This descriptive, exploratory study qualitatively examined parent-child relationships following the discovery that a daughter or son is gay.
My research assistants and I qualitatively interviewed a multicultural sample of sixty-five self-identified gay youth (aged fourteen to twenty-five) and seventy-six of their parents. (It should be noted that during their interviews two of these young women identified as bisexual and two others as transgender—see chapter 7). In order to be eligible for the study, the son or daughter and at least one parent needed to agree to participate.
Respondents were recruited primarily from the New York City and Philadelphia metropolitan areas, including northern, central, and southern New Jersey. Families were recruited via advertisements in local newspapers (n = 21) and on Craigslist (n = 23), a Web site of electronic classified advertisements. Fifteen families were recruited from Parents, Families, and Friends of Lesbians and Gays (PFLAG), which is a support group for parents of gays and lesbians. The remaining six families were recruited through high school and community service organizations that ran support and social groups for gay youth. I would give talks at PFLAG meetings and community organizations about coming out to parents and, at the end of each talk, I would describe the study and distribute flyers, asking the audience members to consider volunteering.
Interested potential respondents were instructed to contact me to schedule an interview. Parents were asked to check with their children to be sure they were willing to be interviewed. If young gay men or women contacted me asking to participate, they were instructed to have their parent call or e-mail me to arrange an interview. This was to ensure that parents were indeed willing to participate. The interviews were done in person and by telephone from the fall of 2003 until the spring of 2005.
Thirty-five of the youth were female and thirty were male. Forty-one of the youth were white, seventeen were black or biracial, six were Latino, and one was Japanese American, with parents born in Japan. Their ages ranged from fourteen through twenty-five, but only four were under eighteen. Their mean age was twenty-one. The time they were out to their parents ranged from 6 months to 9 years with a mean of 3.8 years; 61.5 percent of the youth described experiencing mental health symptoms of anxiety or depression before coming out.
Fifty-nine mothers and seventeen fathers participated in the study. Fifty-four parents were white, sixteen were black, five were Latino, and one mother was Japanese. The parents’ ages ranged from thirty-four to sixty-nine, with a mean age of fifty-one; 53 percent reported symptoms of anxiety and or depression after discovering that their son or daughter was gay or lesbian and 44 percent reported that religion was important or very important in their lives. Families came from a variety of socioeconomic backgrounds and had household incomes ranging from over $1,200 to over $200,000 per year.
For the most part, the parents were fairly adjusted to their children’s sexual orientation by the time of the interviews, so it is important to recognize the external validity limitations of the sample. Because an unknown proportion of the population of lesbian and gay youth do not publicly identify themselves, investigators in this area lack sufficient information to determine and derive a truly representative sample. Nevertheless, it is important to recognize that the goal of this study was not necessarily to discover findings that could generalize to the larger population, but to develop data-based theory about family adjustment that would be useful to practitioners.
This research was approved for human subjects by the Rutgers University Office of Research and Sponsored Programs, Institutional Review Board, in March of 2003.
A combination of grounded theory (Glaser 1978, 1992) and narrative methods (Lincoln and Guba 1985) were used to collect the data. Interviewers were carefully trained to capture the respondents’ true voices and to avoid influencing their interviewees’ responses. They were also trained to be vigilant for social desirability effects. The interviewers (myself included) continually reassured respondents that we wanted participants to be truthful and that they would not be negatively judged for any of their responses, no matter how harsh. Parents and their children were interviewed separately because it was believed that individual respondents might edit their responses in front of family members. Interviews were done mostly in person, but occasionally an interview was done over the telephone if a child or parent lived out of town or was away at school.
Below are the primary research questions, followed by some of the interview questions:
1. How do parents react upon discovery that a son or daughter is gay or lesbian, and what factors are related to changes in their reactions?
Parents were asked to describe parent-child relationships historically and immediately prior to the discovery. They were asked how they learned of their child’s sexual orientation and about their behavioral and emotional reactions, including whether they experienced mental health symptoms. They were also queried regarding any changes immediately following and up to a year after the discovery. In addition, their lesbian daughters or gay sons were also asked for their perceptions of parental reactions.
