• Home
• Courses Listed by Topic
• Courses Listed by Hrs
• Courses Listed by Code
• CE Requirements
• Colleague Share
• Large Group Discount
• Instructors | About Us
• Free Training Videos
• Create an Account

A cross-sectional street-intercept survey method (Miller, Wilder, Stillman, & Becker, 1997) was used to administer the "Sex and Love Survey, Version 2.0" to 2.733 participants at a series of GLB community events in New York City and Los Angeles between the fall of 2003 and the spring of 2004. The intercept survey approach to collecting data has been used in previous studies, including those focused on GLB persons, and has been shown to provide data that is comparable to that obtained from more methodologically rigorous approaches (Halkitis & Parsons, 2002; Koken, Parsons, Bimbi, & Severino, 2005).
The response rate was high, with 82.9% of individuals consenting to complete a survey. As an incentive, those completing the 15-20 minute questionnaire were provided with a voucher for free admission to a movie. The survey instrument included items that assessed experience with a broad range of sexual behaviors, history of sexually transmitted infections, substance use, physical health, and a series of scales related to psychological well-being. This study was conducted by the Center for HIV/AIDS Educational Studies and Training (CHEST) and approved by the Institutional Review Board of the authors.
To protect their confidentiality, participants were given the survey on a clipboard so they could step away from others to complete the questionnaire. Participants were also requested not to include any identifying information on the surveys. Upon completion, participants deposited their survey into a secure box at the event. Data were entered into an SPSS database and subsequently verified by project staff for accuracy.

Demographics. Participants were asked to indicate age, sexual identity, and race/ethnicity (by checking all that apply). Response categories to race/ethnicity included African American, Asian/Pacific Islander, European/White, Hispanic/Latino, and Other (Specify).
Coming out and sexual debut. Three separate questions were included following Floyd and Stein's (2002) example of the different aspects of coming out. Participants were asked to indicate the age they admitted to themselves they were GLB, the age they told another person they were GLB, and whether or not their parent/s (or person/s who raised them) were aware of their sexual activity with same-gendered partners. Finally, participants were asked to indicate the age they first had sex with a same-gendered partner.

We performed cross-tabulations and chi-square statistical tests to compare categorical data such as race/ethnicity, gender, and age cohorts. Mean differences by gender were compared using t-tests. Finally, in cases where categorical variables containing more than two categories (such as race and ethnicity) were compared across a continuous variable (such as age of first sex with someone of the same gender), we used ANOVA multiple comparisons and Bonferroni post-hoc analyses.
Age, originally an open-ended item, was recoded into age-cohort categories: 18 to 24; 25 to 34; 35 to 44; 45 to 54; and over 55. The youngest age cohort contained individuals spanning seven years, while older cohorts span ten years. Young adults, particularly those in their late teens and early twenties, are at a unique developmental phase called emerging adulthood (Arnett, 2004). Although ages in this cohort are unevenly balanced from other cohorts in this analysis, it is developmentally more homogenous than having combined all 18 to 29 year olds together.

On average, men reported coming out to themselves at a younger age (M = 17.5) than did women (M = 19.6), t( 2,519) = 5.66, p < .001. Men were also younger (M = 17.9) than women (M = 19.8) when engaging in their first same-gender sexual experiences, t( 2 ,259) = 5.58, p < .001. There were no significant gender differences between the ages that men and women reported coming out to others, nor whether they had come out to their parents.

