of parent gender, number of out children, and time since disclosure (adjusted R 2 = .10). These results suggest that mothers, parents who have more than one child out to them, and parents who have known about their child?s same-sex orientation for longer periods of time are more likely to perceive positive changes in their parent-child relationship since their child?s disclosure. 121 Table 11 Hierarchical Regression Analysis for Overall Involvement, Involvement in Support, Involvement in Education, and Involvement in Advocacy Predicting Perception of Changes in Parent-Child Relationship Functioning Since Disclosure Controlling for Parent Gender, Number of Out Children, and Time Since Disclosure (N = 151) Variable B SEB X R 2 YR 2 Step 1 .10** Parent Gender .28 .12 .18 Number of Out Children -.27 .17 -.13 Time Since Disclosure .15 .06 .19 Step 2 .22** .12** Overall Involvement .00 .00 .04 Involvement in Support -.06 .06 -.12 Involvement in Education -.06 .06 -.13 Involvement in Advocacy .15** .03 .43** Note. **p < .01 122 Results further indicated that the four support group variables accounted for a significant proportion of the variance in parents? perception of changes above and beyond the three covariates, F(3, 147) = 5.36, p < .01. The multiple correlation coefficient for this model was .47, indicating that approximately 22% of the variance in parents? perception of changes can be accounted for by a combination of all predictors plus the covariates (adjusted R 2 = .22). Approximately 12% of the variance in parents? perception of changes can be accounted for by the four support group variables alone (YR 2 = .12). Furthermore, X values indicated that participation in advocacy activities was found to be a unique significant predictor of parents? perception of changes. These results suggest that the more parents are involved overall, and the more they are involved in support activities, in education activities, and especially in advocacy activities, the more they perceive positive changes in their relationships with their children since their child?s disclosure of her/his same-sex orientation. 123 Chapter 5 Discussion The following paragraphs will discuss in more detail the results pertaining to each research question as well as the strengths and limitations of the study. The summary of findings and comparison to existing literature is organized in two sections: Parental Attitudes and Family Relationships and Support Group Involvement. Implications of the overall findings for future research, clinical work, training and supervision, and advocacy will then be discussed. Summary of Findings and Comparison to Existing Literature Parental Attitudes and Family Relationships As the cultural climate for lesbian, gay, and bisexual (LGB) individuals has undergone dramatic changes in the last three decades, coming out to family may be less crisis provoking for many than it would have been 30 years ago (Beeler & DiProva, 1999). Indeed, the overall picture of the participants in the current study is one of parents who have generally tolerant and non-heterosexist attitudes, who view their current family and parent-child relationship functioning as healthy and positive, and who perceive positive changes in relationships with their children following their child?s disclosure of her/his same-sex orientation. More importantly, and related to this study?s attempt to understand what might facilitate parents? understanding and acceptance of their child?s same-sex orientation, results indicated that heterosexist attitudes were the strongest predictor of parent-child relationship functioning, and parents with less heterosexist attitudes perceived more positive changes in their relationships with their children since their child?s disclosure. 124 Although parents may still experience reactions such as grief and confusion, which is what previous literature typically has focused on, it may be that as the cultural environment within which families find out about a member?s same-sex orientation becomes increasingly gay positive, the challenge that families now face is that of integrating their LGB children into the family with new or changed expectations, rituals, and relationships. In fact, given their perceptions of their parent-child and family relationships, it appears that the parents in the current study fit best in the ?Integration? phase of Herdt and Koff?s (2000) typology of how families with an LGB member differ from each other. Integration is generally characterized by having little if any shame or recrimination, having a sense of enhancement in relationships, experiencing conflicts as generating closeness rather than distance, and often reporting involvement in support organizations such as PFLAG (Herdt & Koff). It may be that the parent-child relationships and family relationships of most of the participants in this study were previously healthy and positive, thus allowing for or possibly setting the stage for continued positive relationship functioning as well as positive changes in relationships following the child?s disclosure of her/his same-sex orientation. Previous authors have suggested that the best predictor of post-disclosure relationships between lesbian and gay young adults and their parents is the quality of their relationships before the disclosure (Cohen & Savin-Williams, 1996; Savin- Williams, 1990). In fact, Aquilino (1997) drew from social learning theory in explaining continuity in parent-child relationships as learned patterns of interaction that continue to influence parents? interactions with their grown children over time. However, it is notable 125 that even in this highly biased sample of parents, important associations were found between heterosexist attitudes and relationship functioning. Consistent with the current study, previous research also has found that parents perceive an improvement in the parent-child relationship following disclosure. For example, Ben-Ari (1995) found that a significant majority of her sample of parents, many of whom also were involved in a support group such as PFLAG at the time of the study, said that the relationship with their child improved following discovery of the child?s same-sex orientation. Ben-Ari noted that it would seem reasonable that parents who belong to these types of groups would be more likely to express relatively positive responses to the knowledge that their child is gay. However, results of the current study indicated a significant positive relationship between parent-child relationship functioning and parents? perception of changes in parent-child relationship functioning since their child?s disclosure. Thus, the more positively parents viewed their current parent-child relationship, the more parents reported positive changes in their relationships with their children following their child?s disclosure of her/his same-sex orientation. It is unclear what these parents? relationships with their children were like before disclosure and even immediately following disclosure, but this finding may hint at the importance of relationship quality in how that relationship withstands certain changes or disruptions, apart from parents? involvement in a support group. In addition to early history and past aspects of the parent-child relationship, other factors may play a part in current as well as changes in parent-child relationship functioning. Parents in the present study had high egalitarian gender role attitudes, neutral (neither traditional nor liberal) attitudes toward sex and dating between men and women, 126 and tended to view themselves as more religiously liberal than conservative. These attitudes, along with the participants? more tolerant heterosexist attitudes, could reflect the general nature of individuals who become involved in organizations and support groups such as PFLAG. It may be important, or even necessary, for individuals to hold certain beliefs, values, or attitudes in order to be open to share their experiences with others, join a group such as PFLAG for support or guidance, and find the new experiences in that group or with similar others useful and helpful. In addition, the overall pattern of participants? attitudes in this study is consistent with an extensive body of literature that has examined key correlates of heterosexism and homophobia. For example, consistent with previous research (see Bohan, 1996), the current study found an inverse relationship between participants? heterosexist attitudes and gender role attitudes, indicating that the more parents endorsed heterosexist attitudes, the less they endorsed more egalitarian gender role attitudes. It has been suggested that individuals? affective and behavioral reactions to gay men and lesbians are influenced by a generalized belief system that includes their perceptions of and stereotypes about women and men more generally and their ideas about appropriate gender roles (Kite, 1994). Gender roles are socially and culturally defined prescriptions and beliefs about the behavior and emotions of women and men, and stereotypes about gender role behaviors and gender role self-concepts may be the processes through which gender role attitudes relate to heterosexist attitudes. Gender alone also appears to be a significant influence on parents? attitudes and family/parent-child relationships in the current study. For example, results indicated that parent gender and child gender were significantly related to gender role attitudes, 127 suggesting that mothers as well as parents with daughters are more likely to have egalitarian gender role attitudes. Results also indicated that mothers were more likely to perceive positive changes in their parent-child relationship since their child?s disclosure than fathers. Many researchers (e.g., Spence & Helmreich, 1972; Beere et al., 1984; King & King, 1993) have posited and demonstrated that females score more liberally than males on measures of gender role attitudes, purportedly because women have more to gain socially, vocationally, educationally, and in other ways by shifts away from traditional gender role expectations and behaviors. This may partly explain why mothers in the current study were more likely to have egalitarian gender role attitudes. Previous research illustrating that parents regard cross-gender boys (boys who engage in more traditionally feminine activities) more negatively than cross-gender girls (girls who engage in more traditionally masculine activities) may help to explain results of the current study which found that parents with daughters had more egalitarian gender role attitudes than parents with sons (e.g., Sandnabba & Ahlberg, 1999). Previous research also has found that LGB children reported greater improvement with their mother after coming out than with their father, and this is consistent with the results of the current study regarding mothers being more likely to perceive positive changes in their parent-child relationship since their child?s disclosure than fathers. This may be related to mothers having more egalitarian gender role attitudes and possibly subsequently less heterosexist attitudes than fathers. It would seem, then, that mothers could react less negatively to their child?s coming out and/or could experience more positive changes in their relationships with their LGB children. However, given the significant overrepresentation of mothers as well as gay sons in the current sample 128 (which is consistent with sample representations in previous research of this nature), it may be that parent gender as well as child gender represent confounds in the study and help to explain the results regarding differences between mothers and fathers. In terms of religious attitudes, the participants in the current study, although part of many different organized religions, indicated that they were more religiously liberal than conservative, and this may have contributed to their ability to effectively utilize a resource like PFLAG. In fact, in this study, religious orientation was explored as one of the attitudinal variables, but religious orientation is more a way of approaching religion and using it in one?s life than an actual religious belief system. The parents in this study indicated that they were more intrinsically religious than extrinsically religious, meaning that they viewed religion less as a way of conforming to social conventions and more as a framework through which they can understand and make meaning out of life. There were no results specifically related to religious orientation in the current study, so it is difficult to draw conclusions about these parents? religious orientation. However, it seems reasonable to suggest that individuals who are more fundamentalist or orthodox in their religious beliefs and practices would likely have a more challenging process of coming to terms with their children?s same-sex orientation. Furthermore, these types of individuals would likely not be involved in organizations like PFLAG and thus are rarely participants in this type of research. In fact, parents who are more fundamentalist or orthodox in their religious beliefs may not even be aware of their children?s same-sex orientation if their children choose to remain closeted out of fear of rejection from their parents or out of their own internalized homophobia stemming from religious beliefs. 