Summary

AUTHOR NOTE: This is an extended summary of a paper by Gregory M. Herek, in A Queer World, edited by Martin Duberman (1997, New York University Press, pp. 331-345). The research was originally reported in G.M. Herek & J.P. Capitanio (1996). "Some of My Best Friends": Intergroup Contact, Concealable Stigma, and Heterosexuals' Attitudes Toward Gay Men and Lesbians. Personality and Social Psychology Bulletin. The research was supported by a grant to Gregory Herek from the National Institute of Mental Health (R01 MH43253).

In the past quarter-century, scientific research on homosexuality has undergone a sea change. Breaking with past studies that defined gay men and lesbians as mentally ill and sought a cure for homosexuality, social and behavioral scientists have instead turned their attention to the many problems and challenges that gay people face in a heterosexist society. These include the problems created by individual and institutional prejudice, often labeled homophobia, heterosexism or, more recently, sexual prejudice.

One of the most consistent findings in this research area has been that heterosexuals who personally know a lesbian or gay man manifest more positive general attitudes toward gay people as a group. This finding is consistent with a long-standing social psychological theory of prejudice called the contact hypothesis. As formulated by Gordon Allport (1954), the contact hypothesis states:

"Prejudice (unless deeply rooted in the character structure of the individual) may be reduced by equal status contact between majority and minority groups in the pursuit of common goals. The effect is greatly enhanced if this contact is sanctioned by institutional supports (i.e., by law, custom, or local atmosphere), and if it is of a sort that leads to the perception of common interests and common humanity between members of the two groups" (Allport, 1954, p. 267).

Unfortunately, most empirical research on the association between heterosexuals' attitudes and their personal contact with gay men or lesbians has not moved beyond simply demonstrating that such a correlation exists. The research described here was designed to explore in greater depth the role that contact plays in shaping heterosexuals' attitudes toward gay people. I report findings obtained from a 2-wave probability sample in a national telephone survey concerning AIDS and stigma. Because of time limitations and other constraints, only the second wave of the survey (conducted in 1991-92) included questions about attitudes toward both lesbians and gay men. The first wave (conducted approximately one year earlier) included questions only about attitudes toward gay men.

With my collaborator, John Capitanio, I sought to replicate the previously reported finding of a correlation between having contact and professing favorable attitudes toward gay people. We also tested three hypotheses, based on the contact hypothesis. First, we hypothesized that contact experiences with two or more gay individuals are associated with more favorable attitudes than are contact experiences with only one person. Because gay people inevitably differ on characteristics irrelevant to their category membership, heterosexuals with multiple contact experiences have increased opportunities for observing such variation and, consequently, individuating outgroup members. Such individuation (i.e., thinking of a group as consisting of varied individuals rather than as a monolithic entity) is likely to reduce intergroup prejudice.

Second, we hypothesized that contact with gay close friends or immediate family members is more likely to be associated with favorable intergroup attitudes than is contact with mere acquaintances or distant family members who are gay. Close relationships with gay men or lesbians can provide heterosexuals with intimate, personally relevant information about gay people. They are likely to foster personalization of gay people -- that is, thinking of gay people as complex human beings rather than as abstract symbols or unidimensional caricatures -- which helps to reduce prejudice.

Third, we hypothesized that a lesbian or gay person's management of information concerning her or his sexual orientation (which, in most cases, is concealable) has important implications for heterosexuals' attitudes. We predicted that heterosexuals who have been told directly by another person that he or she is gay are more likely to have positive attitudes toward gay people generally than are heterosexuals who acquired such information about a friend or relative indirectly (e.g., from a third party). In part, this prediction is based on previous findings that self-disclosure of personal information often leads to greater liking of an individual. In addition, we assumed that most heterosexuals - as a consequence of living in a society in which homosexuality is stigmatized - possess relatively little knowledge about gay people and hold attitudes toward gay people that are more negative than favorable. Upon learning that a friend, relative, or acquaintance is homosexual, they are likely to follow one of three courses: (1) attach their preexisting antigay stereotypes and attitudes to that person, and possibly reinterpret past experiences with her or him in a way that is consistent with those prejudices; (2) maintain positive feelings toward the person while regarding her or him as an atypical case that is not representative of the larger population of gay people; or (3) maintain positive feelings toward the person and, on the basis of those positive feelings, individuate and personalize the larger category of gay and lesbian people.