2. What effects do parents’ reactions have on their gay and lesbian children?
Gay and lesbian respondents were asked the specific effects of parental reactions on their self-esteem and whether they experienced mental health symptoms after their parents learned of their sexual orientation. They were asked to describe their responses to their parents’ reactions so that specific interactions could be identified. They were also asked to describe changes in their parental relationships immediately after they came out and up until the time of their interviews.
3. What helps or hinders parental adjustment?
Parents were asked how their attitudes and behaviors changed since their initial discovery. They were asked about factors that enhanced or interfered with their adjustment, including child-related factors. Children were also asked for their impressions of the changes in their parents’ reactions since the initial discovery and what factors influenced these reactions.
Probes were used to elicit factors that facilitated or hindered parental adjustment. Each interview lasted anywhere from ninety minutes to two hours.
Two research assistants assisted with the interviews: Bethann Albert is a white social worker who is the mother of a lesbian adult; Rita D. Velez Carreras is a bilingual, native Puerto Rican lesbian woman, who was twenty-five at the time of the interviews and had recently graduated from college with a degree in women’s studies.
Respondents initially were paid twenty dollars per interview. However, halfway through the data collection this compensation was raised to forty dollars, and this increase served to attract more respondents from lower socioeconomic backgrounds including working-class and poor African American families.
All interviews were audiotaped, transcribed, and analyzed using a combination of cross-case analysis and grounded theory methods (Padgett 2008; Corbin and Strauss 2008). Using cross-case analysis, transcriptions of answers to the same question were grouped together and the texts of their responses were analyzed. In keeping with grounded theory methods, these grouped responses were read, and preliminary codes were developed. Through constant comparative analysis, axial codes led to the primary themes, which included the family adjustment stages, the importance of strong parent-child boundaries, and the reciprocal interactions of parents and children.
As the principal investigator, I regularly debriefed with peers who were family therapists, social workers, parents of adult gay and lesbian children, and gay lay people. Repeated reading of transcripts, coding, and discussion of the findings with peers and research assistants led to the concepts and theory that described family adjustment. As a reliability check, one of the research assistants coded portions of the interview transcripts into the key codes that I developed, and agreement ranged from 84–100 percent, with a mean of 94.5 percent.
Throughout this book, and especially in chapters 6 and 7, I have diligently tried to point out the limitations of the findings of this research, and in this section I reiterate some of these and add others. It bears repeating that this sample consisted primarily of families with parents who had mostly adjusted to the news that a daughter or son was gay, so the findings must be interpreted with that in mind. It would be helpful, though rather difficult, to recruit and interview parents who had rejected their children in order to examine how child factors such as cross-gendered mannerisms, developmental problems, and parent-child interactions play a role in parents’ decisions to reject their children.
Further, it would be helpful to know more about how families of Latino, Asian, and other cultural backgrounds adjust to having a gay or lesbian child, as it is likely that culture plays a strong role in parent’s and children’s feelings about homosexuality. Principal investigators who are bicultural and bilingual might be more likely than I, a third-generation Italian American white man, to have sufficient insider knowledge as to how to recruit such parents and encourage them to discuss these deeply personal issues.
All but five of the parents were living in the New York City and Philadelphia metropolitan areas, which are known to be fairly liberal parts of the country. It is expected the findings would be different if the study took place in a different region of the country. For example, if families from an area known as the Bible Belt were interviewed, religious conflicts might have played a more central role in family reactions. Thus a similar study that includes a sizable sample from more religious and traditionally conservative areas of the U.S. could prove illuminating.
Last, this study was cross-sectional and asked for retrospective recall, meaning that all respondents were interviewed at one point in time about incidents that occurred in the past. Such reports can be biased, whereby the way in which respondents remember events can differ from what actually occurred. Throughout this book, I discuss how such a bias could affect respondent reports. A longitudinal study of this topic, in which parents and children are interviewed close to the time the child comes out and then at various points in time afterward, could further test the family stage model identified in this study and determine necessary modifications.
Despite these limitations, the findings of this study provide additional insight into an important but understudied area of research. Hopefully, human service professionals will find these results useful in their work with this potentially vulnerable population of families.