Among women, there were significant differences between age cohorts with regard to coming out, F ( 4,377) = 14.15, p < .001 (see Table 3). Women ages 18 to 24 reported coming out to themselves at younger ages than all older cohorts (M = 15.88). Among men, similar differences were identified between age cohorts (F[ 4,2127] = 14.15, p < .001) in that those ages 18 to 24 reported coming out to themselves at a younger age than all older cohorts (M = 15.01). Although males ages 25 to 34 (M = 16.87) and those 35 to 44 (M = 17.31) did not differ statistically in the reported age of coming out to oneself, men ages 25 to 34 came out at an earlier age than those in older cohorts, 45 to 54 (M = 19.26) and 55 and above (M = 20.31).
Women ages 18 to 24 (M = 16.87) and 25 to 34 (M = 20.11) reported a significantly younger age at coming out to others compared to older cohorts, F ( 4,366) = 22.60, p < .001. Meanwhile, all remaining older cohorts of women did not significantly differ from one another. Men ages 18 to 24 (M = 16.94) reported coming out to others at a significantly younger age than all other age cohorts, F ( 4, 2,016) = 46.32, p < .001. In addition, men ages 25 to 34 (M = 19.84) reported coming out to others at younger ages than those 45 to 54 (M = 22.50) and 55 and above (M = 24.11).
Among men, younger cohorts were more likely to report being out to their parents than older cohorts (χ² [ 4] = 30.761, p < .001). Approximately 75% (n = 169) of men ages 18 to 24 reported being out to their parents, compared to 71% (n = 266) of men 45 to 54 and only 55% (n = 99) of men over 55 (see Table 4). Statistical analyses were not performed on women because of low expected counts in one cell.
Among women, significant differences in reported ages of sexual debut with a person of the same gender were identified across age cohorts, F ( 4,368) = 14.75, p < .001. Posthoc analyses revealed women ages 18 to 24 (M = 16.85) reported experiencing same-gender sex for the first time at significantly younger ages than those in all other age categories, with the exception of those ages 25 to 34 (M = 18.78). Women ages 35 to 44 (M = 21.33) and 45 to 54 (M = 23.00) did not significantly differ from one another in reported age of same-gender sexual debut; however, they did differ from those over 55 (M = 28.43) by more than seven years. Similar significant differences were found among men and their first reported same-gender sexual experience. Men ages 18 to 24 (M = 16.08) reported sexual debut at a younger age than those in all older age categories, F ( 4, 1,880) = 6.65, p < .001. Unique to this analysis was that none of the other male age cohorts significantly differed from one another in reported age of first same-gender sexual experience. See Table 3 for all age comparisons.

Because there were uneven distributions of race across age cohorts, racial and ethnic differences in ages of coming out to self and others were assessed within each cohort, split by gender. ANOVAs conducted with these data revealed no significant differences between racial and ethnic groups in reported ages of coming out to self and coming out to others. Essentially, the age cohort differences by gender (see Table 3) were robust across race and ethnicity.
We found significant differences between racial and ethnic groups in being out to one's parents. Separate cross-tabulations were conducted for both men and women (see Table 5). Approximately 80% (n = 164) of White women reported being out to their parents, compared to 61% (n = 39) of African American women, 72% (n = 55) of Latinas, and 68% (n = 30) of women identifying as Other races. Similarly, 77% (n = 1.094) of White men indicated they were out to their parents, compared to 51% (n = 82) of Asian/Pacific Islander men, 62% (n = 118) of African American men, 69% (n = 197) of Latinos, and 71% (n = 89) of men identifying as other races. These relationships remained intact even when we split the analysis by age cohort.
We conducted ANOVAs to examine potential race and ethnicity differences in age of same-gender sexual debut. For the most part, there were no significant differences in reported age of sexual debut with same-gender partners across race and ethnicity; however, among 25- to 34-year-olds, there were significant findings. White women ages 25 to 34 (M = 19.66) reported experiencing same-gender sexual debut approximately three years after Latina women ages 25-34 (M = 16.64), F ( 3 ,125) = 3.16, p < .05. African American men ages 25 to 34 (M = 15.51) reported their first same-gender sexual experience at younger ages than Asian/Pacific Islanders (M = 19.51), Whites (M = 18.03), and men identifying as Other (M = 19.61), F ( 4,607) = 7.08, p < .001. There were no other significant racial or ethnic differences in reported sexual debut within any other age cohort categories among men or women.

Some limitations to the data collected for this study should be considered when evaluating the generalizability of these findings. These results do not adequately represent all facets of the GLB community, but they give a comprehensive picture of the individuals that attend large-scale GLB events. The venues at which data were collected may not represent individuals who live outside of large urban cities. Since the GLB community is not restricted to metropolitan areas, further investigation into less densely-populated areas is warranted.
Lack of comparative racial and ethnic diversity among older GLB cohorts suggests a need for specific outreach in working with older GLB people of color. Since these data were collected at large-scale venues, lack of equal representation suggests older GLB people of color may not have had the desire to attend these events. As discussed in previous research, GLB people of color carry multiple identities. The salience of other identities could be stronger among older individuals (Bergung, 2004). Lack of representation among older GLB people of color could also be attributed to long-standing impacts of racism (Icard, 1986; Loicano, 1989; Martinez & Sullivan, 1998; Savin-Williams, 1997), hence reducing interest in attending the events where data were collected.
We were unsuccessful in obtaining a representative sample of Asian/Pacific Islander women. Further research is needed to understand how Asian identities might play a role in the lives of Asian/Pacific Islander lesbian and bisexual women and their involvement in the GLB community. Finally, although efforts were taken to ensure confidentiality, there was potential for biased responses due to social desirability in the reporting of sensitive information.