Whereas many factors could determine which of these outcomes occurs, we assumed that the third course - which involves changing long-standing beliefs and deeply-felt attitudes - requires the greatest cognitive effort and is therefore the least likely. We hypothesized, however, that one or more direct discussions with a friend or relative about the latter's homosexuality can help to motivate the heterosexual person both to maintain the relationship and to change her or his attitudes toward gay people generally. In addition, such conversations can provide the heterosexual person with information that will assist her or him in regarding the friend or relative as representative (or not atypical) of gay men or lesbians while also individuating the category of gay people.

In addition to testing these hypotheses about the possible effects of contact on attitudes, we also wished to replicate and extend a finding from my previous research with Eric Glunt (Herek & Glunt, 1993). In that earlier study, we found that heterosexuals are more likely to report contact to the extent that they belong to demographic groups that (1) have more opportunities for contact (e.g., heterosexuals living in urban settings) and (2) are perceived by gay men and lesbians as more accepting of gay people (e.g., women, the well educated) than is society as a whole. This pattern suggests the possibility of a reciprocal relationship between contact and attitudes: Not only might intergroup contact reduce prejudice, as predicted by the contact hypothesis, but individuals low in prejudice might also have more opportunities for contact.

Method

The methods used in the national survey are briefly described here. Readers desiring more detailed information should consult my published papers with John Capitanio (Herek & Capitanio, 1993, 1994, 1995, 1996). The Wave 1 sample was drawn from the population of all English-speaking adults (at least 18 years of age) residing in households with telephones within the 48 contiguous states. Telephone numbers were generated using random-digit dialing, or RDD. Interviews were conducted by the staff of the Survey Research Center at the University of California at Berkeley between 12 September 1990 and 13 February 1991, using their computer-assisted telephone interviewing (CATI) system.

Wave 1 interviews were completed with 538 respondents, which represented a response rate of 70%. Interviews lasted an average of 39 minutes. Approximately one year later, we attempted to recontact all Wave 1 respondents. We were able to complete follow-up interviews with 382 (71%) of the original respondents. The Wave 2 interviews lasted an average of 40 minutes. All of the findings reported here are based on data from respondents who self-identified as heterosexual (506 in Wave 1, and 363 in Wave 2).

Measures

The survey included a large number of questions about AIDS-related attitudes and beliefs, as well as respondents' demographic characteristics. Only the items relevant to the present chapter are described here.

Attitudes toward gay men (Waves 1 and 2). Attitudes toward gay men were measured with a 3-item short form of the Attitudes Toward Gay Men (ATG) scale, which has been shown to be a reliable and valid measure of heterosexuals' attitudes toward gay men when administered by telephone and in paper-and-pencil format (see Herek, 1994). For each statement, respondents were provided with four response alternatives (agree strongly, agree somewhat, disagree somewhat, disagree strongly) which were scored on a 4-point scale. Item responses were reversed as necessary and summed to yield a scale score that could range from 3 to 12, with higher scale scores indicating more unfavorable attitudes. The items are listed in Table 1.

Attitudes toward lesbians (Wave 2 only). In the Wave 2 survey, a 3-item short form of the Attitudes Toward Lesbians (ATL) scale was included in the survey protocol. It comprised the same three ATG items, but reworded to apply to lesbians. Response alternatives and scoring were the same as for the ATG items. The items are listed in Table 1.

Contact experiences. Personal contact was assessed through a series of questions. First, respondents were asked whether they had "any male or female friends, relatives, or close acquaintances who are gay or homosexual" and, for those answering in the affirmative, how many. Respondents reporting only one relationship were asked to describe the gay person's gender, how she/he was related to the respondent (immediate family, other family, close friend, other friend, close acquaintance), and how the respondent first learned about the other person's sexual orientation (were told directly by her/him, were told by someone else, just guessed that the person is gay). Those who chose either of the last two alternatives were asked, "Has he/she ever told you directly that he/she is gay?" Respondents who reported knowing two or more gay people were asked the same series of questions about each of "the two gay people you feel closest to."

Results

Wave 1

Of the 538 respondents with complete Wave 1 interviews, 46% were male and 54% were female. Racially, the sample was 81% White, 10% Black, 5% Hispanic, and 3% Asian (less than 1% of respondents did not use one of these labels). The mean age was 43.8 years; median annual household income was between $30,000 and $40,000; and the median level of educational attainment was "some college."

