Attitudes Toward Lesbians and Gay Men Scale

Attitudes Toward Lesbians and Gay Men Scale‌

GREGORY M. HEREK1 AND KEVIN A. MCLEMORE, University of California at Davis

The Attitudes Toward Lesbians and Gay Men (ATLG) Scale is a brief measure of heterosexuals’ attitudes toward gay men and lesbians. The original scale consisted of 20 different statements, 10 about gay men (ATG subscale) and 10 about lesbians (ATL subscale), to which respondents indicated their level of agreement or disagreement. Shorter versions have subsequently been developed, consisting of ATG and ATL subscales with parallel versions of 3, 4, or 5 items. These shorter versions have been found to be highly correlated with the original, longer subscales (e.g., rs >.95 between 5-item versions of the ATG and ATL and their 10-item counterparts), and their use is now recommended instead of the original subscales.

Response Mode and Timing

The ATLG can be self-administered (presented on paper or on a computer) or administered orally (as in a telephone survey). When presented visually, scale items are typically accompanied by a 5-, 7-, or 9-point Likert-type scale with anchor points of Strongly Disagree and Strongly Agree. When administered orally during telephone or face-to- face interviews, four response options are usually offered (Strongly Disagree, Disagree Somewhat, Agree Somewhat, Strongly Agree), and respondents are allowed to volunteer a middle response (e.g., “Neither Agree nor Disagree”). Completion time is roughly 30–60 seconds per item.

Scoring

Scoring is accomplished by assigning numerical values to the response options (e.g., for a 7-point response format, 1 = Strongly Disagree, 7 = Strongly Agree) and summing across items for each subscale. Some items are reverse scored as indicated below. For ease of interpretation, the sum of item values can be divided by the total number of items to yield a score that matches the response scale metric. The possible range of scores depends on the response scale used.

Scores on the original ATL and ATG subscales, which are based on responses to differently worded items, were not directly comparable. Researchers wishing to compare respondents’ attitudes toward gay men with their attitudes toward lesbians were advised to use parallel forms of one subscale (usually the ATG items). The use of such parallel forms (with each item presented once in reference to gay men and once in reference to lesbians) is now recommended for all ATLG scale users, as shown in the Exhibit.

Reliability and Validity

The ATLG subscales have high levels of internal consistency. When self-administered, α > .85 with most college student samples and α > .80 with most nonstudent adult samples. For telephone surveys with oral administration to adult samples, α > .80 for 5-item versions and α > .70 for 3-item versions. Test-retest reliability (rs > .80) has been demonstrated with alternate forms (Herek, 1988, 1994).

Scores on the ATLG subscales are reliably correlated with other theoretically relevant constructs (e.g., Herek, 1994, 2009; Herek & Capitanio, 1996, 1999a). Higher scores are associated with high religiosity, lack of inter- personal contact with gay men and lesbians, adherence to traditional gender-role attitudes, belief in a traditional fam- ily ideology, and endorsement of policies that discriminate against sexual minorities. In addition, ATG scores are reliably correlated with AIDS-related stigma.

The ATLG’s discriminant validity also has been established. Members of lesbian and gay organizations scored at the extreme positive end of the range, and nonstudent adults who publicly supported a gay rights ballot measure scored significantly lower on the ATLG than did community residents who publicly opposed the initiative (Herek, 1988, 1994).

Administration in Other Languages and Outside the United States

Although the ATLG was originally developed for administration to English-speaking heterosexual adults in the United States, it has also been used in research conducted in England (Hegarty, 2002) and Canada (Mohipp & Morry, 2004), and translated versions have been administered in the Netherlands (Meerendonk, Eisinga, & Felling, 2003), Singapore (Detenber et al., 2007), Brazil (DeSouza, Solberg, & Elder, 2007), Chile (Cardenas & Barrientos, 2008; Nierman, Thompson, Bryan, & Mahaffey, 2007), and Turkey (Gelbal & Duyan, 2006). In addition, a Spanish-lan- guage version was created for a study of California adults of Mexican descent (Herek & Gonzalez-Rivera, 2006). In these studies, scale reliability has been consistently accept- able (typically, α > .80), and the patterns of correlations between ATLG scores and theoretically related constructs have been similar to those obtained with U.S. English- speaking samples.