Few researchers have documented the diversity that exists within the GLB community in age, gender, and race/ethnicity. The findings from this analysis speak not only to other researchers who endeavor to study diversity within GLB communities, but also to service providers and health educators who continue to work with and for such communities. Treatment of psychosocial and developmental issues surrounding and within GLB communities must be culturally appropriate in order to meet the needs of the various populations represented.
It is imperative to provide support to GLB individuals during the coming-out process as it has been linked to both psychological and physical well-being (Cole. Kemeny, Taylor, & Visscher, 1996; Kennamer, Honnold. Bradford, & Hendricks, 2000; Peacock, 2000; Ryan & Futterman, 1998; Stokes & Peterson, 1998). Additionally, there is a documented relationship between lack of disclosure of sexual orientation and risk for HIV (Kennamer et al., 2000; Ryan & Futterman, 1998; Stokes & Peterson, 1998). Meanwhile, other researchers have identified a relationship between concealment of homosexual identity and diseases such as cancer (Cole et al., 1996).
Exploring issues of multiple identities among GLB people of color is central to research and program development for these populations. This study found that among both men and women, people of color were better-represented in younger age cohorts. Since these individuals are coining into both adulthood and their sexual identity, this is an important time to provide culturally-appropriate prevention and educational services for facilitating the coming-out process, as well as for fostering healthier behaviors. Researchers have found young men ages 18 to 24 are reporting more sexual risk behavior than in previous years (Bellis, Cook, Clark, Syed, & Hoskins, 2002; Catania et al., 2001). This analysis found the youngest cohort reported sex at earlier ages than all other cohorts. Thus, adolescence and early adulthood is an urgent time to provide support and educational services related to safer sex and other harm-reduction practices.
Acknowledgment and announcement of one's sexual identity to oneself and to others is a process experienced differently depending upon one's age, gender, and racial/ethnic background. The importance of addressing coming-out issues in health and social service provision, as well as in sexuality education, is apparent. These findings speak not only to health and social service providers, but also to policy makers developing sexuality education curriculum for teens and young adults. Further, they inform the arena of sexuality research and the impact historical changes of the last few decades have had on the coming-out process.

‘When working with gay men or lesbians, irrespective of

whether or not the psychotherapist is gay or lesbian

themselves, a therapist needs to have explored their own

sexuality and at least to some degree have transcended the

sex-role stereotypes offered to us in our society’

(p. 121).