Attitudes toward gay men. As shown in Table 1, most respondents expressed negative attitudes toward gay men, with a majority agreeing that "Sex between two men is just plain wrong" and that "I think male homosexuals are disgusting." Only a minority agreed that "Male homosexuality is a natural expression of sexuality in men." When we summed responses to these three items into an ATG score, the overall mean score was 9.08.

Interpersonal Contact. Almost one-third of the sample (31%) reported that they knew at least one person who is gay or lesbian. Within this subgroup, roughly one-third knew one gay person whereas two-thirds knew two or more. Both male and female respondents were more likely to report that their closest relationships were with gay men than with lesbians. Of the 263 reported relationships (with 55 respondents describing one relationship and 104 describing two), only 27% were with a lesbian. Shifting the unit of analysis from the relationship to the respondent, only 34% of those who knew one or more gay people described at least one relationship with a lesbian.¹ Female respondents were more likely than males to know a gay person (of those who knew any gay people, 67% were women).

Respondents were more likely to describe contact with gay or lesbian friends than with relatives. Of the 263 reported relationships, 75% were with a friend or acquaintance (21% with a close friend, and 54% with an acquaintance or more distant friend), whereas only 23% were with a relative (4% with immediate family, and 19% with more distant relatives).²

In slightly more than one-third (38%) of the relationships reported, the heterosexual person learned directly from the friend or relative about the latter's homosexuality. In the other relationships, the heterosexual was told by a third party (32%) or guessed that the person was gay (30%). In one-fourth of the latter situations (told by third party, guessed), the heterosexual subsequently was told directly by the gay person about her or his sexual orientation. Thus 53% of the relationships described by respondents included direct disclosure, either initially or after the heterosexual person learned through another route that the person was gay (note that respondents could describe up to two relationships).

Interpersonal Contact and Attitudes. As expected, heterosexuals reporting contact with a gay person had significantly more favorable attitudes toward gay men than those without contact. Moreover, respondents manifested progressively more favorable attitudes to the extent that they knew more gay people. We also found that intimate contact was more likely than superficial contact to be associated with favorable attitudes. As shown in Figure 1, the most favorable attitudes were manifested by respondents with a gay close friend, whereas the least favorable attitudes were manifested by respondents with a gay distant relative. Scores for those with an immediate family member or gay acquaintance were intermediate between these two extremes. Figure 1 also displays the tendency for respondents with multiple relationships to manifest more positive attitudes than those with only one gay friend, relative, or acquaintance.

Receiving direct disclosure of another's homosexuality was more likely to be associated with positive attitudes toward gay people than was having acquired such information indirectly. Of the 153 respondents with contact, 33% reported disclosure from one friend or relative, 28% reported disclosure from two friends or relatives, and 38% reported no direct disclosure. Respondents who had been told directly by a friend or relative about her or his homosexuality manifested significantly lower ATG scores (more favorable attitudes) than did those who had guessed or had been told by a third party; the effect was even stronger if respondents had received disclosure from two gay men or lesbians.

We observed that whether or not heterosexuals were recipients of direct disclosure was strongly related to the closeness of their relationship with the gay person. Almost all (93%) of the respondents with a gay close friend were recipients of direct disclosure, compared to 86% of those in the immediate family group, 57% in the other friend group, and 9% of those in the distant relative group. Because of this strong association, the observed intergroup differences in ATG scores may have resulted from the closeness of relationships rather than from receiving disclosure. Statistically disentangling the disclosure and relationship variables was not possible with the current data set. However, one category of relationships - acquaintances and friends who were not described as "close" - included roughly equal numbers of respondents with and without disclosure experiences (57% and 43%, respectively). ATG scores for respondents in this group differed significantly according to disclosure category, with the lowest (most favorable) scores manifested by those reporting two or more disclosures. Thus, disclosure appears to be associated with more favorable attitudes independently of its association with type of relationship - at least among respondents with two or more gay acquaintances.

Who has contact? We found that certain groups of heterosexuals were more likely than others to experience contact and direct disclosure. Having contact with a gay man or woman was predicted by being female, having a higher educational level, not attending religious services frequently, being younger, living in one of the Pacific coast states, and having a higher income. Among those who knew a gay person, the recipients of direct disclosure were more likely than others to be politically liberal, single, and an urban resident.

Before discussing the implications of these findings, I present the results from the Wave 2 survey.