129 Although it is important to acknowledge the possibility of and potential influence of preexisting attitudes (positive or negative) and relationship quality, results from the current study further found that the attitudinal variables as a whole did significantly influence current parent-child relationship functioning, and, as noted above, heterosexist attitudes was found to be the strongest predictor of parent-child relationship functioning. Thus, the less heterosexist bias parents endorsed, the more likely they perceived their relationships with their children positively. Results also indicated a significant inverse relationship between heterosexist attitudes and parents? perception of changes in parent- child relationship functioning since their child?s disclosure. Again, these results could be an indication of the general nature of the types of participants who would be involved in a group like PFLAG and/or who would participate in this kind of research. It is apparent, however, that parents? attitudes, especially heterosexist attitudes, have some kind of direct influence on their relationships with their LGB children, and this is consistent with previous research, although from the perspective of the children (e.g., Cramer & Roach, 1988; Savin-Williams, 1998; Floyd et al., 1999). Regardless of whether or not parents initially had more tolerant and less traditional attitudes, or of how parents may have developed and shifted their attitudes over time, it may be that having these attitudes has allowed the parents not only to love and accept their children regardless of their sexual orientation, but also to understand their children?s lives and experiences in new and different ways, thus enhancing their relationships. Support Group Involvement In this study, participants? experiences in a support group designed to provide support and education to family members and friends of LGB individuals were explored. 130 Most previous research on parental reactions to their child?s disclosure of a same-sex identity has focused on parents? initial reactions, often negative, and the ultimate progression toward acceptance for parents (Savin-Williams & Dube, 1998). Research has not directly addressed the process that parents experience in accepting their LGB child, and results from the current study regarding support group involvement may shed some understanding of this process, particularly within the context of previously discussed results regarding parents? perceptions of their current family relationships. First, it may be helpful to get a sense of who participates in a support group such as PFLAG and how they compare to participants of support and self-help groups in general. Although there is very little previous research that has formally assessed the prevalence and membership characteristics of self-help group participants, the characteristics of the participants in the current study appear to be distinct as well as similar to some demographic characteristics that have been reported. According to previous research with large community surveys across the U.S. (N = 20,000), men are more likely than women to use self-help services (rather than professional mental health services), individuals who are divorced or separated are more likely to use self-help services than those who are married, and Caucasians are more likely to participate in self-help groups than any other racial/ethnic group (Lieberman & Snowden, 1994). In addition, individuals with low incomes (0- 20,000/year) are more likely to use self-help services than are middle-class and affluent individuals, and the highest use of self-help groups is among those 30 to 45 years old, with the next highest being the 46-59 age group, and the lowest group being those over 60 years of age (Klaw & Humphreys, 2004). 131 Compared to these demographics, self-help group participants in the current study were similar in that the majority of participants were Caucasian. Participants in the current study were different, however, in that the majority were female, between the ages of 50 and 70, married, and middle- to upper-class based on their education level, salary, and employment status. These differences may reflect the nature of the type of group that PFLAG represents as well as the nature of groups that are more typically explored and discussed in the support and self-help group literature. The most common types of groups are generally those that address substance abuse, chronic physical or mental illnesses, and other forms of distress experienced by people or their caregivers. The lack of racial/ethnic minority representation is consistent with previous research demonstrating low utilization rates by racial/ethnic minorities, especially African Americans, of self help groups (Snowden & Lieberman, 1994). This could be the result of a lack of awareness of existing self-help groups as well as a lack of access to groups, along with the possibility that both cultural norms and a discomfort with strangers limit the willingness to engage in self-disclosure. For a group such as PFLAG, it may also be that the structure and style of helping, the issues and topics that are emphasized, and, perhaps most importantly, the lack of similar others, serve as deterrents for racial/ethnic minority families. For example, disclosure of intimate details about one?s life in the company of persons who are neither family members nor close friends may hold less appeal for African Americans. This may also hold true for other racial/ethnic minority groups. One explanation for the overrepresentation of females in this sample is that extensive research has illustrated that significantly higher percentages of children come out to mothers than to fathers, and some research has demonstrated that mothers are 132 generally more accepting of their LGB children than are fathers (D?Augelli et al., 1998; Savin-Williams, 1998). More broadly, extensive research has illustrated that women tend to have less heterosexist bias than men, especially toward gay men, and this may partly be a function of the relationship between heterosexist attitudes and gender role attitudes that has been consistently found. In fact, as noted above, there is also an overrepresentation of gay male children in this sample, and given the particularly higher levels of heterosexist bias of men towards gay men, this could also explain the overrepresentation of females (mothers) in this sample. In terms of age, the majority of this sample?s children were at least 30 years of age, and most participants indicated that they have known about their children?s same-sex orientation for more than three years. It seems reasonable, then, that the members of a support group such as PFLAG would be older than the members of other types of support groups, especially groups for individuals struggling with a particular issue themselves. In the current study, participants reported moderate overall involvement in PFLAG, which indicates that they have been involved for several years and attend PFLAG chapter meetings approximately once per month. Results indicating parents? moderate involvement as well as their perception that their involvement in PFLAG has had minor to some influence on both their family and parent-child relationships could reflect the length of time these parents have been involved. It may be that after having been a part of PFLAG for several years, parents? involvement has decreased over time, particularly as they might have become more comfortable with their child?s same-sex identity. Also, over time parents? perceptions that their involvement has some direct influence on their family and parent-child relationships could have changed or become less salient. In 133 addition, participants indicated that they have engaged in a moderate to a great deal of participation in supportive and educational activities, and a moderate amount of participation in advocacy activities. Given the biased nature of this sample as well as participants? moderate levels of participation in PFLAG, this study found some important results regarding associations among participants? involvement in a support group, their attitudes, and their relationships with their children. For example, one research question found that participants? involvement in support, education, and advocacy as well as their overall involvement significantly influenced their perception of change in both their parent-child relationship and family relationships since becoming involved in PFLAG, with overall involvement being the strongest predictor. Given participants? level of overall involvement as discussed above, the fact that overall involvement was the strongest predictor of parents? perception of change might indicate that length of time, as opposed to more specific forms of involvement, makes the most significant difference in the effects of support group experiences. Indeed, more general research looking at responses to support groups for family members and caregivers (e.g., of individuals with schizophrenia, AIDS, or other disabilities) has found that one important variable that seems to have a pervasive effect is amount of experience as a caregiver or amount of time since diagnosis (e.g., Winefield, Barlow, & Harvey, 1998). In fact, results of the current study also found that parents who have known about their child?s same-sex orientation for longer periods of time are more likely to indicate that their involvement in PFLAG has had an influence on changes in their parent-child relationship and family functioning. Results also indicated that parents who have known about their child?s same-sex 134 orientation for longer periods of time are more likely to perceive positive changes in their parent-child relationship since their child?s disclosure. More generally, previous research on self-help group involvement and psychoeducational programming has indicated that these are effective service delivery alternatives and have been found to be significantly more effective in producing positive change than were control groups not involved in any type of group intervention or psychoeducation (Schreier & Werden, 2000; Lipsey & Wilson, 1993). In the current study, significant positive relationships were found between parents? involvement in educational activities and parents? perception of changes in both family functioning and parent-child relationship functioning following their involvement in PFLAG. Similarly, there were significant positive relationships between parents? involvement in advocacy activities and parents? perception of changes in both family functioning and parent-child relationship functioning following their involvement in PFLAG. These results may hint at the more specific components of involvement that could be beneficial to parents, such as educational and advocacy activities. In fact, attitudes are often changed or influenced through actual experience. For example, previous research has found correlations between heterosexism and having little contact with LGB people as well as between heterosexism and believing that peers hold similar negative attitudes (see Bohan, 1996). It could be expected that positive interactions, including the support and comfort of others who share similar experiences, especially over time, would facilitate reductions in heterosexism and potentially, then, improvements in relationship functioning. Indeed, another research question in the current study found that participants? involvement in support, education, and advocacy as 135 well as their overall involvement significantly influenced their heterosexist attitudes, with involvement in advocacy being the strongest predictor. These results suggest that parents with higher levels of participation in advocacy activities are least likely to endorse heterosexist attitudes. Thus, positive interactions with other parents as well as the LGB community through advocacy could facilitate positive changes in heterosexist attitudes as well as relationships among family members in a number of ways. Contact in the form of education as well as interpersonal interactions might change cognitions by providing information that refutes common misconceptions and stereotypes. Contact also could affect feelings by providing a positive emotional experience, not only with other parents but also with other LGB individuals. Bohan (1996) noted that for such contact to reap maximum benefit, it needs to be continuous, to occur in an atmosphere that encourages cooperative efforts toward shared goals, and to involve common interests, beliefs, and expectations. Results of the current study also indicated that parents who have more than one child out to them are more likely to perceive positive changes in their parent-child relationship since their child?s disclosure. It seems reasonable to suggest that parents who have experienced the coming out of one child would find it easier to experience the coming out of a second child, particularly when considering the above discussion regarding the influence of previous contact and interactions with LGB individuals on heterosexist attitudes. Perhaps effects of exposure, learning, changed attitudes, and integration of their children are positively enhanced for parents with more than one child who is same-sex oriented. 136 In addition, participants? involvement in support, education, and advocacy as well as their overall involvement significantly influenced their perception of change in their parent-child relationship since their child disclosed her/his same-sex orientation, again with involvement in advocacy as the strongest predictor. For the participants in this study, involvement may have directly influenced their relationships with their children through improved communication, increased involvement in their child?s life, and/or an enhanced closeness with their child. However, given the results described above regarding the relationships between support group variables and attitudinal variables, it may also be that involvement in PFLAG, especially involvement in advocacy, influenced participants? relationships with their children through the effects that their involvement had on their attitudes, especially their heterosexist attitudes. One of the ways that previous research has illustrated the effectiveness of support group involvement is having the experience of control (Gartner, Gartner, & Ouellette-Kobasa, 1988). According to this research, self-help and support groups can promote this experience in several ways: through having control and access to information and knowledge, through the common sharing of such feelings as sadness, confusion, and fear, and through providing a source of ?behavioral? control in helping others and providing education and support. Perhaps the results of the current study regarding the significance of involvement in advocacy might partly be explained by this notion that having a deeper sense of control through action (e.g., in supporting others, providing education, modeling advocacy, speaking, lobbying) is one of the main processes through which support group involvement can be effective. 