It is Hitchings' belief that practitioners should not only have resolved their own prejudices with respect to lesbian, gay and bisexual clients, but that they should also be open to the idea of adopting affirmative models of psychotherapy, and be ready to acknowledge the courage and resilience of their clients who have broken with perceived social norms and to a certain degree ‘created’ or ‘recreated’ their own lifestyles. He has argued further that, for those practitioners who are themselves lesbian, gay or bisexual, they should have reached a point in their own identity development or ‘coming out’ (i.e. disclosure of sexual orientation to others) where they will have moved from a point of identity confusion and questioning through to tolerance, acceptance and pride and ultimately to identity synthesis or resolution (see Cass, 1979, 1984, 1990; D'Augelli, 1994; Rivers, 1997).
Interestingly, it may be argued that the concept of social difference is still viewed by many practitioners as an entity external to the counselling or supportive domain — as Widdowson (1999) suggested — and where it is mentioned, it is often viewed as a potential ‘problem’. For example, in the BAC (British Association for Counselling) Counselling Reader, Palmer et al. (1996) gave considerable space to issues of social difference, but only in the context of problem areas such as those including alcohol/drug abuse, HIV, PTSD and sexual abuse. Indeed, the section referring to sexual orientation contains only two chapters: one addressing bereavement counselling and the other on the challenge of HIV/AIDS. Unfortunately, no consideration is given to the experience of being outside the mainstream in terms of sexual orientation, or on the process of ‘coming out’, and living with the potential consequences — both positive and negative. This has occurred despite the numerous texts published in the field of psychology and mental health that address issues such as coming out since the late 1970s and early 1980s (see, for example, Coleman 1981/82; McDonald, 1982; McWhirter & Mattison, 1984; Plummer, 1981; Troiden, 1979).
Although Woolfe & Dryden (1996), in their editorial commentary for their counselling psychology text, argued that the practice of counseling or offering support to an individual should be grounded in values that respect difference, feelings and subjective truths, and that emphasize choice and responsibility, they did not address sexual orientation as a counselling or support issue, and this begs the question: what provision is made within professional or everyday contexts (e.g. educational and workplace settings) to offer counseling, support or, indeed, pastoral care for lesbian, gay and bisexual clients? Furthermore, it is also worth questioning the appropriateness of the informational base upon which advice, support or counseling is offered.
As D'Augelli (1994) has pointed out, one of the difficulties in understanding lesbian, gay and bisexual client issues relates to the fact that we do not, as yet, have a model of ‘normative’ development for lesbians, gay men and bisexual men and women. Much of the information and research on lesbian, gay and bisexual issues has focused and continues to focus upon the implications of discrimination and victimization, and given that these issues may only apply to approximately 40% of lesbians, gay men and bisexual men and women at any given time (see Rivers, 1999), it would seem appropriate to consider the process of growing up within the context of being a member of a sexual minority group, but without necessarily having been exposed to overt aggression, discrimination or prejudice.
In this article, we consider the experiences of sixteen individuals whose sexual orientation is (or is perceived to be) different from the mainstream. We explore the social experience of growing up in a world that often rejects same-sex sexual attraction, and consider the role of the practitioner in facilitating the client in developing a positive view of her/his own sexuality.

• the decision to come out or disclose one's sexual orientation to others;
• becoming aware of the implications of coming out for relationships with friends and family, and in the wider community; and
• moving towards a more integrated sense of identity.

‘A feeling of missing out… a feeling of hearing

people in school talking about these wild parties and their

… the sexual decathalons that they were … they

were involved in. And, the … the fun playing …

playing in the park after … after school with their

friends from home. A feeling that I was being deprived of

this’ (Simon, aged 27).

‘I remember I had a girlfriend for about a year …

and it was a completely useless time for me, completely,

because I would like meet [her] at barn dances, but really I

just wanted all the others to see me with my girlfriend, so I

would look… you know… peer pressure. It was that

wonderful freedom of living in the countryside and in a good

community atmosphere, and then the private torture of knowing

that you would never be part of that community. You were alien

to it’ (Paul aged, 27).

The period immediately prior to coming out was often an experience marked by a period of increased insecurity and vulnerability:

‘I think by the time I'd got to 19, 20 or 21, I realised

that if I was going to have any future, I was going to have to

tell someone I was gay and it all hung on that…

everything hung on that. So, I told someone and then I told

another person. The whole "coming out" process took

years, but I did it bit by bit by bit, but I had to go really

low, sink to a very, very, low, low point when I was about

— a bit older than I said — 22 when I did overdose

on sleeping tablets accidentally. It wasn't an attempt to die,

but I did overdose and that would be the lowest you would

get-the absolute rock bottom — there was seriously

something wrong. I knew what was wrong, it  was just that I

wasn't being honest with myself. But once I made the leap or

whatever, that changed by telling somebody then things very,

very slowly and very, very gradually began to get better bit by

bit by bit’ (Matthew, aged 36).

However, as the following extracts indicate, a positive response from counselors, friends and, importantly, identified gay support groups did have a significant effect upon the individual's mental state:

‘I started ringing the gay switchboard, I mean they must

be credited there, they were fantastic’ (Matthew, aged

36).

‘I first came out shortly after I came up to university.

By the time I got to Christmas of my first term I was in a

pretty rotten mental state and I knew that something had to be

done. I talked to a number of student counselors, I talked to

a couple of friends in roundabout circuitous ways, and

eventually I phoned the local gay switchboard, talked to them

and the next day [I] felt as if I was walking on air’

(Mark, aged 22).

‘I've only ever come out to good friends, or people

who've asked. There's only ever been one person I've done it to

and never heard or seen again. The response has generally been

good’ (Liam, aged 16).