Wave 2

Attitudes toward lesbians and gay men. The response distributions were remarkably similar for attitudes toward lesbians and toward gay men (see Table 1). Most respondents expressed negative attitudes. A majority agreed that "Sex between two women is just plain wrong" and that "I think lesbians are disgusting." Roughly one-fourth of respondents agreed that "Female homosexuality is a natural expression of sexuality in women" and with the comparable item about male homosexuality. The overall mean scale scores were 9.0 for the ATL and 9.1 for the ATG.

Interpersonal contact. As in Wave 1, almost one-third of the sample (32%) knew at least one person who is gay or lesbian. Also as in Wave 1, respondents within this subgroup tended to know more than one gay person, were more likely to report that their closest relationships were with gay men than with lesbians, and were more likely to describe contact with gay or lesbian friends than with relatives.

Interpersonal contact and attitudes. As in Wave 1, contact with a gay person was associated with significantly more favorable attitudes toward lesbians and gay men. Once again, respondents manifested progressively more favorable attitudes to the extent that they knew multiple gay people. This relationship is shown in Figure 2, which displays mean ATG and ATL scores for respondents reporting one, two, three or more, and no relationships with gay people. As in Wave 1, we also observed that intimacy of the relationship was related to attitudes, with the least favorable attitudes manifested by respondents reporting a gay distant relative. As in Wave 1, respondents reporting direct disclosure had more favorable attitudes than did respondents reporting contact without disclosure (see Figure 3). The difference was statistically significant, however, only for respondents who reported disclosures from at least two friends or relatives.

Discussion

Heterosexuals who had experienced interpersonal contact with gay men or lesbians expressed significantly more favorable general attitudes toward gay people than did heterosexuals without contact. This pattern was generally consistent across both waves of data collection, and for attitudes toward lesbians and gay men alike. Because most previous research in this area has not directly assessed heterosexuals' attitudes toward lesbians, the extent to which findings about attitudes toward gay men or toward "homosexuals" (a term likely to evoke attitudes toward gay men) could be generalized to attitudes toward lesbians has been in doubt. The present research suggests that such attitudes closely resemble attitudes toward gay men.

The relationship between contact and attitudes was affected by three different aspects of the contact experience. First, favorable attitudes were more likely among heterosexuals who reported multiple contacts with lesbians or gay men. Although knowing one gay person was associated with more positive attitudes than was knowing none, only respondents who knew at least two gay people were consistently significantly different from those with no contacts. Perhaps knowing multiple members of a stigmatized group is more likely to foster recognition of that group's variability than is knowing only one group member. Knowing multiple members of a group may also reduce the likelihood that their behavior can be discounted as atypical.

The two other dimensions of contact examined here - degree of intimacy and direct disclosure - were highly correlated. Having a close gay or lesbian friend was almost always associated with direct disclosure, whereas heterosexuals who knew lesbians or gay men only as distant relatives were likely to have learned about the individual's sexual orientation indirectly. One interpretation of this pattern is that gay people come out to their close friends but not to distant relatives or acquaintances (with whom their homosexuality may be common knowledge but not openly discussed). Alternatively, disclosing one's stigmatized sexual orientation may strengthen a relationship, whereas not disclosing - despite the heterosexual's knowledge that one is homosexual - may weaken a relationship. In either case, the results are consistent with the contact hypothesis: Interpersonal relationships characterized by intimacy, shared values, and common goals are more likely to be associated with favorable attitudes toward gay people as a group than are superficial or distant relationships.

Although the strong correlation between closeness of relationship and receipt of disclosure makes it difficult to evaluate the individual contribution of each, both variables appear to affect intergroup attitudes. Closer relationships were consistently associated with more favorable attitudes. Furthermore, in the one relationship category for which disclosure experiences were nearly equally divided (acquaintances/distant friends), respondents reporting at least two disclosure experiences had significantly more favorable attitudes toward gay men than did other respondents.

The importance of disclosure and relationship type is also highlighted by our analysis of ATG scores among the 26 respondents who reported knowing one or more gay people at Wave 2 but none at Wave 1. This group did not manifest a significant attitude change across waves, a finding that could be interpreted as disconfirming the contact hypothesis. I believe, however, that this pattern is better understood as demonstrating that type of contact, not contact per se, shapes intergroup attitudes. For all but one of the 26 respondents, the relationship newly described at Wave 2 was distant: with a distant relative, an acquaintance, or a friend described as "not close." Furthermore, two-thirds of the 26 respondents did not report direct disclosure. Thus, although this subsample experienced new intergroup contact between Waves 1 and 2, that contact was of the sort least likely to reduce prejudice. Consequently, the subsample does not provide an adequate test of the contact hypothesis.