137 Additionally, previous research has indicated that the benefits of social contact may extend beyond received support to include other aspects of the interpersonal relationship, such as giving support to others (Brown, Nesse, Vinokur, & Smith, 2003). Providing help is associated with significant improvement in psychosocial adjustment (Roberts et al., 1999), and even perceptions that are likely to be associated with giving (or advocacy), such as a sense of meaning, purpose, belonging, and mattering, have been shown to increase happiness and decrease depression (Brown et al., 2003). Thus, the positive changes participants in the current study indicated in their parent-child relationships and in their heterosexist attitudes, particularly as they were influenced by involvement in advocacy, may be explained by this notion that participants in self-help groups benefit from both receiving support, understanding, and information and providing these things to others. It may be that the empowerment they might feel through gaining a deeper sense of control or being able to give back to others is a key component of what self-help and support groups provide for members. It may also be that over time the parents in the current study developed new understandings of heterosexism and even gender roles, and this may have motivated them to become activists for the rights of sexual minorities, and of their LGB children more specifically. As noted above, the results indicating participants? general attitudes toward lesbian, gay, and bisexual people, attitudes toward gender roles, and religious orientation might simply reflect the possibility that parents who become involved in organizations like PFLAG (and find it helpful) are already characterized as having more tolerant and liberal attitudes prior to involvement. However, results of the current study indicated significant relationships between parents? attitudes and parents? involvement in PFLAG. Thus, it 138 does seem that parents? attitudes are influenced by their level and type of involvement, and it may be that the attitudes of parents in this study have shifted over time as a result of their involvement. Moreover, perhaps the parents in this study, more than already having more tolerant and liberal attitudes prior to involvement, were simply motivated to understand their experience and children better and cope more effectively with their child?s disclosure. This, again, could reflect the relationships that parents had prior to their child?s coming out, and, in fact, results also indicated that the more parents reported positive changes in their relationships with their children following their child?s disclosure, the more they perceived that their involvement in PFLAG had influenced their family and parent-child relationships. Finally, perhaps a more appropriate model (than the grief/loss models typically associated with parents? coming out experience) for understanding the process by which this population of parents come to understand, accept, and advocate for their children could be a model such as the Social Identity Development Model (Jackson & Hardiman, 1982), a general model that describes the development of people in ?dominant? social groups and is based upon an examination of other models of learning and unlearning oppression, as well as models of racial and gender identity development. This model has five stages: 1) Na?ve (Heterosexual Na?ve), 2) Acceptance (Heterosexual Acceptance), 3) Resistance (Heterosexual Resistance), 4) Redefinition (Heterosexual Redefinition), and 5) Internalization (Heterosexual Internalization). Currently this model has no empirical support, but the results of the present study regarding perceptions of change in family/parent-child relationships and the influence of involvement in support, education, and especially advocacy may be more consistent with this model of acceptance and 139 change than models of grief and loss. It may be that the parents in the current study, and the types of parents who become involved in support groups such as PFLAG, go through similar stages as those in the Social Identity Development Model. It appears that the parents in the current sample might be in the fourth stage, Redefinition, which signifies that these individuals could be in the process of redefining new ways of being heterosexual parents of a sexual minority child and working to bring about social change using new awareness as well as their own privilege. Limitations and Strengths Although several limitations are inherent to the research design employed in this study, several aspects of the study?s methodology also improve upon previous research. One limitation is the generalizability of results based on sample bias. A challenge previously faced by researchers of the coming out process of parents of lesbians and gay men is accessing a wide range of individuals in this process, from parents who have just learned that their child is lesbian or gay to parents who have known for several years, or from parents who have more negative reactions to their child?s same-sex orientation to parents who have more positive reactions. In particular, as illustrated by the majority of participants in the current study who were relatively non-heterosexist, nontraditional in gender role attitudes, and more religiously liberal than conservative, restricted range in attitudinal variables is a common problem. In addition, most of the participants in the current study have known about their children?s same-sex orientation for several years and also have been involved in a support group for many years. Individuals who join groups such as PFLAG might generally be more open, secure, and accepting, and they are likely not representative of many parents 140 with a sexual minority child, especially shortly after the child?s disclosure. Also, although a specific effort was made to recruit a more diverse sample, this study?s sample was predominantly White, middle- to upper-class, highly educated, and married. This is quite consistent with samples in previous research of this nature, but it, again, greatly limits the generalizability of the current results and presents significant confounds in the data. Also, along with other demographic characteristics such as race/ethnicity, SES, age, and marital status, gender could represent a confound in this study. The gender differences found in the study could be a function of the overrepresentation of females and gay sons in the sample. Given the exploratory nature of this study, some of the measures in the study were developed by the principal investigator and thus did not have previous psychometric validity or reliability. Given that there is so little research on support groups such as PFLAG, all of the measurement for support group involvement was developed for this study, as were the measures of parents? perceptions of change in parent-child and family relationships following involvement in PFLAG. In addition, the Intrinsic/Extrinsic and Quest religious orientation scales may not have been the most suitable measures of religious attitudes given the low reliability of the Extrinsic subscale and the degree to which participants indicated that they were uncertain about their Quest orientation. Other statistical considerations which present limitations for the current study include assumptions for tests which were not met. First, the assumption of normality was not met for the family functioning and parent-child relationship functioning variables. This violation may have compromised results that included these variables because the variables? distributions were not normal and thus may have affected regression estimates. 141 Second, the predictor variables in several of the regression analyses were significantly correlated with each other, thus contributing to multicollinearity among the variables. Third, some of the cases may not be independent of each other because they came from mothers and fathers in the same family. There were 25 matched pairs of participants, for a total of 50 participants coming from married couples who each completed a separate questionnaire. It is unclear whether or not data from these participants are confounded because the participants came from the same family and are responding to questions about their family relationships (which are not independent of each other). One final limitation is the retrospective nature of the study, which limits the conclusions drawn about parents? perceptions of change in their family and parent-child relationships over time. Although this study is consistent with and builds on previous research with regards to relationships among attitudes, relationship functioning, and utilization of social support resources, some of the research questions more specifically asked about changes in relationship functioning as well as parents? histories of involvement in PFLAG. These questions might be better explored or understood prospectively, particularly beginning earlier in parents? ?coming out? experiences or very shortly after learning that their child is lesbian, gay, or bisexual. As discussed above, although this study?s exploratory nature contributes to some methodological and conceptual limitations, one strength of the current study is that it adds to the literature by looking at three areas that have not been previously explored together (i.e., attitudes, relationship functioning, and support group involvement). Results of the current study not only contribute to the literature in combining these areas, but also in their respective literatures, such as efficacy research in the support group/self-help 142 literature. Another strength of the current study is that it explored the coming out experience from the perspective of parents themselves. Much previous research is from the perspective of LGB individuals who were asked about their perceptions of how their coming out has affected their parents and/or their relationships with their parents. In addition to adding to the literature with the parent perspective, the current study looked more specifically at the experiences of mothers as well as fathers and differentiated between parent gender (as well as child gender). Indeed, results indicated significant differences between mothers and fathers and also between parents with sons and parents with daughters. Finally, the current study builds upon previous research by taking into account other confounding variables, such as time since disclosure and number of out children, and looking at their particular influence on the variables of interest. Implications Research Based on the current study as well as previous studies looking at parents in PFLAG and similar support groups, it seems apparent that support groups such as PFLAG mainly serve a particular demographic group (i.e., White, married, middle-aged, well-educated, middle- to upper-class parents). In order to broaden our understanding of how this type of resource can be useful to parents and family members of LGB individuals, it seems imperative that future research explore not only the experiences of various racial/ethnic minority parents and families but also the experiences of different types of families (e.g., divorced parents, single parents, parents and families with adolescent LGB members). Increasing numbers of adolescents are disclosing their sexual orientation to parents at much earlier ages (Maguen, Floyd, Bakeman, & Armistead, 2002), and this may be 143 attributed to the relatively greater availability of information and supportive role models for gay, lesbian, and bisexual identities. In addition, the assertion that young gay people of color can encounter additional problems and obstacles while coming out to family members due to their ethnic, racial, and cultural backgrounds has been documented (e.g., Savin-Williams, 1996), but the experience of having a child come out as LGB for racial/ethnic minority parents has not been explored. Thus, recruitment for participants (not only for research but also for membership in support groups such as PFLAG) needs to be expanded and tailored more specifically for different communities and families. Given that involvement in PFLAG did have a significant influence on parents? attitudes and perceptions of changes in their relationships with their children, it might be beneficial to examine more specific aspects of involvement. Although the current study did attempt to look at involvement in support, education, and advocacy in addition to overall involvement, perhaps asking participants about their involvement in more specific activities at different stages of their overall involvement over time might provide a clearer understanding of what processes are helpful, particularly for this specific type of organization (which clearly serves a different kind of population than other self-help groups). Future research might conduct more experimental or pseudo-experimental studies exploring the effects of didactic, psychoeducational workshops compared to participation in support alone or no participation at all. The current study did find that involvement in advocacy had a unique influence on attitudinal and relationship variables, and it might be useful to further explore what implications advocacy activities can have for programming and psychoeducational initiatives. Moreover, the apparent usefulness of 144 involvement in this type of self-help group might justify more general research into the nature of, ideology of, and efficacy of self-help groups