Attitude toward Lesbians and Gay Men Scale (ATLG)

Abstract

The Attitude toward Lesbians and Gay Men Scale (ATLG), developed by social psychologist Dr. Gregory M. Herek in 1984, is one of the most widely utilized and psychometrically validated self-report instruments designed to assess explicit sexual prejudice and anti-gay bias among heterosexual populations. Moving conceptually beyond early pathologizing formulations of “homophobia” as an irrational clinical phobia, the ATLG operationalizes anti-gay attitudes as socially learned, ideological, and functionally motivated belief systems rooted in cultural heterosexism. The scale comprises 20 items partitioned into two correlated 10-item subscales: the Attitudes Toward Lesbians (ATL) subscale and the Attitudes Toward Gay Men (ATG) subscale. Items are traditionally evaluated using a 9-point Likert response format ranging from 1 (“Strongly Agree”) to 9 (“Strongly Disagree”), though abbreviated 5-point and short-form variants (e.g., ATLG-S, ATLG-R) have also been established in the literature. Psychometric evaluations consistently demonstrate robust internal consistency, with Cronbach's alpha coefficients regularly exceeding .85 for the ATL, .90 for the ATG, and .90 for the total ATLG index across diverse demographic cohorts. Factor analytic research validates a robust two-factor structure corresponding to the gender of the target group, frequently revealing nuanced second-order factors related to moral condemnation, social/civil equality, and personal discomfort. Convergent, discriminant, and construct validities are thoroughly corroborated through robust correlations with Right-Wing Authoritarianism (RWA), Social Dominance Orientation (SDO), traditional gender-role endorsement, intrinsic religiosity, and interpersonal contact with sexual minority individuals. The ATLG remains an indispensable instrument in social psychology, clinical assessment, sociology, public health, and intergroup relations research.

Keywords

Attitude toward Lesbians and Gay Men Scale, ATLG, Gregory M. Herek, sexual prejudice, heterosexism, homophobia measurement, Attitudes Toward Gay Men, Attitudes Toward Lesbians, social psychology, psychometrics, scale validation, gender roles

Authors

The scale was conceptualized, developed, and validated by Gregory M. Herek, Ph.D., an internationally recognized authority on prejudice against sexual minorities, hate crimes, and LGBTQ+ health psychology. Dr. Herek is Professor Emeritus of Psychology at the University of California, Davis. Across four decades of empirical investigation, his pioneering theoretical scholarship dismantled clinical and psychoanalytic interpretations of “homophobia”—originally popularized by George Weinberg in the early 1970s—and substituted them with rigorous social-psychological models of sexual stigma, heterosexism, and sexual prejudice. Dr. Herek has authored over one hundred peer-reviewed publications, served as an expert witness in pivotal legal cases regarding marriage equality and military service (e.g., “Don't Ask, Don't Tell”), and received numerous distinguished honors, including the American Psychological Association (APA) Award for Distinguished Contributions to Psychology in the Public Interest.

Purpose

The fundamental purpose of the Attitude toward Lesbians and Gay Men Scale (ATLG) is to provide empirical researchers, educators, and clinicians with a reliable, theoretically anchored, and multidimensional measure of explicit attitudes, cognitive evaluations, and emotional aversions directed toward sexual minority individuals. Prior to the scale's introduction in 1984, psychological literature predominantly relied on idiosyncratic, unstandardized questionnaires or instruments that conflated attitudes toward gay men with attitudes toward lesbian women under an undifferentiated umbrella construct of “homophobia.” Dr. Herek recognized that heterosexual individuals frequently exhibit distinct psychological responses depending on the gender of the target: heterosexual men, for instance, systematically express markedly more hostility, moral condemnation, and visceral aversion toward gay men than toward lesbians, whereas heterosexual women often evaluate both groups with comparable, less punitive attitudes.

In research contexts, the ATLG serves several critical functions:

• Evaluating Intergroup Bias: It enables precise tracking of sexual prejudice within social psychology, allowing investigators to observe how demographic variables (e.g., gender, age, socioeconomic status, educational attainment, political ideology) intersect with explicit outgroup hostility.
• Assessing Intervention Efficacy: It provides a sensitive pre- and post-test metric to evaluate the effectiveness of diversity training, educational workshops, contact-hypothesis interventions, and curricular reforms aimed at reducing institutional and individual bias.
• Correlating with Systemic Injustice: It facilitates cross-disciplinary research investigating how individual levels of sexual prejudice predict endorsements of discriminatory public policies, opposition to relationship recognition, healthcare access disparities, and susceptibility to perpetrating or excusing anti-LGBTQ+ hate crimes.