When we examined trends across the two waves of data collection, we found that heterosexuals who knew a gay man or lesbian when we first interviewed them subsequently tended to develop more positive attitudes toward gay people as a group, a conclusion that is consistent with the contact hypothesis. Yet, we also observed that heterosexuals with favorable attitudes at Wave 1 were subsequently more likely than others subsequently to experience contact. When possible, lesbians and gay men appear to be selective in associating with heterosexuals and revealing their sexual orientation.

Another theoretically interesting finding concerns the apparent relationship between intergroup attitudes and receipt of disclosure. In a close relationship, we speculate that a minority individual's direct disclosure about her or his concealable stigma can provide the majority group member with the necessary information and motivation to restructure her or his attitudes toward the entire minority group. This seems most likely to occur when the gay man or lesbian carefully manages the disclosure process so that the heterosexual can receive information (e.g., about what it means to be gay, about the gay person's similarity to other gay people) in the context of a committed relationship. For example, the gay person may disclose in a series of gradual stages, frame the disclosure in a context of trust and caring, explain why she or he did not disclose earlier, answer the heterosexual person's questions, and reassure the heterosexual that her or his past positive feelings and favorable judgments about the gay friend or relative are still valid.

Such interactions may assist the heterosexual person in keeping salient the relevant ingroup-outgroup distinction (i.e., heterosexual-homosexual) while observing behaviors that are inconsistent with her or his stereotypic beliefs, thereby facilitating the rejection of those stereotypes while fostering attitude change. If this experience leads the heterosexual person to accept that the friend or relative is indeed representative of the larger community of gay people (i.e., the friend or relative is not regarded as an anomaly), the heterosexual is likely to experience cognitive dissonance: On the one hand, she or he has strong positive feelings toward the gay friend or relative; on the other hand, she or he probably has internalized society's negative attitudes toward homosexuality. If the dissonance is resolved in favor of the friend or relative - an outcome that is more likely when the gay person plays an active role in imparting information about her or his stigmatized status - the heterosexual's attitudes toward gay people as a group are likely to become more favorable.

In contrast, a readily apparent stigma (such as race or physical disability) can usually be detected without such disclosure. Consequently, contact between the bearers of such stigma and members of the majority group may be less likely to reduce the latter's prejudice than when a stigma is concealable. This is exemplified in the assertion by a White person that "Some of my best friends are Black." Although having a best friend from a minority group should be associated with an absence of prejudice toward the group, making such a statement is commonly perceived as a defensive attempt to disavow racist attitudes. Rather than simply dismissing the statement (many individuals who make such a statement probably do not actually have best friends from the minority group), we can draw a potentially important insight from it. Because of the visible nature of race, a White person can have a Black friend but never discuss issues related to race in any depth. Without such discussion, even a White who personally knows Blacks might still retain negative stereotypes and attitudes toward African Americans as a group (e.g., if her or his Black friends are not perceived as representative of African Americans generally). In contrast, because homosexuality represents a concealable stigma, knowing that some of one's best friends are gay probably means that a heterosexual has directly discussed homosexuality with gay individuals and consequently has acquired greater insight and empathy for their situation, which can be generalized to gay people as a group. Rather than concluding that her or his friends are unlike other gay people, for example, such discussions might lead a heterosexual to regard sexual orientation as irrelevant to one's qualities as a human being.

The results reported here suggest directions for future studies. Collecting heterosexuals' first-person accounts of their contact experiences with gay people would be useful for identifying different patterns of contact and developing hypotheses about their role in attitude change. Similarly, descriptions by gay men and lesbians of their coming out experiences could be useful for describing how gay people decide to disclose to others, how they manage the disclosure process, and what happens when they lose control of that process. Controlled field experiments and longitudinal survey studies of heterosexuals' attitudes will be important for understanding the causal relationships between contact and attitude change. They also will permit description of the cognitive processes that underlie these relationships.

The findings also have important policy implications. At the most basic level, they demonstrate that heterosexuals can and do establish close relationships with openly gay people. This conclusion is contrary to one of the U.S. government's principal objections to allowing gay people in the military, namely, that heterosexual personnel cannot overcome their prejudices against homosexuality. A second implication of the findings is that heterosexuals' antigay prejudices are likely to be reduced in the course of close, ongoing contact that involves direct disclosure about sexual orientation. Thus, institutional policies are more likely to reduce prejudice to the extent that they encourage gay people to disclose their homosexual orientation to heterosexual peers. Conversely, policies that discourage or punish such disclosure may perpetuate prejudice.