such as PFLAG. For the current study, attempts were made to acquire as much information and feedback from consultants and chapter leaders within the PFLAG organization throughout the conceptualization of and development of the study, particularly because there is so little research available regarding this type of population in a self-help group such as PFLAG. Perhaps one methodology that future research can employ is participatory action research (PAR), a scientific paradigm that has been shown to be most relevant for research and work with self-help groups (Chesler, 1991). PAR can be characterized as ?applied research? which directly involves the individuals and organizations whose issues, experiences, or problems are under study. As noted by Chesler (1991), the tremendous variety in organizational form and function, even in local units of a single national umbrella organization such as PFLAG, means that preestablished and standardized research questions, measures, and approaches may fail to adequately tap or be relevant to real-life heterogeneity. Thus, being more collaborative with the organization, in this case with PFLAG and/or its local chapters, would mean involving chapter leaders, professionals in the national office, and chapter members themselves in the design, conduct, and utilization of the research. Through participation, these individuals can increase the researchers? knowledge base as well as learn new skills in gathering and analyzing information that could prove useful in recruiting/maintaining members, programming for diverse needs, running meetings and using members? resources, and influencing the broader community. In fact, results from the current study 145 regarding the important influence of involvement, and especially advocacy, have implications for this notion of allowing members to participate in research of this nature. Finally, given this study?s results indicating parents? perceptions of changes in their relationships with their children since disclosure and also since they became involved in PFLAG, future research should incorporate more prospective designs that follow participants over time. Perhaps future studies could target parents who have just learned that their child is lesbian, gay, or bisexual and follow them throughout the early and later courses of their ?coming out? experience. Similarly, future studies could target parents who have just joined a support group such as PFLAG and follow their involvement in that group over time to explore the effects of involvement on attitude change, relationship change, and the efficacy of support and self-help group interventions. In addition, future research might explore other psychosocial outcomes associated with involvement in a support group such as PFLAG. For example, other racial or prejudicial attitudes could be explored as well as other relationship variables, such as integration of children?s same- sex partners and other forms of social support for children. Practice Results of the current study demonstrated that the availability of support resources and educational information about coming out and same-sex identity is an important component of parents? own coming out experience and influences their attitudes and family relationships. Practitioners should be knowledgeable about local resources (like PFLAG) to link families with other individuals who are dealing with coming out issues. This knowledge could also be useful for LGB clients themselves who are struggling with coming out to family. For example, informed practitioners could help clients (through 146 individual counseling, group counseling, and possibly consulting relationships) recognize that understanding and integrating an LGB child after disclosure requires time. It may take time to locate educational and support resources as well as to reap maximum benefit from participation in educational and supportive interventions. As has been previously suggested, it is of course important for clinicians working with LGB clients and their families not to minimize the ramifications of coming out (Merighi & Grimes, 2000). However, given the significantly positive results of the current study regarding parents? attitudes and changes in their relationships with their children, it may be equally as important to offer hope to LGB clients as well as parents who have just learned that their child is LGB. For example, in counseling people who are preparing to come out, practitioners can facilitate exploration of LGB clients? perceptions of their current family and parent-child relationships. Practitioners could help clients explore their reluctance to disclose to family and whether or not that is based more in fear than in reality. Family members? current attitudes could also be explored, and practitioners might help LGB clients as well as families or parents discuss the possibility that whatever difficulties, if any, they are experiencing, those can be alleviated or changed, especially with the passage of time. The current study may also have implications for group psychotherapy in terms of what elements of self-help group participation are most beneficial. For example, it appears that some elements that might be similar in both self-help groups and professional or therapy groups are emotional support, role modeling, and instillation of hope (all elements that have already been shown to be critical components of group psychotherapy). One element that is different and that might enhance our understanding 147 of what could be beneficial in group psychotherapy is the opportunity that self-help group members have to provide help or services (in contrast to receiving help or services) through advocacy, education, and peer support. Furthermore, involvement in advocacy seemed to play a unique and significant role in the current study?s results regarding parents? attitudes and relationships with their children, and these results may have implications for helping parents support their children in addition to providing a support for parents themselves. Finally, for practitioners who serve as organizational consultants, the results of the current study could be used to help social support organizations such as PFLAG better understand the population they serve, communities they are not being accessible to, how they might provide the most appropriate and relevant resources and interventions, and how they might assess and evaluate their services. For example, based on results of the current study, it seems important that these organizations have and disseminate accurate information regarding heterosexism, gender, gender role attitudes, and family and parent-child relationship functioning. Furthermore, these organizations could benefit from information regarding what aspects of involvement are most helpful and how they can develop programming and activities with that knowledge. Training and Supervision The results of the current study highlight the potential usefulness and effectiveness of resources such as support groups for family members, and particularly parents, of LGB individuals. For graduate students working with LGB clients, such as college students who have recently come out or who are contemplating coming out, it seems important that they be encouraged to explore social support resources not only for 148 themselves but also for their parents and family. As noted above, it would indeed be important for trainees to be encouraged to explore the consequences of coming out and the range of reactions that they and their families may experience. It might also be beneficial, however, for trainees and their supervisors to be aware of other important areas for exploration, such as strengths in their families and positive relationship quality that could buffer negative reactions, as well as local and community resources that could be helpful to clients. Additionally, the current study and its focus on parents? perspectives might have implications for the importance of incorporating more varied practica in training programs that allow graduate students to work with more diverse populations that could benefit from additional support, expertise, and resources. For example, training programs might incorporate more of a community-focused practicum in which students have the opportunity to work within organizations and communities under supervision and in conjunction with a didactic portion that focuses on diverse populations, their needs, and interventions that would be most appropriate for them. Based on results from the current study, one such practicum could take place in a local PFLAG chapter or the national PFLAG office and could involve trainees working with parent groups and PFLAG leaders in a variety of ways, such as group counseling, individual or couples counseling with parents, workshop and program development, and evaluation of services. Thus, this training experience could be in the form of a consultation seminar which focuses more on outreach, program development, and/or program evaluation, or it could have more of a clinical focus and involve individual, couples, or group counseling with families. In both cases, areas of focus might be current beliefs/values/attitudes and how those can or need 149 to change, family and parent-child relationship functioning and how those relationships are affected or changed, and more general issues around giving and receiving social support. Finally, current research, clinical issues, and resources related to LGB clients and their families should be addressed during in-services and continuing education programs for professionals. In order to serve clients effectively as well as to serve as good role models and teachers for trainees, it is imperative that psychologists be current and knowledgeable about potential interventions and resources. Such programs should emphasize local and community resources and how they can be helpful to clients and community members, and knowledge about what works in those types of resources or groups (such as PFLAG) could be used in work with clients (e.g., paying attention to and assessing social support resources, beliefs/attitudes/values, relationships with others). Advocacy A better understanding of as well as involvement in organizations such as PFLAG has important implications for advocacy on a variety of levels, including individual, family, community, and socio-political levels. For example, results of the current study regarding the significant influence of involvement in advocacy on parents? attitudes as well as perceptions of change in their relationships point to the potential usefulness of advocacy not only for LGB people but also for their family members and family relationships. Advocacy for the rights and privileges of LGB people has obvious implications for those individuals? well-being, but clearly advocacy on behalf of their family members also has implications for the well-being of parents and family members of LGB people. In fact, psychologists and practitioners themselves might consider 150 becoming more involved in advocacy through education, consultation, and support in the community, not only for the benefit of those individuals they are advocating for, but also for the ways in which their own professional knowledge, networks, and impact are enhanced. In addition, the apparent usefulness of involvement in this type of group intervention might justify more professional attention to and lobbying for allocation of federal, state, and local funding. Many of these types of organizations serve as important resources in communities across the U.S. in addition to their influence on national awareness and public policy related to LGB issues. Self-help and support groups can serve as an accessible, low-cost resource, particularly in rural or other communities in which mental health services are not as readily available. Moreover, results of the current study indicated associations between attitudes, especially heterosexist attitudes, and family relationships as well as between attitudes and involvement in organizations like PFLAG. Thus, these organizations? potential for influencing societal attitudes toward LGB individuals has further implications for the importance of lobbying for their existence, maintenance, and success. Conclusion In conclusion, the current study sought to address questions of how parents? attitudes as well as relationships with their LGB children are related to their involvement in a support group, more specifically the organization PFLAG. It appears that the results of this study confirm previous qualitative research that also has explored some of the steps that families, and parents in particular, take as they attempt to understand, acknowledge, accept, and integrate their gay or lesbian child. As indicated in previous research as well 151 as the current study, it appears that accessing resources in the community, participating in supportive or self-help groups, and engaging in social advocacy around gay and lesbian issues are some of these important steps for parents. More specifically, the current study indicated that involvement in PFLAG, especially involvement in advocacy activities, has significant influences on heterosexist attitudes as well as on parents? perceptions of changes in their relationships with their child since their child came out. Moreover, results of the current study indicated that parents? attitudes are significantly related to their current parent-child relationship functioning. In addition to its potential contributions to the coming out/disclosure literature and the self-help/support group literature, the current study has potential implications for practitioners, consultants, and community organizations. It seems that an organization like PFLAG can have several potential benefits for individuals, families, communities, and society. From helping to shape attitudes, to helping to influence relationships, to providing important resources for families and communities, and to having an impact on society through individuals? advocacy efforts, PFLAG and other support groups can be an important step in the coming out processes not only of LGB individuals but also of their parents, families, and communities. 