‘When I got back in October I thought, "Right, you

know, new people around me, still got my old friends, but I'm

going to make new friends". My friends are really

supportive, but you can't just live with straight

friends’ (Alex, aged 19).

For others, such as Marcus and Paul, acknowledging a gay identity brought with it a number of personal and social conflicts:

‘I got involved in a social youth group and one of the

guys there asked me if I wanted to go on a pilgrimage to

Lourdes, and I said, "Yes, I quite fancy that". At

that time I was at the stage where I thought, well yeah, it

might cure me you know … I go and it might…

because I was struggling with it [being gay] so much, I

thought, well, I might at least get some respite and come to

terms with things a bit’ (Marcus, aged 31).

‘Sometimes I worry … I know that… say if

he [boyfriend] does something in public with straight people

that's particularly camp a part of me cringes, and I hate

cringing because that's the internal homophobia. That's my

problem’ (Paul, aged 27).

It is worth noting that while being an ‘out’ gay man meant that Paul's family and friends had come to accept both his sexual orientation and his same-sex partner, he remained sensitive to the way in which society viewed homosexuality, and particularly public demonstrations of sexual orientation. His own discomfort, which he recognised as his own ‘internal homophobia’, was described in terms of a ‘problem’, and was something which he felt needed to be addressed. Yet the question arises: how can Paul address his ‘problem’ when public demonstrations of same-sex attraction are not deemed acceptable, except in certain identified gay locales?
Paul's concerns were a common feature in the majority of interviews conducted by the second author. For example, Matthew (below) pointed out that it is only in the last few years that public figures have felt able to talk openly about their homosexuality or bisexuality. In the past there were very few positive role models accessible to young lesbians, gay men and bisexual men and women as they came out. Furthermore, since little information or guidance was generally available relating to how an individual should behave in public (for example, the perceived acceptability or unacceptability of two men holding hands), the absence of positive lesbian, gay or bisexual role models meant that there were few benchmarks upon which participants were able to structure their emergent alternative identities. Yet, as Matthew attested, role models are an invaluable resource:

‘I listened to a radio programme. It was Ian McKellen

talking about how he came out to his family, and he came out to

his sort of 92 year old maiden aunt and she said sort of,

"What are you worrying about?" and that…

that sort of then really made me certain that I was doing the

right thing’ (Marcus, aged 31).

For all 16 interviewees, the support they received from their family and friends was an important aspect of their development. However, in some cases they recalled that they did not receive the unconditional support they had hoped for:

‘I went home and told my mother. She just sat there and

she nodded her head and she said, "Right, fine",

and then started … we talked about it. Then she said,

"When you get older, if you feel that you've made a

mistake and you're not really gay then it's OK to tell

everybody that you're not gay"’ (Liam, aged 16).

While family members may not have been wholly supportive, local or university societies, or regional gay and lesbian switchboards were considered valuable resources that offered both friendship and advice for those experiencing difficulty in coming to terms with their sexual orientation.

‘Once I got to the university, I got involved in a lot of

feminism [and] tended to meet women through women's groups or

friends of friends, and I actually moved into an all-women

household at one point screaming to meet another lesbian’

(Susan, aged 30.)

It is arguable that, for the lesbian, gay and bisexual practitioners, having worked through their identity formation and having reached a point were they are comfortable with their sexual orientation, their experiences may actually provide a foundation upon which it is possible to offer guidance to clients in the absence of other positive role models (Davies, 1996). In fact, where such support was actively sought by participants, they reported that it had been beneficial. The interviews made it clear that both practitioners and volunteers are particularly effective where there is a shared understanding of the issues facing participants, although as Davies points out, it raises a number of practice issues for the individual providing support for the lesbian, gay or bisexual client.

‘I don't think anyone who isn't gay can ever understand

the complete, one hundred percent humiliation you feel because

all you know is you are yourself (Paul, aged 27).

Education was also seen by participants as a way forward in supporting the individual as they came to terms with their sexual orientation, and this not only related to those responsible for pastoral care and support, but also the general community:

‘Just teach people and educate them to say that it's fine

[to be gay], it's not wrong, it's not a disease’ (Tessa,

aged 16).