In clinical and training contexts, the ATLG has been employed in the supervision of mental health professionals, medical trainees, and counselors to detect implicit or explicit countertransference, value conflicts, and residual biases that might compromise ethical, affirmative care for lesbian, gay, and bisexual clients. The theoretical rationale rests on the recognition that prejudice is not simply an individual pathology, but a socially conditioned, value-expressive cognitive schema requiring systematic measurement.

Psychological Construct

The primary psychological construct assessed by the ATLG is explicit sexual prejudice, defined by Herek as negative evaluations, hostile affect, and stigmatizing beliefs held toward an individual or group based solely on their non-heterosexual orientation. Central to the construction of the ATLG is the conceptual bifurcation of sexual prejudice into two discrete, target-specific dimensions: attitudes toward female homosexuality and attitudes toward male homosexuality.

Attitudes Toward Lesbians (ATL)

The ATL subscale comprises 10 items assessing cognitive, affective, and ideological evaluations of lesbian women. This dimension captures multiple components of anti-lesbian sentiment:

• Moral and Religious Condemnation: Items evaluate perceptions of female homosexuality as sinful, morally corrupting, or contrary to ethical codes (e.g., Item 5: “Female homosexuality is a sin”; Item 6: “The growing number of lesbians indicates a decline in American morals”).
• Violation of Gender Roles: The scale taps into anxieties regarding the subversion of traditional patriarchal divisions and biological essentialism (e.g., Item 3: “Female homosexuality is detrimental to society because it breaks down the natural divisions between the sexes”).
• Pathologizing and Social Threat: Items probe beliefs that lesbians represent a psychological defect or a destructive threat to foundational societal institutions (e.g., Item 8: “Female homosexuality is a threat to many of our basic social institutions”; Item 10: “Lesbians are sick”).
• Civil Rights and Non-Discrimination: Reverse-scored items assess the degree of endorsement for egalitarian protections in employment and legal autonomy (e.g., Item 2: “A woman’s homosexuality should not be a cause for job discrimination in any situation”; Item 4: “State laws regulating private, consenting lesbian behavior should be loosened”).

Attitudes Toward Gay Men (ATG)

The ATG subscale comprises 10 items assessing explicit psychological orientations toward gay men. Empirical research across decades highlights that the ATG frequently elicits stronger, more polarized affective responses than the ATL, particularly among heterosexual male respondents. The subscale operationalizes several psychological facets:

• Visceral Affect and Moral Disgust: The ATG measures direct revulsion and normative moral condemnation (e.g., Item 12: “I think male homosexuals are disgusting”; Item 14: “Male homosexuality is a perversion”; Item 18: “Homosexual behavior between two men is just plain wrong”).
• Parental and Interpersonal Anxiety: Items capture psychological distress related to close-contact familial scenarios or pedagogical roles, reflecting persistent societal stereotypes associating gay men with threats to youth (e.g., Item 13: “Male homosexuals should not be allowed to teach school”; Item 17: “I would not be upset if I learned that my son were a homosexual”).
• Institutional Equality and Family Legitimacy: Items address legal recognition, marital rights, and adoption equality (e.g., Item 11: “Male homosexual couples should be allowed to adopt children the same as heterosexual couples”; Item 19: “The idea of male homosexual marriages seems ridiculous to me”).
• Naturalness versus Curative Imperatives: The subscale taps biological essentialism and the belief that sexual orientation should be actively suppressed or altered (e.g., Item 15: “Just as in other species, male homosexuality is a natural expression of sexuality in human men”; Item 16: “If a man has homosexual feelings, he should do everything he can to overcome them”).

By differentiating the ATL from the ATG, the scale permits researchers to detect subtle asymmetrical patterns in intergroup bias, establishing that prejudice against gay individuals is systematically mediated by the respondent's own gender and the gender of the evaluated group.

Theoretical Framework

The ATLG is theoretically grounded in Herek’s neofunctional theory of attitudes (Herek, 1986a, 1987), integrated with social identity theory, hegemonic masculinity, and theories of cultural heterosexism. Traditional functional theories of attitudes (such as those developed by Daniel Katz and M. Brewster Smith) proposed that attitudes serve psychological needs, including utilitarian, ego-defensive, value-expressive, and knowledge functions. Herek synthesized and refined these formulations to explain why individuals harbor hostility toward sexual minorities.