152 Appendix A Recruitment Email to PFLAG Chapter Leaders Dear PFLAG Chapter Leader, This is a follow-up to an email you recently received from Roy Gilbert-Higginson at National PFLAG. My name is Susanna Gallor, and I am a graduate student in Counseling Psychology at the University of Maryland. I am currently working on a project for my dissertation in which I am exploring the experiences of parents who utilize PFLAG support and resources. I previously contacted you in April requesting information about what kinds of resources/services are offered to parents through PFLAG, and I want to thank you for your feedback and assistance. I am now ready to begin seeking participation from parents across various chapters, and I would greatly appreciate any assistance you could offer me. Participation requires that members of PFLAG who have a child who identifies as lesbian or gay complete a questionnaire and return it to me in a pre-paid stamped envelope that will be provided. I would be happy to come to one of your chapter meetings to briefly explain the project and request participation. However, I am also interested in hearing what your thoughts are on what might be the most convenient and appropriate way to reach parent members in your chapter. I would like to respect members? privacy as well as your time and availability. Another option might be for me to mail you several packets at a time, each packet containing a cover letter, questionnaire, and return envelope. I very much appreciate your taking the time to read this email and assist me in any way your can. I look forward to hearing from you and further discussing the possibility for your member participation in this project. Again, thank you for your time and consideration, Susanna Gallor 153 Appendix B DEMOGRAPHIC INFORMATION Age: _______ Gender: ____ Female ____ Male Your Race/Ethnicity: Your Sexual Orientation: ____ African-American/Black ____ Heterosexual ____ Arab/Middle Eastern American ____ Lesbian or Gay ____ Asian American/Pacific Islander ____ Bisexual ____ Caucasian/Euroamerican ____ Uncertain ____ Hispanic or Latino/a ____ Native American/American Indian ____ Other (indicate:____________________) ____ Other (indicate:______________________) Highest level of education you completed: Your employment status: ____ Less than high school ____ Employed ____ High school ____ Retired ____ Technical/Vocational training ____ Unemployed ____ Some college ____ Associate Degree If employed, your work status: ____ Bachelor?s Degree ____ Full time ____ Graduate Degree ____ Part time ____ Professional Degree ____ Hold more than one job Your occupation:_________________________ Approximate family annual income: ____ $0 ? 9,999 Your marital/spousal status: ____ $10,000 ? 19,999 ____ Single ____ $20,000 ? 29,999 ____ Married ____ $30,000 ? 39,999 ____ Partnered/in a committed relationship ____ $40,000 ? 49,999 ____ Divorced ____ $50,000 ? 59,999 ____ Widowed ____ $60,000 ? 69,999 ____ $70,000 ? 79,999 Spouse/Partner?s occupation (if applicable): ____ $80,000 ? 89,999 _____________________________________ ____ $90,000 or greater In which U.S. state do you live? Do you consider your current environment __________________________________ to be: ____ Rural Urban Suburban Do you have more than one child who has come out to you? ____ Yes ____ No 154 Please indicate the following information for each child who has come out to you: Child 1 Male ____ Female ____ Age ____ How your child identifies her/his sexual orientation: ____ Lesbian and/or primarily attracted to women ____ Gay and/or primarily attracted to men ____ Bisexual and/or attracted to men and women approximately equally ____ Other (indicate:____________________________________________) Length of time since your child disclosed her/his sexual orientation to you: ____ Less than 1 month ____ 1 to 3 months ____ 4 to 6 months ____ 7 to 9 months ____ 10 to 12 months ____ More than 1 year ____ More than 2 years ____ More than 3 years ____ Other (indicate:_________________________) Does your child currently live at home? Yes____ No____ Child 2 (if applicable) Male ____ Female ____ Age ____ How your child identifies her/his sexual orientation: ____ Lesbian and/or primarily attracted to women ____ Gay and/or primarily attracted to men ____ Bisexual and/or attracted to men and women approximately equally ____ Other (indicate:____________________________________________) Length of time since your child disclosed her/his sexual orientation to you: ____ Less than 1 month ____ 1 to 3 months ____ 4 to 6 months ____ 7 to 9 months ____ 10 to 12 months ____ More than 1 year ____ More than 2 years ____ More than 3 years ____ Other (indicate:_________________________) Does your child currently live at home? Yes____ No____ 155 Child 3 (if applicable) Male ____ Female ____ Age ____ How your child identifies her/his sexual orientation: ____ Lesbian and/or primarily attracted to women ____ Gay and/or primarily attracted to men ____ Bisexual and/or attracted to men and women approximately equally ____ Other (indicate:____________________________________________) Length of time since your child disclosed her/his sexual orientation to you: ____ Less than 1 month ____ 1 to 3 months ____ 4 to 6 months ____ 7 to 9 months ____ 10 to 12 months ____ More than 1 year ____ More than 2 years ____ More than 3 years ____ Other (indicate:_________________________) Does your child currently live at home? Yes____ No____ Do you have other family members who identify as lesbian, gay, bisexual, or transgendered? If so, please indicate your relationship with her/him. ____ No ____ Yes (indicate:_____________________________) NOTE: Many of the items in the remainder of this questionnaire refer to your experiences and relationship with your child. If more than one child has come out to you, please complete the remainder of this questionnaire based on your experiences with any one of your lesbian or gay children. Comments (optional): 156 Appendix C PFLAG INVOLVEMENT This part of the survey inquires about your participation in PFLAG. Please answer all questions based on your own personal level of involvement with PFLAG. How long have you been participating in PFLAG? How often do you attend PFLAG meetings? ____ Less than 1 month ____ More than once per month ____ 1 to 3 months ____ Once per month ____ 4 to 6 months ____ Once every two months ____ 7 to 9 months ____ Once every three months ____ 10 to 12 months ____ Once or twice per year ____ 1 to 2 years ____ Once every year or two ____ 2 to 3 years ____ Other (indicate: ___________________) ____ 3 to 5 years ____ 5 to 10 years ____ More than 10 years ____ Other (indicate: ___________________) The following questions ask about what kinds of experiences you have had in your PFLAG chapter. PFLAG offers opportunities for support, education, and advocacy through a variety of resources and activities. Considering the definitions and examples for support, education, and advocacy given below, please rate the extent to which you have participated in such activities. In other words, to what extent have you participated in supportive activities, educational activities, and advocacy activities? Support: Supportive activities connect you to other individuals and groups for the purpose of obtaining emotional support and understanding from others whose circumstances may be similar to your own and who have experience with similar kinds of issues or concerns. Examples include: Support group discussion Sharing experiences with others Meeting individually with other PFLAG members or staff) Social event (e.g., potluck, movie night) ?New Parents? meeting (for new PFLAG members) 1 2 3 4 5 6 7 No Moderate A great deal participation participation of participation Comments (optional): 157 Education: Educational activities include exposure to materials and resources that are designed to convey information about the gay and lesbian community, gay and lesbian issues, family issues related to having a family member who has come out, and services for gay men, lesbians, and their families. Examples include: Reading about information related to gay and lesbian issues through brochures, books, and other literature Learning of community information and resources, such as local events, news stories, services for gays and lesbians and their families Hearing a speaker from the community Viewing or listening to videotapes/audiotapes related to gay and lesbian issues Looking at the PFLAG website 1 2 3 4 5 6 7 No Moderate A great deal participation participation of participation Comments (optional): Advocacy: Advocacy activities include activities organized around promoting public awareness about gay and lesbian issues and the gay and lesbian community, educating groups, communities, and the general public about gay and lesbian issues, and helping to change negative attitudes in others. Examples include: Participating in marches, pride events, charities, fundraisers, etc. Receiving assistance with organizing events, talks, or seminars led by parents (e.g., going to a local high school or college to talk about a particular topic, organizing a radio program, providing a workshop at your place of employment) Legislative lobbying and involvement 1 2 3 4 5 6 7 No Moderate A great deal participation participation of participation Comments (optional): 158 Please indicate the extent to which your PFLAG chapter emphasizes supportive, educational, and advocacy activities: Support: 1 2 3 4 5 6 7 None Some A great deal Education: 1 2 3 4 5 6 7 None Some A great deal Advocacy: 1 2 3 4 5 6 7 None Some A great deal What PFLAG resource/activity/publication(s) has been MOST helpful to you? ______________________________________________________________________________ What PFLAG resource/activity/publication(s) has been LEAST helpful to you? ______________________________________________________________________________ What would you like to see PFLAG do to better support parents with lesbian or gay children? ______________________________________________________________________________ ______________________________________________________________________________ After your child came out to you, how did you discover PFLAG? ______________________________________________________________________________ ______________________________________________________________________________ After your initial contact with PFLAG, what motivated you to become more involved or remain in the organization? ______________________________________________________________________________ ______________________________________________________________________________ Has the nature or kind of involvement you have in PFLAG changed over time? ____ Yes ____ No If yes, how? ______________________________________________________________________________ ______________________________________________________________________________ 159 How satisfied are you with your PFLAG chapter?s supportive activities and resources? 1 2 3 4 5 6 7 Very Moderately Very Dissatisfied Satisfied Satisfied Comments (optional): How satisfied are you with your PFLAG chapter?s educational activities and resources? 1 2 3 4 5 6 7 Very Moderately Very Dissatisfied Satisfied Satisfied Comments (optional): How satisfied are you with your PFLAG chapter?s advocacy activities and resources? 1 2 3 4 5 6 7 Very Moderately Very Dissatisfied Satisfied Satisfied Comments (optional): 160 Appendix D ATTITUDES ABOUT LESBIAN, GAY, AND BISEXUAL PEOPLE Please rate the extent to which you agree or disagree with the following statements about lesbians, gay men, and bisexual individuals. 1 2 3 4 5 Strongly Disagree Neither Disagree Agree Strongly Disagree or Agree Agree Strongly Strongly Disagree Agree 1. All sexual orientations are natural expressions of 1 2 3 4 5 human sexuality. 2. Positive aspects of various sexual orientations 1 2 3 4 5 should be included in public education. 3. I believe that the lives of lesbian, gay, and bisexual 1 2 3 4 5 individuals could not be as fulfilling as those of heterosexual individuals. 4. Only heterosexual individuals are appropriate 1 2 3 4 5 religious leaders. 5. I think society will benefit from fostering equal 1 2 3 4 5 opportunity employment for lesbian, gay, and bisexual individuals. 6. Heterosexual couples make better candidates for 1 2 3 4 5 parents than do same-sex couples for adoption. 7. I would accept my sibling?s partner regardless of 1 2 3 4 5 his or her sex. 8. No one sexual orientation is better than any other 1 2 3 4 5 sexual orientation. 9. An anti-discrimination law is incomplete without 1 2 3 4 5 the inclusion of sexual orientation. 10. There is no reason to restrict lesbian, gay, or 1 2 3 4 5 bisexual individuals from working in the military. 11. I think lesbian/gay/bisexual people are unfit as 1 2 3 4 5 teachers. 12. I would not think less of my co-worker if I found 1 2 3 4 5 out that he or she was a lesbian, gay, or bisexual individual. 13. My relationship with my friend would change if I 1 2 3 4 5 found out that he or she was not heterosexual. 14. I make sure to invite the partner of my lesbian or 1 2 3 4 5 gay friend to social functions. 15. In general, heterosexual people are more 1 2 3 4 5 psychologically adjusted than lesbian, gay, and bisexual people. 16. Legalization of same-sex marriage will dismantle 1 2 3 4 5 the fundamental foundations for society. 161 Appendix E ATTITUDES ABOUT GENDER ROLES Below are statements about men and women. Read each statement and decide how much you agree or disagree. We are not interested in what society says. We are interested in your personal opinions. For each statement, circle the number that describes your opinion. Please do not omit any statements. Remember to circle only one of the five choices for each statement: 1 2 3 4 5 Strongly Disagree Neutral or Agree Strongly Disagree undecided Agree Strongly Strongly Disagree Agree 1. Women should have as much right as men to go 1 2 3 4 5 to a bar alone. 