Recognizing the ongoing dangers posed by societal prejudice, lesbian and gay activists nevertheless have often called upon gay people to disclose their sexual orientation publicly, that is, to come out of the closet. Perhaps the most noted political leader to advocate this strategy was Harvey Milk, San Francisco's first openly gay Supervisor, who was assassinated in 1978. For example, in a message that he had recorded to be played in the event of his death, Milk expressed the belief that coming out would eliminate prejudice: "I would like to see every gay lawyer, every gay architect come out, stand up and let the world know. That would do more to end prejudice overnight than anybody could imagine" (Shilts, 1982, p. 374).

Such calls to come out reflect a conviction that the tenets of the contact hypothesis are applicable to heterosexuals' attitudes toward lesbians and gay men. Although not definitive, the findings of the present study suggest that this belief is fundamentally correct. Coming out to heterosexuals - especially to close friends and immediate family - appears to reduce prejudice against gay people as a group. Furthermore, the finding that heterosexuals with multiple contacts and disclosures hold the most favorable attitudes of any group suggests that coming out will be most effective as a strategy for reducing prejudice when it is practiced by large numbers of lesbians and gay men. Thus, although coming out to loved ones exposes gay men and lesbians individually to the possibility of ostracism, discrimination, and even violence, it appears to be one of the most promising strategies for promoting the kind of societal change that will ultimately end such prejudice.

Bibliography

Allport, G. (1954). The nature of prejudice. New York: Addison Wesley.

Herek, G. M. (1994). Assessing heterosexuals' attitudes toward lesbians and gay men: A review of empirical research with the ATLG scale. In B. Greene & G. Herek (Eds.), Contemporary perspectives in lesbian and gay psychology (Vol. 1, pp. 206-228). Newbury Park, CA: Sage.

Herek, G. M., & Capitanio, J. P. (1993). Public reactions to AIDS in the United States: A second decade of stigma. American Journal of Public Health, 83, 574-577.

Herek, G.M., & Capitanio, J.P. (1994). Conspiracies, contagion, and compassion: Trust and public reactions to AIDS. AIDS Education and Prevention, 6, 367-377.

Herek, G.M., & Capitanio, J.P. (1995). Black heterosexuals' attitudes toward lesbians and gay men. Journal of Sex Research.

Herek, G.M., & Glunt, E.K. (1993). Interpersonal contact and heterosexuals' attitudes toward gay men: Results from a national survey. Journal of Sex Research, 30, 239-244.

Shilts, R. (1982). The mayor of Castro Street: The life and times of Harvey Milk. New York: St. Martin's.

Notes

1. Because respondents were asked to describe their two closest relationships with a gay/lesbian person (rather than all relationships), more distant relationships may have manifested a different gender distribution. Respondents' closest relationships, however, tended to be with gay men.

2. A few individuals reported that they knew at least one gay man or lesbian in response to the initial screening question, but then declined to answer subsequent questions about the relationship(s). Consequently, the numbers described here do not total 100%.

Table 1

Heterosexuals' Attitudes Toward Gay Men (Waves 1 and 2) and Lesbians (Wave 2)

ITEM 1990-91 1991-92

____________________________________________________________________________

Sex between two men

is just plain wrong.

% Agree (somewhat/strongly) 69.8 68.3

% Disagree (somewhat/strongly) 28.7 31.4

I think male homosexuals

are disgusting.

% Agree (somewhat/strongly) 54.1 59.9

% Disagree (somewhat/strongly) 44.8 39.7

Male homosexuality is a natural

expression of sexuality in men.

% Agree (somewhat/strongly) 23.6 24.6

% Disagree (somewhat/strongly) 74.4 75.4

____________________________________________________________________________

Sex between two women is

just plain wrong.

% Agree (somewhat/strongly) NA 64.3

% Disagree (somewhat/strongly) NA 35.3

I think lesbians

are disgusting.

% Agree (somewhat/strongly) NA 59.9

% Disagree (somewhat/strongly) NA 39.7

Female homosexuality is a

natural expression of

sexuality in women.

% Agree (somewhat/strongly) NA 26.6

% Disagree (somewhat/strongly) NA 73.2

____________________________________________________________________________

NA = item was not administered in that wave of the study. Reprinted from Herek (1994).