Other Information

Item order effects have been observed among heterosexual males in telephone surveys, which may indicate a gender- linked pattern in heterosexuals’ attitudes (Herek, 2002; Herek & Capitanio, 1999b). For more information about the ATLG’s development and usage, see Herek (1994, 2009). Researchers need not obtain the author’s permission to use the ATLG in not-for-profit research that is consistent with the American Psychological Association’s Ethical Principles of Psychologists.

1Address correspondence to Gregory Herek, Department of Psychology, University of California, Davis, CA 95616-8686; e-mail through http://psychology.ucdavis.edu/Herek/

References

Attitudes Toward Lesbians and Gay Men Scale, Revised 5-Item Version

Attitudes Toward Gay Men (ATG-R-S5) Subscale

1. Sex between two men is just plain wrong.*

2. I think male homosexuals are disgusting.*

3. Male homosexuality is a natural expression of sexuality in men.* (Reverse scored)

4. Male homosexuality is a perversion.

5. Male homosexuality is merely a different kind of lifestyle that should not be condemned. (Reverse scored)

Attitudes Toward Lesbians (ATL-R-S5) Subscale

1. Sex between two women is just plain wrong.*

2. I think female homosexuals (lesbians) are disgusting.*

3. Female homosexuality is a natural expression of sexuality in women.* (Reverse scored)

4. Female homosexuality is a perversion.

5. Female homosexuality is merely a different kind of lifestyle that should not be condemned. (Reverse scored)

*This item is included in the 3-item version (ATLG-R) of the subscale.

Attitudes Toward Lesbians and Gay Men Scale: Spanish-Language Version

Attitudes Toward Gay Men (ATG)

1. Las relaciones sexuales entre dos hombres simplemente están mal. [Sex between two men is just plain wrong.]

2. Yo pienso que los hombres homosexuales son repugnantes. [I think that male homosexuals are disgusting.]

3. La homosexualidad masculina es una expresión natural de la sexualidad del hombre. [Male homosexuality is a natural expression of sexuality in men.]

4. La homosexualidad masculina es una perversión. [Male homosexuality is a perversion.]

Attitudes Toward Lesbians (ATL)

1. Las relaciones sexuales entre dos mujeres simplemente están mal. [Sex between two women is just plain wrong.]‌

2. Yo pienso que las lesbianas son repugnantes. [I think that lesbians are disgusting.]

3. La homosexualidad femenina es una expresión natural de la sexualidad de la mujer. [Female homosexuality is a natural expression of sexuality in women.]

La homosexualidad femenina es una perversión. [Female homosexuality is a perversion.]

Note. The Spanish-language version was adapted for use by Herek and Gonzales-Rivera (2006).

Component Measure of Attitudes Toward Homosexuality

MARY E. KITE,1 Ball State University

Researchers have demonstrated that attitudes toward gay men and lesbians are multidimensional and that, to fully understand antigay prejudice, these different aspects of people’s attitudes and perceptions should be considered (see Kite & Whitley, 1996; LaMar & Kite, 1998). Because attitudes toward homosexuality can serve different functions for different people (Herek, 1986), antigay prejudice can best be understood by considering these differing perspectives. One attitude component reflects the general belief that homo- sexuality violates traditional values, for example, whereas another component reflects the possibility that people fear contact with gays and lesbians in general or sexual advances from a same-sex other specifically. These attitudes are sepa- rate from the stereotypic beliefs people hold about gay men and women and from people’s tendency to support or deny gay men and lesbians’ civil rights.

Description

To construct this measure, established measures of attitudes toward homosexuality, summarized in Beere (1990), were reviewed. Initially, 174 items were selected that conceptually represented five components of attitudes toward homosexuality: Civil Rights, Condemnation/Tolerance, Contact, Morality, and Stereotypes. These items were categorized by component, and overlapping items were eliminated. For those categories not well represented by previous measures, new items were written.

A sample of 270 college students completed the measure, and their responses were analyzed using a varimax

factor analysis. Results revealed five factors that accounted for 56.8% of the variance. Most parallel items for lesbians and gay men loaded on the same factor; the exception was that the gay male and lesbian contact items emerged as separate factors. Contrary to expectations, civil rights did not emerge as a separate factor. Rather, those items loaded on the Condemnation/Tolerance factor, thereby resulting in four final components.