2. Clubs for students in nursing should admit only 1 2 3 4 5 women. 3. Industrial training schools ought to admit more 1 2 3 4 5 qualified females. 4. Women ought to have the same chances as men 1 2 3 4 5 to be leaders at work. 5. Keeping track of a child?s activities should be 1 2 3 4 5 mostly the mother?s task. 6. Things work out best in a marriage if the husband 1 2 3 4 5 stays away from housekeeping tasks. 7. Both the husband?s and wife?s earnings should be 1 2 3 4 5 controlled by the husband. 8. A woman should not be President of the United 1 2 3 4 5 States. 9. Women should feel as free to ?drop in? on a male 1 2 3 4 5 friend as vice versa. 10. Males should be given first choice to take courses 1 2 3 4 5 that train people as school principals. 11. When both husband and wife work outside the 1 2 3 4 5 home, housework should be equally shared. 12. Women can handle job pressures as well as men 1 2 3 4 5 can. 13. Male managers are more valuable to a business 1 2 3 4 5 than female managers. 14. A woman should have as much right to ask a man 1 2 3 4 5 for a date as a man has to ask a woman for a date. 15. The father, rather than the mother, should give 1 2 3 4 5 teenage children permission to use the family car. 16. Sons and daughters ought to have an equal 1 2 3 4 5 chance for higher education. 162 17. A marriage will be more successful if the 1 2 3 4 5 husband?s needs are considered first. 18. Fathers are better able than mothers to decide 1 2 3 4 5 the amount of a child?s allowance. 19. The mother should be in charge of getting 1 2 3 4 5 children to after-school activities. 20. A person should be more polite to a woman than 1 2 3 4 5 to a man. 21. Women should feel as free as men to express their 1 2 3 4 5 honest opinion. 22. Fathers are not as able to care for their sick 1 2 3 4 5 children as mothers are. 23. An applicant?s sex should be important in job 1 2 3 4 5 screening. 24. Wives are better able than husbands to send 1 2 3 4 5 thank you notes for gifts. 25. Choice of college is not as important for women 1 2 3 4 5 as for men. Please rate the extent to which you agree or disagree with the following additional statements about men and women. 1 2 3 4 5 6 7 Disagree Disagree Disagree Neither Disagree Agree Agree Agree Strongly Somewhat Slightly nor Agree Slightly Somewhat Strongly Disagree Agree Strongly Strongly 1. The initiative in asking for a date should come 1 2 3 4 5 6 7 either from the man or the woman. 2. In a relationship, the woman as well as the man 1 2 3 4 5 6 7 should be free to initiate sexual activity. 3. It is acceptable for a man to have sex with a casual 1 2 3 4 5 6 7 acquaintance. 4. I would have no respect for a man who engages in 1 2 3 4 5 6 7 sexual relationships without any emotional involvement. 5. I would have no respect for a woman who engages 1 2 3 4 5 6 7 in sexual relationships without any emotional involvement. 6. A woman should allow the man to take charge of 1 2 3 4 5 6 7 their sexual relationship. 7. It is acceptable for a woman to have sex with a 1 2 3 4 5 6 7 casual acquaintance. 8. The initiative in dating should come from the man. 1 2 3 4 5 6 7 Comments (optional): 163 Appendix F RELIGIOUS EXPERIENCES Each of the following questions and statements has to do with various issues related to religion and spirituality. Please try to answer all of the questions in this section regardless of how religious you consider yourself to be. Your religious affiliation: How often do you attend a place of worship? ____ Agnostic ____ Not at all ____ Atheist ____ More than once per week ____ Baptist ____ Once per week ____ Catholic ____ Several times per month ____ Episcopalian ____ Once per month ____ Hindu ____ Several times per year ____ Jehovah?s Witness ____ Once every 1-3 years ____ Buddhist ____ Other (Indicate:____________________) ____ Jewish ____ Lutheran ____ Methodist ____ Mormon ____ Muslim ____ Presbyterian ____ Pentecostal ____ Unitarian ____ Evangelical ____ Quaker ____ Seventh Day Adventist ____ Christian Scientist ____ Metropolitan Community Church ____ Not Applicable ____ Other (Indicate:____________________) Overall, how important is religion in your life today? 1 2 3 4 5 Not at all A little bit Moderately Very Extremely important important important important important Overall, regardless of affiliation, how would you describe your level of religious conservatism/liberalism? 1 2 3 4 Very Somewhat Somewhat Very Conservative Conservative Liberal Liberal 164 Think about each of the following items carefully. Does the attitude or behavior described in the statement apply to you? No Not Certain Yes 1. I enjoy reading about my religion. 1 2 3 2. I go to my place of worship because it helps me to make 1 2 3 friends. 3. It doesn?t much matter what I believe so long as I am good. 1 2 3 4. Sometimes I have to ignore my religious beliefs because of 1 2 3 what other people might think of me. 5. It is important to me to spend time in private thought and 1 2 3 prayer. 6. I have often had a strong sense of God?s presence. 1 2 3 7. I pray mainly to gain relief and protection. 1 2 3 8. I try hard to live all my life according to my religious beliefs. 1 2 3 9. What religion offers me most is comfort in times of trouble 1 2 3 and sorrow. 10. My religion is important to me because it answers many 1 2 3 questions about the meaning of life. 11. I would rather join a group that studies the sacred book(s) 1 2 3 of my religion than attend a social group in my place of worship. 12. Prayer is for peace and happiness. 1 2 3 13. Although I am religious, I don?t let if affect my daily life. 1 2 3 14. I go to my place of worship mostly to spend time with my 1 2 3 friends. 15. My whole approach to life is based on my religion. 1 2 3 16. I go to my place of worship mainly because I enjoy seeing 1 2 3 people I know there. 17. I pray mainly because I have been taught to pray. 1 2 3 18. Prayers I say when I am alone are as important to me as 1 2 3 those I say in my place of worship. 19. Although I believe in my religion, many other things are 1 2 3 more important in life. 165 Think about each of the following items carefully. Does the attitude or behavior described in the statement apply to you? No Not Certain Yes 1. I was not very interested in religion until I began to ask 1 2 3 questions about the meaning and purpose of my life. 2. I have been driven to ask religious questions out of a 1 2 3 growing awareness of the tensions in my world and in my relation to my world. 3. My life experiences have led me to rethink my religious 1 2 3 convictions. 4. God wasn?t very important to me until I began to ask 1 2 3 questions about the meaning of my own life. 5. It might be said that I value my religious doubts and 1 2 3 uncertainties. 6. For me, doubting is an important part of what it means 1 2 3 to be religious. 7. I do not find religious doubts upsetting. 1 2 3 8. Questions are more central to my religious experience than 1 2 3 are answers. 9. As I grow and change, I expect my religion also to grow and 1 2 3 change. 10. I am constantly questioning my religious beliefs. 1 2 3 11. I expect my religious convictions to change in the next few 1 2 3 years. 12. There are many religious issues on which my views are still 1 2 3 changing. 166 Appendix G FAMILY AND PARENT-CHILD RELATIONSHIPS The following statements apply to your overall relationships with your family and with your child(ren). Please respond by circling the number with the response that best describes your current family and parent-child relationships in general. In our family: Strongly Strongly Disagree Disagree Agree Agree 1. Planning family activities is difficult because we 1 2 3 4 misunderstand each other. 2. In times of crisis we can turn to each other for support. 1 2 3 4 3. We cannot talk to each other about the sadness we feel. 1 2 3 4 4. Individuals are accepted for who they are. 1 2 3 4 5. We avoid discussing our fears and concerns. 1 2 3 4 6. We can express feelings to each other. 1 2 3 4 7. There are lots of bad feelings in the family. 1 2 3 4 8. We feel accepted for who we are. 1 2 3 4 9. Making decisions is a problem for our family. 1 2 3 4 10. We are able to make decisions about how to solve 1 2 3 4 problems. 11. We don?t get along well together. 1 2 3 4 12. We confide in each other. 1 2 3 4 Please rate your current level of satisfaction with your family in each of the following areas: 1 2 3 4 5 Extremely Dissatisfied Neither Satisfied Extremely dissatisfied satisfied nor satisfied dissatisfied 13. Communication among family members 1 2 3 4 5 14. Level of affection among family members 1 2 3 4 5 15. Level of involvement in each other?s lives 1 2 3 4 5 16. Amount of support for each other 1 2 3 4 5 17. Problem solving in the family 1 2 3 4 5 167 The following statements and questions specifically apply to your relationship with your child who identifies as lesbian or gay. Please respond to each item by circling the number with the response that best describes your current relationship with your child. 18. It?s easy for me to laugh and have a good time with my child. 1 2 3 4 5 Strongly Disagree Neither Agree Strongly Disagree agree nor Agree disagree 19. I feel on edge or tense when I?m with my child. 1 2 3 4 5 Strongly Disagree Neither Agree Strongly Disagree agree nor Agree disagree 20. I would like more influence over my child?s decisions. 1 2 3 4 5 Strongly Disagree Neither Agree Strongly Disagree agree nor Agree disagree 21. During the last 3 months, how often did you argue or fight or have a lot of difficulty with your child? 1 2 3 4 Not at all Infrequently Somewhat Very frequently (once a month frequently (more than once or less) (once every a week) week or two) 22. Taking things all together, how would you describe your relationship with your child? 0 1 2 3 4 5 6 7 8 9 10 Really Absolutely Bad Perfect Please indicate to what extent (if any) your relationship with your child has changed in each of the following areas as a result of your child coming out as gay, lesbian, or same-sex oriented: 1 2 3 4 5 Has changed for Has not changed Has changed for the worse at all the better 23. Enjoyment/Closeness 1 2 3 4 5 24. Tension 1 2 3 4 5 25. Level of involvement in child?s life 1 2 3 4 5 26. Fights/arguments 1 2 3 4 5 27. Overall relationship 1 2 3 4 5 168 The next two sets of questions relate to how PFLAG involvement has influenced your family in general and your relationship with your lesbian/gay child in particular. Please use the following scale to answer each question: 1 2 3 4 5 No influence Minor Some Significant Very strong influence influence influence influence Please indicate to what extent your involvement in PFLAG has influenced your family in general in each of the following areas: 28. Communication among family members 1 2 3 4 5 29. Level of affection among family members 1 2 3 4 5 30. Level of involvement in each other?s lives 1 2 3 4 5 31. Amount of support for each other 1 2 3 4 5 32. Problem solving in the family 1 2 3 4 5 Please indicate to what extent your involvement in PFLAG has influenced your current relationship with your lesbian/gay child in each of the following areas: 33. Enjoyment/Closeness 1 2 3 4 5 34. Tension 1 2 3 4 5 35. Level of involvement in child?s life 1 2 3 4 5 36. Fights/arguments 1 2 3 4 5 37. Overall relationship 1 2 3 4 5 Comments (optional): 169 Appendix H Cover Letter Dear Participant: Thank you for your willingness to participate in this study. This study examines the experiences of parents with a child who identifies as lesbian, gay, or oriented toward individuals of the same sex. The study is being conducted by myself, an advanced level doctoral student at the University of Maryland at College Park under the supervision of Dr. Ruth Fassinger and in cooperation with Dr. Roy Gilbert-Higginson and the PFLAG National Office. In this study, we are interested in learning about the personal experiences of parents who have learned that their child identifies as lesbian, gay, or oriented toward individuals of the same sex. We understand that sexual identity and sexual orientation are complex and include a wide variation. For the purposes of this study, we are interested in parents with children who have specifically declared themselves as gay, lesbian, or attracted to persons of the same sex. We hope that results will be used to address the needs of parents who have a gay or lesbian child and to develop interventions that can help parents and their families dealing with issues related to having a gay or lesbian child. We hope to survey as large and diverse a group of participants as possible. We need your help in completing our survey, which will take approximately 45 minutes to complete. This study was approved by the university Institutional Review Board, and your responses will be treated anonymously. There are no known risks associated with participation in this study. The benefits of this study are not intended to help you personally, but rather to help the researchers learn more about how parents and their relationships with their children are influenced by their child?s coming out as lesbian or gay. By completing and returning this questionnaire in the self-addressed, stamped envelope provided, you will indicate consent to participate and agree that: Your participation in this research is voluntary You are not required to answer every question that might be asked You are free to stop participating at any point without penalty At the end of participation in this study, you may have your responses withdrawn from the study by contacting the researchers (see contact information below) If you are interested in learning the results of this study after its completion, please retain the following contact information to request a copy of the results. Results should be available in the Fall of 2005. Also, please feel free to contact the researchers at any time if you have any questions or concerns about this study. Thank you for your time and consideration. Susanna M. Gallor Ruth E. Fassinger sgallor@wam.umd.edu rf36@umail.umd.edu University of Maryland, College Park University of Maryland, College Park 170 References Allport, G.W., & Ross, J.M. (1967). Personal religious orientation and prejudice. Journal of Personality and Social Psychology, 5, 432-443. Altemeyer, B., & Hunsberger, B. (1992). Authoritarianism, religious fundamentalism, quest, and prejudice. International Journal for the Psychology of Religion, 2, 113 133. American Psychological Association. (2000). Guidelines for psychotherapy with lesbian, gay,and bisexual clients. Washington, DC: American Psychological Association. Anselmi, D.L., & Law, A.L. (1998). Questions of Gender: Perspectives and Paradoxes. New York: McGraw-Hill. Aquilino, W.S. (1997). From adolescent to young adult: A prospective study of parent child relations during the transition to adulthood. Journal of Marriage & the Family, 59, 670-686. Aquilino, W.S. (1999). Two views of one relationship: Comparing parents? and young adult children?s reports of the quality of intergenerational relations. Journal of Marriage & the Family, 61, 858-870. Ashmore, R.D. (1990). Sex, gender, and the individual. In L.A. Pervin (Ed.), Handbook of personality theory and research (pp. 486-526). New York: Guilford Press. Ashmore, R.D., Del Boca, F.K., & Bilder, S.M. (1995). Construction and validation of the Gender Attitude Inventory, a structured inventory to assess multiple dimensions of gender attitudes. Sex Roles, 32, 753-785. Baker, M., & Gorsuch, R. (1982). Trait anxiety and intrinsic-extrinsic religiousness. Journal for the Scientific Study of Religion, 21, 119-122. 