Realistically, of course, legislation such as Section 28 of the 1988 Local Government Act has prevented many schools from providing effective pastoral care for young people, although the Act itself only precludes schools from ‘promoting’ homosexuality as a ‘pretended family relationship’ (see Rivers, 2000). It is notable that, for all the participants in this study, being able to talk to someone without fear of rejection or prejudice was a valuable commodity, and one which was rarely available to them as they were developing their own identities as lesbians, gay men or bisexual men and women. As Susan pointed out, when she was coming out, those around her had little idea of how to support her effectively, and as she went on to say:

‘They had no advice, no information, nowhere to offer. I

wasn't necessarily expecting them to solve my problems, but had

they — any of them — told me there was anything

like a gay switchboard or a gay support group, or anything like

that it would have been an enormous help’ (Susan, aged

30).

Building upon the extracts quoted above, it is clear that those acting in supporting roles have a part to play in assisting lesbians, gay men and bisexual men and women address issues such as ‘coming out’ and the development of positive lesbian, gay or bisexual identities. They also have a role in maintaining those identities in a world that has yet to accept fully the lesbian, gay man or bisexual man and woman as an equal citizen. As some of the comments (above) illustrate, the residual influences of living as a lesbian, gay man or bisexual man or woman in a world of heterosexual stereotypes has a profound influence upon the individual, who without recourse to affirmative role models may have begun a process of constructing an identity which is largely founded on prevalent negative images of homosexuality.

‘What is important is the therapist's ability to

empathize with and accept the client. To be able to empathize

fully, the therapist needs to be able to set aside their own

fears and prejudices; to do this they need to be able to

understand the roots of them’ (p. 38).

While it is beyond the scope of this article to recommend any particular theoretical approach, texts such as the ‘Pink Therapy’ series provide a useful overview of the therapeutic perspectives currently utilized when working with lesbian, gay and bisexual clients (Davies & Neal, 1996, 2000; Neal & Davies, 2000). However, in developing the supportive role to facilitate lesbian, gay and bisexual clients, the practitioner's ability to acknowledge social difference and work with it is important. Through affirmation and support, the client's potential for personal development may be enhanced, and her or his capacity to make choices may also be enhanced. Parker (1998) proposes that through the deconstruction of the meaning of relationships practitioners can be helped to rework the supportive relationship. It may be possible for them to analyze and address issues of power objectively, and to reflect critically upon their roles of alleviating distress through providing effective and non-judgemental support. It may be argued that the lesbian, gay or bisexual practitioner's use of personal insights based upon their own experiences of growing up and identity formation may provide a way of offering guidance for clients where role models or modes of coping are not apparent. However, personal insights have limited use as they may also blur the professional nature of the practitioner-client relationship. Indeed, the fact that a number of lesbian, gay and bisexual counselors and psychotherapists specialize in lesbian, gay and bisexual issues suggests that, although not necessarily avowed, the use of shared understanding and shared insight is beneficial to the provision of support and counseling (see Davies, 1996).
Where the practitioner is not lesbian or gay, she or he should always acknowledge the client's ‘expertise’ on her or his situation, and try to obtain a sense of how the world might view the client, and how the client views the world. Ultimately, this is where Hitchings' (1994) argument is most salient: practitioners should have worked through their own feelings with respect to homosexuality and bisexuality, and through a process of self-evaluation and personal reflection new insights may develop.

celebrate and advocate the validity of lesbian, gay,and bisexual

 persons and their relationships. Such a therapist goes beyond a

 neutral or null environment to counteract the life-long messages

 of homophobia and heterosexism that lesbian, gay, and bisexual

 individuals have experienced and often internalized, (p. 736)

Thus, an absence of homophobia is not sufficient to practice affirmatively. Rather, affirmative practice requires that practitioners celebrate and validate the identities of gay men and lesbians and actively work with these clients to confront their internalized homophobia to develop positive identities as gay and lesbian individuals.
Gay affirmative practice is well suited for the many settings in which social workers assist clients. According to Appleby and Anastas (1998), "There is no particular approach to psychotherapy or other forms of mental health treatment nor any particular modality of treatment — individual, couple, family, or group — that cannot be made useful for lesbian, gay, or bisexual people if approached affirmatively" (p. 286).
In addition to being applicable across a variety of social work settings, such as case management, substance abuse treatment, child welfare, and private practice, gay affirmative practice is consistent with approaches familiar to many social workers (see Figure 1):