The final measure consists of 91 items assessing attitudes toward gay men and lesbians. Of these, 42 items address attitudes toward gay men and 42 parallel items address attitudes toward lesbians. An additional seven items refer to homosexuality or same-sex interactions (and do not vary by sex of target). Number of items per subscale and Cronbach’s alpha, based on LaMar and Kite (1998), are as follows: Gay Male Condemnation/Tolerance (11 items; alpha = .92); Lesbian Condemnation/Tolerance (11 items; alpha = .89); Gay Male Morality (10 items; alpha = .92); Lesbian Morality (10 items; alpha = .93); Neutral Morality (3 items; alpha = .80); Gay Male Contact (14 items; alpha= .96); Lesbian Contact (14 items; alpha = .95); Neutral Contact (4 items; alpha = .75); Gay Male Stereotypes (7 items; alpha = .78); Lesbian Stereotypes (7 items; alpha =.75). This measure was developed using a college student sample; however, it can be used with any population.

Response Mode and Timing

Participants evaluate these items using a 5-point Likert scale ranging from 1, Strongly Agree, to 5, Strongly

Address correspondence to Mary E. Kite, Department of Psychological Science, Ball State University, Muncie, IN 47306; e-mail: [email protected]

Correlations Among Subscales of Component Measure of Attitudes Toward Homosexuality

	Gay Condemnation- Tolerance	Lesbian Condemnation- Tolerance	Gay Morality	Lesbian Morality	Gay Contact	Lesbian Contact	Neutral Contact	Gay Stereotypes	Lesbian Stereotypes
Gay Condemnation-
Tolerance		.90	.83	.75	.83	.55	.46	.62	.61
Lesbian Condemnation-
Tolerance			.79	.79	.78	.67	.50	.62	.63
Gay Morality				.91	.85	.64	.58	.63	.61
Lesbian Morality					.76	.75	.59	.62	.63
Gay Contact						.61	.67	.60	.56
Lesbian Contact							.63	.54	.54
Neutral Contact								.42	.42
Gay Stereotypes									.79
N	263	264	264	264	263	263	264	264	264

Note. All correlations are significant at p < .001. Reprinted from LaMar and Kite (1998), Table 2, with permission from Taylor & Francis.

Disagree. It is recommended that the item order be randomized and parallel items not appear contiguously on the measure. At administration, items from the four factors can be intermixed. However, each factor can be used as a stand- alone instrument to assess a specific component of attitudes toward homosexuality. The instrument takes approximately 15 minutes to complete.

Scoring

Subscales are created for each of the 10 factors by first reverse scoring items that are worded negatively. Items that are reverse scored are marked on the instrument (see the Exhibit). Subscale scores are then created by averaging participants’ ratings on the items. LaMar and Kite (1998) reported correlations among the subscales. Correlations between the gay male and lesbian versions of the sub- scales are high for Condemnation/Tolerance (r = .90) and Morality (r = .91) and smaller for Contact and Stereotype (see Table 1).

Reliability

Information concerning the internal consistency of the sub- scales is presented in the description of the development of the scale (see above). Additional information is provided in Table 1. Test-retest reliability of the measure has not been established.

Validity

To establish validity of the factors, LaMar and Kite (1998) compared mean subscale scores for each target sex for male and female respondents. Results showed that men evalu- ated gay men more negatively than they evaluated lesbians on both the Condemnation/Tolerance and the Morality subscales. On these two subscales, women respondents evaluated gay men and lesbians similarly. For the Contact subscale, men rated contact with gay men more negatively than contact with lesbians, and women rated contact with

lesbians more negatively than contact with gay men. There were no sex of respondent differences on the Stereotype subscale, but gay men were evaluated more negatively than lesbians on that subscale. These results provide support for the idea that attitudes toward homosexuality are multidimensional.

Whitley (1999) found that, averaged across two samples, the Stereotype subscale (for gay men and lesbians combined) was correlated with Right-Wing Authoritarianism (Altemeyer, 1988; r= .44) and Social Dominance Orientation (Pratto, Sidanius, Stallworth, & Malle,1994; r=.37) but was unrelated to affective responses to gays and lesbians (r = −.14). Wilkinson (2006) reported correlations between subscales assessing attitudes toward lesbians and attitudes toward gender roles. Specifically, he found the Attitude Toward Women Scale (AWS; Spence & Hahn, 1997) was correlated with the Contact subscale (r = .41), the Morality subscale (r = .47), and the Stereotype subscale (r =.43). He also reported that Hostile Sexism (HS), as assessed by the Ambivalent Sexism Inventory (ASI; Glick & Fiske, 1996), was correlated with the Contact (r = .26), Morality (r = .30), and Stereotype (r = .44) subscales. Benevolent Sexism (BS), as assessed by the ASI, was correlated with the Contact (r = .35), Morality (r = .37), and Stereotype (r= .43) subscales. Finally, he reported that the Benevolence Toward Men (BM) subscale of the Ambivalence Toward Men Inventory (Glick & Fiske, 1999) correlated with the Contact (r = .47), Morality (r = .47), and Stereotype (r =.54) subscales. Similar relationships between HS and BM are reported in Wilkinson (2008).