171 Batson, C.D. (1976). Religion as prosocial: Agent or double agent? Journal for the Scientific Study of Religion, 15, 29-45. Batson, C.D., Flink, C.H., Schoenrade, P.A., Fultz, J., & Pych, V. (1986). Religious orientation and overt and covert racial prejudice. Journal of Personality and Social Psychology, 50, 175-181. Batson, C.D., & Schoenrade, P. (1991a). Measuring religion as Quest: 1) Validity concerns. Journal for the Scientfic Study of Religion, 30, 416-429. Batson, C.D., & Schoenrade, P. (1991b). Measuring religion as Quest: 2) Reliability concerns. Journal for the Scientific Study of Religion, 30, 430-447. Batson, C.D., Schoenrade, P.A., & Ventis, W.L. (1993). Religion and the individual: A social-psychological perspective. New York: Oxford University Press. Batson, C.D., & Ventis, W.L. (1982). The religious experience: A social-psychological perspective. New York: Oxford University Press. Beeler, J., & DiProva, V. (1999). Family adjustment following disclosure of homosexuality by a member: Themes discerned in narrative accounts. Journal of Marital and Family Therapy, 25, 443-459. Beere, C.A., King, D.W., Beere, D.B., & King. L.A. (1984). The Sex-Role Egalitarianism Scale: A measure of attitudes toward equality between the sexes. Sex Roles, 10, 563- 576. Bem, S.L. (1974). The measurement of psychological androgyny. Journal of Consulting and Clinical Psychology, 42, 155-162. Bem, S.L. (1981). Gender schema theory: A cognitive account of sex typing. Psychological Review, 88, 354-364. 172 Ben-Ari, A. (1995). The discovery that an offspring is gay: Parents?, gay men?s, and lesbians? perspectives. Journal of Homosexuality, 30, 89-112. Billingham, P., & King, L.A. (1991). Sex-role egalitarianism and heterosexual violence. Paper presented at the annual meeting of the Midwestern Psychological Association, Chicago. Blumstein, P., & Schwartz, P. (1998). Intimate relationships and the creation of sexuality. In D.L. Anselmi & A.L. Law (Eds.), Questions of gender: Perspectives and paradoxes (pp. 345-354). New York: McGraw-Hill. Bohan, J.S. (1996). Psychology and sexual orientation coming to terms (pp. 31-59). New York: Routledge. Boxer, A.M., Cook, J.A., & Herdt, G. (1991). Double jeopardy: Identity transitions and parent-child relations among gay and lesbian youth. In K.A. Pillemer & K. McCartney (Eds.), Parent-child relations throughout life (pp. 59-92). Hillsdale, NJ: Lawrence Erlbaum Associates. Broido, E.M. (1999). Ways of being an ally to lesbian, gay, and bisexual students. In V.A. Wall & N.J. Evans (Eds.), Toward Acceptance: Sexual Orientation Issues On Campus (pp. 345-369). Lanham, MD: University Press of America. Brown, L.S. (1997). New voices, new visions: Toward a lesbian/gay paradigm for psychology. In M.M. Gergen & S.N. Davis (Eds.), Toward a new psychology of gender (pp. 295-308). New York: Routledge. Brown, S.L., Nesse, R.M., Vinokur, A.D., & Smith, D.M. (2003). Providing social support may be more beneficial than receiving it: Results from a prospective study of mortality. Psychological Science, 14, 320-327. 173 Cash, T.F., Ancis, J.R., & Strachan, M.D. (1997). Gender attitudes, feminist identity, and body images among college women. Sex Roles, 36, 433-447. Cass, V.C. (1979). Homosexual identity formation: A theoretical model. Journal of Homosexuality, 4, 219-235. Cass, V.C. (1996). Sexual orientation identity formation: A western phenomenon. In R.P. Cabaj & T.S. Stein (Eds.), Textbook of homosexuality and mental health (pp. 227- 251). Washington, D.C.: American Psychiatric Press. Chan, C.S. (1989). Issues of identity development among Asian-American lesbians and gay men. Journal of Counseling and Development, 68, 16-21. Chao, R.K. (1994). Beyond parental control and authoritarian parenting style: Understanding Chinese parenting through the cultural notion of training. Child Development, 65, 1111-1119. Chesler, M.A. (1991). Participatory action research with self-help groups: An alternative paradigm for inquiry and action. American Journal of Community Psychology, 19, 757-768. Clark, J.M., Brown, J.C., & Hochstein, L.M. (1990). Institutional religion and gay/lesbian oppression. In F.W. Bozett & M.B. Sussman (Eds.), Homosexuality and family relations (pp. 265-284). New York: Haworth Press. Cohen, K.M., & Savin-Williams, R.C. (1996). Developmental perspectives on coming out to self and others. In R.C. Savin-Williams & K.M. Cohen (Eds.), The Lives of Lesbians, Gays, and Bisexuals: Children to Adults (pp. 113-151). New York: Harcourt Brace. Coopersmith, S. (1982). Self-esteem inventories. Palo Alto, CA: The Consulting 174 Psychologists Press. Cramer, D.W., & Roach, A.J. (1988). Coming out to mom and dad: A study of gay males and their relationships with their parents. Journal of Homosexuality, 15, 79-91. Crandall, J.E., & Rasmussen, R.D. (1975). Purpose in life as related to specific values. Journal of Clinical Psychology, 31, 483-485. Dahlheimer, D., & Feigal, J. (1994). Community as family: The multiple-family contexts of gay and lesbian clients. In C.H. Huber (Ed.), Transitioning from individual to family counseling (pp. 63-74). Alexandria, VA: American Counseling Association. D?Augelli, A.R., Hershberger, S.L., & Pilkington, N.W. (1998). Lesbian, gay, and bisexual youth and their families: Disclosure of sexual orientation and its consequences. American Journal of Orthopsychiatry, 68, 361-371. Davidson, M.G. (2000). Religion and spirituality. In R.M. Perez, K.A. DeBord, & K.J. Bieschke (Eds.), Handbook of counseling and psychotherapy with lesbian, gay, and bisexual clients (pp. 409-433). Washington, DC: American Psychological Association. Deaux, K. (1999). An overview of research on gender: Four themes from 3 decades. In W.B. Swann, J.H. Langlois, & L.A. Gilbert (Eds.), Sexism and stereotypes in modern society (pp. 11-33). Washington, DC: American Psychological Association. Deaux, K., & Kite, M.E. (1987). Thinking about gender. In B.B. Hess & M.M. Ferree (Eds.), Analyzing gender: A handbook of social science research (pp. 92-117). Newbury Park, CA: Sage. Deaux, K., & Major, B. (1998). A social-psychological model of gender. In D.L. Anselmi & A.L. Law (Eds.), Questions of gender: Perspectives and paradoxes (pp. 367-375). 175 New York: McGraw-Hill. DeVine, J.L. (1984). A systematic inspection of affectional preference orientation and the family of origin. Journal of Social Work & Human Sexuality, 2, 9-17. Dixon, L., Adams, C., & Lucksted, A. (2000). Update on family psychoeducation for schizophrenia. Schizophrenia Bulletin, 26(1), 5-20. Dube, E.M., & Savin-Williams (1999). Sexual identity development among ethnic sexual-minority male youths. Developmental Psychology, 35, 1389-1396. Epstein, N.B., Baldwin, L.M., & Bishop, D.S. (1983). The McMaster Family Assessment Device. Journal of Marital and Family Therapy, 9, 171-180. Fassinger, R.E. (1987). Use of structural equation modeling in counseling psychology research. Journal of Counseling Psychology, 34, 425-436. Fernald, J.L. (1995). Interpersonal heterosexism. In B. Lott & D. Maluso (Eds.), The social psychology of interpersonal discrimination (pp. 80-117). New York: Guilford Press. Floyd, F.J., Stein, T.S., Harter, K.S.M., Allison, A., & Nye, C.L. (1999). Gay, lesbian, and bisexual youths: Separation-individuation, parental attitudes, identity consolidation, and well-being. Journal of Youth and Adolescence, 28, 719- 739. Francis, L.J., & Stubbs, M.T. (1987). Measuring attitudes toward Christianity: From childhood to adulthood. Personality and Individual Differences, 8, 710-743. Frank, S.J., Avery, C.B., & Laman, M.S. (1988). Young adults? perceptions of their relationships with their parents: Individual differences in connectedness, competence, and emotional autonomy. Developmental Psychology, 24, 729-737. 176 Fulton, A.S., Maynard, E.A., & Gorsuch, R.L. (1999). Religious orientation, antihomosexual sentiment, and fundamentalism among Christians. Journal for the Scientific Study of Religion, 38(1), 14-22. Garcia, C.T., Meyer, E.C., & Brillon, L. (1995). Ethnic and minority parenting. In M.H. Bornstein (Ed.), Handbook of parenting, Vol. 2: Biology and ecology of parenting (pp. 189-209). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc. Gartner, A., Gartner, A.J., & Ouellette-Kobasa, S.C. (1988). Self-help. In E.A. Blechman & K.D. Brownell (Eds.), Handbook of Behavioral Medicine for Women (pp. 330 342). New York: Pergamon Press. Gelso, C.J., & Fretz, B.R. (1992). Counseling psychology. San Diego, CA: Harcourt Brace Jovanovich. Genia, V., & Shaw, D.G. (1991). Religion, intrinsic-extrinsic orientation, and depression. Review of Religious Research, 32, 274-283. Goldfried, M.R., & Goldfried, A.P. (2001). The importance of parental support in the lives of gay, lesbian, and bisexual individuals. Journal of Clinical Psychology, 57, 681-693. Gorsuch, R.L., & McPherson, S.E. (1989). Intrinsic/extrinsic measurement: I/E-revised and single-item scales. Journal for the Scientific Study of Religion, 28, 348-354. Gorsuch, R.L., & Venable, G.D. (1983). Development of an ?Age Universal? I-E Scale. Journal for the Scientific Study of Religion, 22, 181-187. Greene, B. (2003). Beyond heterosexism and across the cultural divide ? developing an inclusive lesbian, gay, and bisexual psychology: A look to the future. In L.D. Garnets & D.C. Kimmel (Eds.), Psychological perspectives on lesbian, gay, and 177 bisexual experiences (pp. 357-400). New York: Columbia University Press. Hall, T.W., Tisdale, T.C., & Brokaw, B.F. (1994). Assessment of religious dimensions in Christian clients: A review of selected instruments for research and clinical use. Journal of Psychology and Theology, 22, 395-421. Helmreich, R.L., & Stapp, J. (1974). Short forms of the Texas Social Behavior Inventory (TSBI), an objective measure of self-esteem. Bulletin of Psychonomic Society, 4(5A), 473-475. Herdt, G., & Boxer, A. (1993). Children of horizons: How gay and lesbian teens are leading a new way out of the closet. Boston: Beacon Press. Herdt, G., & Koff, B. (2000). Something to tell you: The road families travel when a child is gay. New York: Columbia University Press. Herek, G. (1984b). Attitudes toward lesbians and gay men: A factor analytic study. Journal of Homosexuality, 10, 39-51. Herek, G.M. (1986). On heterosexual masculinity: Some psychical consequences of the social construction of gender and sexuality. American Behavioral Scientist, 29, 563 577. Herek, G.M. (1987). Religious orientation and prejudice: A comparison of racial and sexual attitudes. Personality and Social Psychology Bulletin, 13(1), 34-44. Herek, G.M. (1988). Heterosexuals? attitudes toward lesbians and gay men: Correlates and gender differences. Journal of Sex Research, 25, 451-477. Herek, G.M. (1990). The context of anti-gay violence. Journal of Interpersonal Violence, 5, 316-333. Herek, G.M. (1994). Assessing heterosexuals? attitudes toward lesbians and gay men: A 178 review of empirical research with the ATLG scale. In B. Greene & G.M. Herek (Eds.), Lesbian and gay psychology: Theory, research, and clinical applications (pp. 206-228). Thousand Oaks: Sage Publications. Herek, G.M. (1995). Psychological heterosexism in the United States. In A.R. D?Augelli & C.J. Patterson (Eds.), Lesbian, gay, and bisexual identities over the lifespan: Psychological perspectives (pp. 321-346). New York: Oxford University Press. Herek, G.M. (1996). Heterosexism and homophobia. In R.P. Cabaj & T.S. Stein (Eds.), Textbook of homosexuality and mental health (pp. 101-113). Washington, DC: American Psychiatric Press. 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, 412-424. Herek, G.M., & Capitanio, J.P. (1999). Sex differences in how heterosexuals think about lesbians and gay men: Evidence from survey context effects. Journal of Sex Research, 36, 348-360. Herek, G.M., & Glunt, E.K. (1993b). Interpersonal contact and heterosexuals? attitudes toward gay men: Results from a national survey. Journal of Sex Research, 30, 239 244. Hoffman, J.A. (1984). Psychological separation of late adolescents from their parents. Journal of Counseling Psychology, 31, 170-178. Holtzen, D.W., & Agresti, A.A. (1990). Parental responses to gay and lesbian children: Differences in homophobia, self-esteem, and sex-role stereotyping. Journal of Social and Clinical Psychology, 9, 390-399. 