• Person in environment — Gay and lesbian individuals are considered in the context of the many environments in which they interact and the many roles they play. For example, when working with gay and lesbian individuals, affirmative practitioners pay attention to gay men's and lesbians' work and family settings and the degree to which they disclose their sexual orientation to others along with the roles that gay and lesbian individuals play in these environments.
• Strengths perspective — Gay and lesbian individuals are viewed as having many strengths that can assist them in addressing their presenting issue. Where appropriate, affirmative practitioners also use other components of the strengths model, including self-determination, by supporting gay and lesbian individuals in their decisions regarding when and to whom to disclose their sexual orientation (Appleby & Anastas, 1998); a focus on health, not pathology, by viewing identities as gays or lesbians as equally healthy as heterosexual identities (Davies, 1996); and consciousness raising, by encouraging gay men and lesbians to examine the impact of homophobic forces in their lives.
• Cultural competence models — Many of these models suggest chat culturally sensitive practice with diverse populations requires a unique knowledge base, set of attitudes and beliefs, and skill base for a given population (Sue et al., 1982). Van Den Bergh and Crisp (2004) asserted that gay affirmative practice is a form of cultural competence, similar to culturally sensitive practice with racial and ethnic minority groups.

Other aspects of gay affirmative practice should be noted. Appleby and Anastas (1998) discussed six principles of gay affirmative practice:

1. Do not assume that a client is heterosexual.
2. Believe that homophobia in the client and society is the problem, rather than sexual orientation.
3. Accept an identity as a gay, lesbian, or bisexual person as a positive outcome of the helping process.
4. Work with clients to decrease internalized homophobia that they may be experiencing so that clients can achieve a positive identity as a gay or lesbian person.
5. Become knowledgeable about different theories of the coming out process for gay men and lesbians.
6. Deal with one's own homophobia and heterosexual bias.

Hunter and colleagues (1998) presented guidelines for affirmative practice, including, but not limited to, the following: Understand and abide by one's professional code of ethics; value clients' sexual orientations; do not view sexual orientation as the problem; do not attempt to change clients' sexual orientation; support clients in deciding how "out" to be; and do not attempt to identify the cause of clients' sexual orientation, as doing so may be destructive to the client.
Davies (1996) claimed that holding any of the following beliefs precludes practitioners from working affirmatively with gay and lesbian individuals: Homosexuality is sinful or against God's wishes; homosexuality is sick, unnatural, or perverted; homosexuality is inferior to heterosexuality; monogamy is the only healthy way to have a relationship; gay and lesbian relationships can only be short-term, sexual, or lacking depth; gay men and lesbians are more likely to sexually abuse children; gay and lesbian parents are inferior to heterosexual parents; and bisexual individuals can decide to be gay or lesbian or heterosexual. Practitioners who hold these beliefs cannot work with gay men and lesbians in ways that convey respect for them and seek to affirm their identities as gay and lesbian individuals. Concern arises that these beliefs about gay men and lesbians lead to beliefs about practice with gay men and lesbians. For example, a belief that homosexuality is against God's wishes may lead to a belief that gay men and lesbians should be treated for their homosexuality so that they may no longer be sinners; a belief that gay men and lesbians are more likely to sexually abuse children may lead practitioners to treat gay men and lesbians for pedophilia when there is no other indication of such a disorder.

• Tinsley and Tinsley (1987) recommend that a minimum of five to 10 respondents per item are needed to conduct the factor analysis. Given that the factor analysis was conducted on 80 items, a minimum sample size of 400 usable surveys was needed.
• Earlier research using mailed surveys to assess social workers' attitudes toward gay men and lesbians yielded response rates between 32 percent and 63 percent (Hardman, 1997; Harris, Nightengale, & Owen, 1995). Using this minimum response rate of 32 percent, a return of 960 usable surveys could be expected.
• In view of the terrorist attacks on the World Trade Center and the Pentagon on September 11, 2001, and the subsequent disruption in the mail system when envelopes containing anthrax were sent to members of Congress and other individuals were infected by anthrax, concern arose that respondents might be reluctant to open mail from a source unfamiliar to them. Thus, it seemed wise to expect a much lower response rate than previous studies and to assume that the response rate might be as low as half the lowest response rate noted earlier.
• Given that 400 responses were needed and a minimum response rate of 16 percent was assumed, a minimum sample of 2,500 was needed. To further safeguard against a low response rate, this number was rounded up to a total of 3,000 total packets mailed.