Component Measure of Attitudes Toward Homosexuality

Record your responses to each item using the following scale:

1	2	3	4	5
Strongly Agree				Strongly Disagree

Condemnation/Tolerance

1. 1. Apartment complexes should not accept lesbians (gay men) as renters.

   2. Lesbians (gay men) should be required to register with the police department where they live.

   3. Lesbians (gay men) should not be allowed to hold responsible positions.

   4. Job discrimination against lesbians (gay men) is wrong.*

   5. Lesbians (gay men) are a danger to young people.

   6. Lesbians (gay men) are more likely to commit deviant acts such as child molestation, rape, voyeurism (peeping Toms) than are heterosexuals.

   7. Lesbians (gay men) dislike members of the opposite sex.

   8. Finding out an artist was a gay man (lesbian) would have no effect on my appreciation of his (her) work.*
   9. Lesbians (gay men) should be allowed to serve in the military.*
   10. Lesbians (gay men) should not be discriminated against because of their sexual preference.*

   11. Lesbians (gay men) should not be allowed to work with children.

Gay Male/Lesbian Social Norms/Morality

1. The increasing acceptance of gay men (lesbians) in our society is aiding in the deterioration of morals.

2. Gay men (lesbians) endanger the institution of the family.

3. Many gay men (lesbians) are very moral and ethical people.*
4. Gay male (lesbian) couples should be able to adopt children the same as heterosexual couples.*

5. The idea of marriages between gay men (lesbians) seems ridiculous to me.

6. State laws regulating private, consenting behavior between gay men (lesbians) should be loosened.*

7. Gay men (lesbians) just can’t fit into our society.

8. Gay men (lesbians) do need psychological treatment.

9. Gay men (lesbians) are a viable part of our society.*

10. Homosexual behavior between two men (women) is just plain wrong.

Neutral Morality

1. Homosexuality, as far as I am concerned, is not sinful.*

2. Homosexuality is a perversion.

3. I find the thought of homosexual acts disgusting.

Gay Male/Lesbian Contact

1. I enjoy the company of gay men (lesbians).*

2. It would be upsetting to me to find out I was alone with a gay man (lesbian).

3. I avoid gay men (lesbians) whenever possible.

4. I would feel nervous being in a group of gay men (lesbians).

5. I think gay men (lesbians) are disgusting.

6. I would enjoy attending social functions at which gay men (lesbians) were present.*

7. Bars that cater solely to gay men (lesbians) should be placed in a specific and known part of town.

8. I would feel comfortable working closely with a gay man (lesbian).*

9. If a gay man (lesbian) approached me in a public restroom, I would be disgusted.

10. I would not want a gay man (lesbian) to live in the house next to mine.

11. Two gay men (lesbians) holding hands or displaying affection in public is revolting.

12. I would be nervous if a gay man (lesbian) sat next to me on a bus.

13. I would decline membership in an organization if I found out it had gay male (lesbian) members.

14. If I knew someone was a gay male (lesbian), I would go ahead and form a friendship with that individual.*

Neutral Contact‌

1. If a member of my sex made advances toward me, I would feel angry.

2. I would feel comfortable knowing I was attractive to members of my sex.*
3. I would be comfortable if I found myself attracted to a member of my sex.*

4. I would feel uncomfortable if a member of my sex made an advance toward me.

Gay Male/Lesbian Stereotypes

1. Lesbians (gay men) prefer to take roles (passive or aggressive) in their sexual behavior.

2. The love between two lesbians (gay men) is quite different from the love between two persons of the opposite sex.

3. Lesbians (gay men) have weaker sex drives than heterosexuals.

4. A lesbian’s (gay man’s) mother is probably very domineering.

5. Most lesbians (gay men) have a life of one-night stands.‌‌

6. Most lesbians (gay men) like to dress in opposite-sex clothing.

7. Most lesbians (gay men) have identifiable masculine (feminine) characteristics.

*Items reverse scored.

References