179 Hort, B.E., Fagot, B.I., & Leinbach, M.D. (1990). Are people?s notions of maleness more stereotypically framed than their notions of femaleness? Sex Roles, 23, 197-212. Hudson, W.W., & Ricketts, W.A. (1980). A strategy for the measurement of homophobia. Journal of Homosexuality, 5, 357-372. Hunsberger, B. (1996). Religious fundamentalism, right-wing authoritarianism, and hostility toward homosexuals in non-Christian religious groups. The International Journal for the Psychology of Religion, 6(1), 39-49. Iasenza, S., Colucci, P.L., & Rothberg, B. (1996). Coming out and the mother-daughter bond: Two case examples. In J. Laird & R.J. Green (Eds.), Lesbians and gays in couples and families (pp. 123-136). San Francisco: Jossey-Bass. Jackson, L.A., & Cash, T.F. (1985). Components of gender stereotypes: Their implications for inferences on stereotypic and nonstereotypic dimensions. Personality and Social Psychology Bulletin, 11, 326-344. Jackson, L.A., Fleury, R.E., & Lewandowski, D.A. (1996). Feminism: Definitions, support, and correlates of support among female and male college students. Sex Roles, 34, 687-693. Jackson, B., & Hardiman, R. (1982). Social identity development model. Unpublished manuscript. Jensen, L.D., Gambles, D., & Olsen, J. (1988). Attitudes toward homosexuality: A cross- cultural analysis of predictors. International Journal of Social Psychiatry, 34, 47-57. Kahoe, R.D. (1974). Personality and achievement correlates of instrinsic and extrinsic religious orientations. Journal of Personality and Social Psychology, 29, 812-818. Kalin, R. (1979). Method for scoring androgyny as a continuous variable. Psychological 180 Reports, 44, 1205-1206. Kerns, J.G., & Fine, M.A. (1994). The relation between gender and negative attitudes toward gay men and lesbians: Do gender role attitudes mediate this relation? Sex Roles, 31, 297-307. King, L.A., & King, D.W. (1993). Sex-Role Egalitarianism Scale Manual. Michigan: Sigma Assessment Systems. King, L.A., & King, D.W. (1997). Sex-Role Egalitarianism Scale: Development, psychometric properties, and recommendations for future research. Psychology of Women Quarterly, 21, 71-87. Kingree, J.B., & Ruback, R.B. (1994). Understanding self-help groups. In T.J. Powell (Ed.), Understanding the Self-Help Organization: Frameworks and Findings (pp. 272-292). Thousand Oaks: SAGE Publications. Kinsey, A., Pomeroy, W., & Martin, C. (1948). Sexual behavior in the human male. Philadelphia: W.B. Saunders. Kirkpatrick, L.A. (1989). A psychometric analysis of the Allport-Ross and Feagin measures of intrinsic-extrinsic religious orientation. In D.O. Moberg & M.L. Lynn (Eds.), Research in the social scientific study of religion (Vol. 1, pp. 1-31). Greenwich, CT: JAI Press. Kite, M.E. (1994). When perceptions meet reality: Individual differences in reactions to lesbians and gay men. In B. Greene & G.M. Herek (Eds.), Lesbian and gay psychology: Theory, research, and clinical applications (pp. 25-53). Thousand Oaks: Sage Publications. Kite, M.E., & Deaux, K. (1987). Gender belief systems: Homosexuality and the implicit 181 inversion theory. Psychology of Women Quarterly, 11, 83-96. Kite, M.E., & Whitley, B.E. (1996). Sex differences in attitudes toward homosexual persons, behavior, and civil rights: A meta-analysis. Personality and Social Psychology Bulletin, 22, 336-353. Kite, M.E., & Whitley, B.E. (1998). Do heterosexual women and men differ in their attitudes toward homosexuality? A conceptual and methodological analysis. In G.M. Herek (Ed.), Stigma and sexual orientation: Understanding prejudice against lesbians, gay men, and bisexuals (pp. 39-61). Thousand Oaks, CA: Sage Publications. Klaw, E., & Humphreys, K. (2004). The role of peer-led mutual help groups in promoting health and well-being. In J.L. DeLucia-Waack, D.A. Gerrity, C.R. Kalodner, & M.T. Riva (Eds.), Handbook of Group Counseling and Psychotherapy (pp. 630-640). Thousand Oaks, CA: Sage Publications. Kojentin, B.A. (1988). Separating the seekers from the doubters: Quest, self consciousness, and religious experience. Paper presented at the annual convention of the Society for the Scientific Study of Religion, Chicago. Kojentin, B.A., McIntosh, D.N., Bridges, R.A., & Spilka, B. (1987). Quest: Constructive search or religious conflict? Journal for the Scientific Study of Religion, 26, 111-115. Kubler-Ross, E. (1969). On death and dying. New York: MacMillan. Kurdek, L.A. (1988). Correlates of negative attitudes toward homosexuals in heterosexual college students. Sex Roles, 18, 727-738. Laird, J. (1996). Invisible ties: Lesbians and their families of origin. In J. Laird & R.J. Green (Eds.), Lesbians and gays in couples and families (pp. 89-122). San Francisco, 182 CA: Jossey-Bass. Laner, M.R., & Laner, R.H. (1980). Sexual preference or personal style? Why lesbians are disliked. Journal of Homosexuality, 5, 339-356. Larson, K.S., Cate, R., & Reed, M. (1983). Anti-Black attitudes, religious orthodoxy, permissiveness, and sexual information: A study of the attitudes of heterosexuals toward homosexuality. The Journal of Sex Research, 19, 105-118. Lieberman, M.A., & Snowden, L.R. (1994). Problems in assessing prevalence and membership characteristics of self-help group participants. In T.J. Powell (Ed.), Understanding the Self-Help Organization: Frameworks and Findings (pp. 32-49). Thousand Oaks: SAGE Publications. Lipsey, M.W., & Wilson, D.B. (1993). The efficacy of psychological, educational, and behavioral treatment: Confirmation from meta-analysis. American Psychologist, 48, 1181-1209. Lomax, R.G. (1998). Statistical Concepts: A Second Course for Education and the Behavioral Sciences. New Jersey: Lawrence Erlbaum Associates. Lottes, I.L., & Kuriloff, P.J. (1992). The effects of gender, race, religion, and political orientation on the sex role attitudes of college freshman. Adolescence, 27, 675-688. 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. Maltby, J., & Day, L. (1998). Amending a measure of the Quest Religious Orientation: Applicability of the scale?s use among religious and non-religious persons. Personality and Individual Differences, 25, 517-522. 183 Maltby, J., & Lewis, C.A. (1996). Measuring intrinsic and extrinsic orientation toward religion: Amendments for its use among religious and non-religious samples. Personality and Individual Differences, 21, 937-946. Maton, K., Leventhal, G., Madara, E., & Julien, M. (1989). The birth and death of mutual-help groups: The role of national affiliation, professional involvement, and member focal problem. American Journal of Community Psychology, 17, 643-671. Matthews, C.R., & Lease, S.H. (2000). Focus on lesbian, gay, and bisexual families. In R.M. Perez, K.A. DeBord, & K.J. Bieschke (Eds.), Handbook of counseling and psychotherapy with lesbian, gay, and bisexual clients (pp. 249-273). Washington, DC: American Psychological Association. Mattison, A.M., & McWhirter, D.P. (1995). Lesbians, gay men, and their families: Some therapeutic issues. Psychiatric Clinics of North American, 18, 123-137. McFarland, S.G. (1989). Religious orientations and the targets of discrimination. The Journal for the Scientific Study of Religion, 28, 324-336. McIntosh, P. (1988). White privilege: Unpacking the invisible knapsack. Peace and Freedom (1988, July/August), 10-12. Merighi, J.R., & Grimes, M.D. (2000). Coming out to families in a multicultural context. Families in Society, 81(1), 32-41. Miller, I.W., Epstein, N.B., Bishop, D.S., & Keitner, G.I. (1985). The McMaster Family Assessment Device: Reliability and validity. Journal of Marital and Family Therapy, 11, 345-356. Minton, B., & Spilka, B. (1976). Perspectives on death in relation to powerlessness and form of personal religion. Journal of Death and Dying, 7, 261-268. 184 Newman, B.S., & Muzzonigro, P.G. (1993). The effects of traditional family values on the coming out process of gay male adolescents. Adolescence, 28, 213-226. Oswald, R.F. (2002). Resilience within the family networks of lesbians and gay men: Intentionality and redefinition. Journal of Marriage & the Family, 64, 374-384. Park, J. (2004). Development of the Heterosexism Scale. Manuscript submitted for publication. Patterson, C.J. (2000). Family relationships of lesbians and gay men. Journal of Marriage and the Family, 62, 1052-1069. Pomeroy, E.C., Rubin, A., & Walter, R.J. (1996). A psychoeducational group intervention for family members of persons with HIV/AIDS. Family Process, 35, 299-312. Putney, S., & Middleton, R. (1961). Dimensions and correlates of religious ideologies. Social Forces, 39, 285-290. Raja, S., & Stokes, J. P. (1998). Assessing attitudes toward lesbians and gay men: The Modern Homophobia Scale. Journal of Gay, Lesbian, and Bisexual Identity, 3, 113- 134. Roberts, L.J., Salem, D., Rappaport, J., Toro, P., Luke, D., & Seidman, E. (1999). Giving and receiving help: Interpersonal transactions in mutual help meetings and psychosocial adjustment of members. American Journal of Community Psychology, 27, 841-868. Robinson, B.E., Walters, L.H., & Skeen, P. (1989). Response of parents to learning that their child is homosexual and concern over AIDS: A national study. Journal of Homosexuality, 18(1-2), 59-80. 185 Rust, P.C. (2003). Finding a sexual identity and community: Therapeutic implications and cultural assumptions in scientific models of coming out. In L.D. Garnets & D.C. Kimmel (Eds.), Psychological perspectives on lesbian, gay, and bisexual experiences (pp. 227-269). New York: Columbia University Press. Saltzburg, S. (2004). Learning that an adolescent child is gay or lesbian: The parent experience. Social Work, 49, 109-118. Sandnabba, N.K., & Ahlberg, C. (1999). Parents? attitudes and expectations about children?s cross-gender behavior. Sex Roles, 40, 249-263. Savin-Williams, R.C. (1990). Gay and lesbian youth: expressions of identity. New York: Hemisphere. Savin-Williams, R.C. (1996). Ethnic- and sexual-minority youth. In R.C. Savin-Williams & K. M. Cohen (Eds.), The lives of lesbians, gays, and bisexuals: Children to adults (pp. 152-165). Ft. Worth, TX: Harcourt Brace College Publishers. Savin-Williams, R.C. (1998). Lesbian, gay, and bisexual youths? relationships with their parents. In C.J. Patterson & A.R. D?Augelli (Eds.), Lesbian, gay, and bisexual identities in families (pp. 75-98). New York: Oxford University Press. Savin-Williams, R.C. (2001). Mom, dad. I?m gay. How families negotiate coming out. Washington, DC: American Psychological Association. Savin-Williams, R.C. (2003). Lesbian, gay, and bisexual youths? relationships with their parents. In L.D. Garnets & D.C. Kimmel (Eds.), Psychological perspectives on lesbian, gay, and bisexual experiences (pp. 299-326). New York: Columbia University Press. Savin-Williams, R.C., & Dube, E.M. (1998). Parental reactions to their child?s disclosure 186 of a gay/lesbian identity. Family Relations, 47, 7-13. Savin-Williams, R.C., & Ream, G.L. (2003). Sex variations in the disclosure to parents of same-sex attractions. Journal of Family Psychology, 17, 429-438. Schaefer, C.A., & Gorsuch, R.L. (1992). Dimensionality of religion: Belief and motivation as predictors of behavior. Journal of Psychology and Christianity, 11, 244-254. Schreier, B.A., & Werden, D.L. (2000). Psychoeducational programming: Creating a context of mental health for people who are lesbian, gay, or bisexual. In R.M. Perez, K.A. DeBord, & K.J. Bieschke (Eds.), Handbook of counseling and psychotherapy with lesbian, gay, and bisexual clients (pp. 359-382). Washington, DC: American Psychological Association. Schubert, M.A., & Borkman, T.J. (1991). An organizational typology for self-help groups. American Journal of Community Psychology, 19, 769-787. Sciarra, D.T., & Gushue, G.V. (2003). White racial identity development and religious orientation. Journal of Counseling and Development, 81, 473-482. Simoni, J.M., & Walters, K.L. (2001). Heterosexual identity and heterosexism: Recognizing privilege to reduce prejudice. Journal of Homosexuality, 41, 157-172. Snowden, L.R., & Lieberman, M.A. (1994). African-American participation in self-help groups. In T.J. Powell (Ed.), Understanding the Self-Help Organization: Frameworks and Findings (pp. 50-62). Thousand Oaks: SAGE Publications. Spence, J.T. (1985). Gender identification and its implications for the concepts of masculinity and femininity. In R.A. Dienstbier (Series Ed.) & T.B. Sonderegger (Ed.), Nebraska symposium on motivation and achievement. Vol. 32: Psychology and 187 gender (pp. 59-95). Lincoln: University of Nebraska Press. Spence, J.T. (1993). Gender-related traits and gender ideology: Evidence for a multifactorial theory. Journal of Personality and Social Psychology, 64, 624-635. Spence, J.T., & Helmreich, R.L. (1972). The Attitudes Toward Women Scale: An objective instrument to measure the attitudes toward the rights and roles of women in contemporary society. JSAS: Catalog of Selected Documents in Psychology, 2, 66- 67. Strommen, E.F. (1989). ?You?re a what??: Family member reactions to the disclosure of homosexuality. Journal of Homosexuality, 18, 37-58. Strommen, E.F. (1990). Hidden branches and growing pains: Homosexuality and the family tree. In F.W. Bozett & M.B. Sussman (Eds.), Homosexuality and family relations (pp. 9-34). New York: Harrington Park Press. U.S. Department of Health and Human Services. (1987). Surgeon-General?s workshop on self-help and public health. Washington, DC: U.S. Government Printing Office. Walsh, F. (1998). Strengthening family resilience. New York: Guilford Press. Watson, P.J., Morris, R.J., & Hood, R.W. (1990). Attributional complexity, religious orientation, and indiscriminate proreligiousness. Review of Religious Research, 32, 110-121. Whitley, B.E. (2001). Gender-role variables and attitudes toward homosexuality. Sex Roles, 45, 691-721. Winefield, H., Barlow, J., & Harvey, E. (1998). Responses to support groups for family caregivers in schizophrenia: Who benefits from what? Australian and New Zealand Journal of Mental Health Nursing, 7, 103-110.