Measurement Package. Along with the 80-item Gay Affirmative Practice Scale (GAP) developed from the expert review stage, a packet consisting of the following items was mailed to each potential respondent:

• The Attitudes Toward Lesbians and Gay Men Scale (ATLG) (short form), developed by Herek (1988), was administered to assess the convergent construct validity of the GAP's behavior domain. The ATLG has been validated with college students and members of gay and lesbian organizations and demonstrated high internal consistency (alpha of .90). Herek found that the scale correlates significantly in the expected direction with attitudes toward sex roles, with traditional family ideology, and with reports of positive contact with lesbians and gay men. Although the research correlating attitudes and behaviors provides mixed results, it was expected that people who held more positive attitudes toward gay men and lesbians would be more affirming in their practice with this population than those who held more negative attitudes toward gay men and lesbians. Thus, the behavior domain of the GAP was expected to correlate significantly with the ATLG.
• The Heterosexual Attitudes Toward Homosexuals Scale (HATH), developed by Larsen, Reed, and Hoffman (1980), was administered to assess the convergent construct validity of the GAP's belief domain The HATH has been validated with college students, resulting in a split-half reliability of .86, and when corrected with the Spearman-Brown prophecy formula, has a split-half reliability of .92 (Larsen et al.). Larsen and colleagues found that the scale correlated significantly in the expected direction with instruments that assess religious ideology, authoritarianism, and feelings of inadequacy. As with the ATLG and the GAP's behavior domain, significant correlations between the HATH and the GAP's belief domain were expected.
• The Marlowe-Crowne Social Desirability Scale (SDS), developed by Crowne and Marlowe (1960) and reduced to a shorter version by Strahn and Gerbasi (1972) using 10 of the original 33 items, was used to assess discriminant construct validity. The short version of the scale has high internal consistency (.876) and a correlation of .958 with the original scale (Fischer & Fick, 1993). A nonsignificant correlation with the SDS and the entire GAP was expected and would thus provide initial evidence of discriminant construct validity.
• Twenty demographic items were chosen based on the literature review and other items of interest. These items inquired about the respondents' personal characteristics such as gender, race, and sexual orientation; religious and political affiliation; and contact with and feelings about gay men and lesbians.

Mailing Procedure. On January 25, 2002, the University of Texas at Austin University Mailing Services (UMS) distributed the instrument packets to the potential respondents using mailing labels provided by NASW and APA. A cover letter and a self-addressed business reply envelope in which to return the survey were included in the packets. According to UMS, materials should have been received by the respondents no later than February 1, 2002. Respondents were requested to return the materials to the researcher by February 15, 2002.
Assessing Reliability, Internal consistency reliability was established through an analysis of responses obtained in the administration to clinicians. Internal consistency reliability is considered a more effective method for computing reliability (Nunnally & Bernstein, 1994) and was thus the most logical method for establishing reliability in this study.
The standard error of measurement (SEM) was also computed for each domain. The SEM is an estimate of the standard deviations of error of measurement and is less influenced by differences in variance and standard deviation in different samples or populations than coefficient alpha (Springer, Abell, & Nugent, 2002). The SEM should be computed to compensate for differences in sample standard deviations. A small SEM provides evidence that the scale is reliable.
Assessing Validity. Content validity was examined and can be thought of as the degree to which items in an instrument represent the construct of interest. It is often established by having independent experts assess whether the items adequately assess the construct (Springer, 1997). Convergent construct and discriminant construct validity were assessed to examine the degree to which the GAP correlated with measures that were theoretically related to it. Construct validity is closely tied to theory and is concerned with theoretical relationships between variables (DeVellis, 1991; Springer). As multidimensional scales are collections of unidimensional scales (Hudson, 1985), convergent construct validity was established for each domain rather than for the scale as a whole, and discriminant construct validity was examined for the entire scale. Confirmatory factor analysis (CFA) using the multiple groups method (Nunnally & Bernstein, 1994) was used to establish factorial validity. This method allows the researcher to examine the correlation between each individual item and each domain in the scale. This in turn allows the researcher to confirm or disconfirm a priori hypotheses about factor loadings. The number of domains was fixed at two and were titled "belief" and "behavior." As a general rule, item correlations are considered moderately high when they load on their intended domain at or about .60 (Nunnally & Bernstein). As discussed by several authors, validating a scale is an ongoing process and does not end on completion of the aforementioned types of analyses.