The Functional Approach to Stigma

According to Herek, attitudes toward lesbians and gay men do not stem from a monolithic origin; instead, they serve distinct psychological functions that vary across individuals and situations:

• Value-Expressive Function: For many individuals, expressing hostility toward gay men and lesbians serves to assert and validate core identity markers, such as conservative religious beliefs, traditional family values, or political allegiance. Holding anti-gay attitudes functions as a symbolic public reaffirmation of one's worldview.
• Social-Expressive (Social-Adjustive) Function: Prejudicial attitudes often serve to gain social approval, strengthen ingroup cohesion, and avoid social ostracism from peer groups that endorse heteronormative standards. Expressing disdain toward sexual minorities acts as a badge of membership in orthodox social circles.
• Defensive (Ego-Defensive) Function: Rooted in psychodynamic formulations, ego-defensive attitudes protect the individual's self-esteem against internal conflicts, unconscious anxieties regarding their own sexual adequacy, or perceived failures to conform to rigorous cultural gender expectations. By denigrating gay men, heterosexual men may symbolically affirm their own hyper-masculinity and distance themselves from any appearance of effeminacy.
• Experiential (Utilitarian) Function: Attitudes may directly reflect generalized, firsthand past experiences. Individuals who have had constructive, close interpersonal contact with openly gay or lesbian individuals typically develop more benevolent, empathetic attitudes, whereas those whose interactions were negative or non-existent rely heavily on stereotypic societal scripts.

Cultural Heterosexism and Gender-Role Socialization

Herek further positioned the ATLG within the macro-sociological framework of cultural heterosexism—the societal ideology that institutionalizes, rewards, and renders normative heterosexual desires and relationships while subordinating, pathologizing, and rendering invisible any non-heterosexual behavior or identity. The ATLG presumes that individual attitudes (psychological heterosexism) are internalized manifestations of this broader cultural heterosexism. Crucially, the theoretical framework emphasizes that anti-gay attitudes are deeply entangled with gender-role enforcement. Hostility toward gay men often serves as a policing mechanism for traditional male role norms (emotional stoicism, anti-femininity, dominance), explaining why male respondents consistently display higher scores on the ATG than on the ATL.

Validity

Extensive psychometric investigations over four decades have firmly established the construct, criterion, convergent, and discriminant validity of the ATLG scale.

Construct and Convergent Validity

Construct validity has been repeatedly corroborated by evaluating the ATLG against empirically related psychometric scales:

• Right-Wing Authoritarianism (RWA) and Social Dominance Orientation (SDO): Research consistently reveals substantial positive correlations between ATLG scores and RWA (typically ranging from $r = .50$ to $r = .72$) as well as SDO ($r = .35$ to $r = .55$). Individuals who value conventionalism, authoritarian submission, and group-based social hierarchies reliably score higher in anti-gay prejudice.
• Traditional Gender Ideology: The ATLG correlates strongly with measures of traditional gender-role attitudes, such as the Attitudes Toward Women Scale (AWS) and the Male Role Norms Inventory (MRNI). Strong adherence to rigid gender divisions correlates positively with elevated scores on both the ATL ($r \approx .45$) and the ATG ($r \approx .60$).
• Religiosity: Extensive empirical evidence demonstrates moderate to strong positive correlations between the ATLG and intrinsic religiosity, religious fundamentalism, and frequent attendance at religious services, reflecting the theological condemnation embedded in conservative denominations.
• Interpersonal Contact: In accordance with Gordon Allport's contact hypothesis, scores on the ATLG are consistently and significantly negatively correlated with having gay or lesbian friends, family members, or acquaintances. Heterosexual individuals reporting close, positive contact exhibit markedly lower ATLG scores.

Discriminant Validity

Discriminant validity is supported by research demonstrating that although the ATLG correlates with other prejudice measures (e.g., modern racism, ethnic prejudice), it remains conceptually and statistically distinct. Factor analyses incorporating racial prejudice scales and sexual prejudice scales consistently show that items load onto separate, clean factors, confirming that the ATLG does not merely reflect generalized misanthropy or generic social intolerance, but rather a distinct, domain-specific ideological posture.

Predictive and Criterion-Related Validity

The ATLG demonstrates strong behavioral and policy-predictive validity. Longitudinal and cross-sectional studies show that elevated ATLG scores reliably predict opposition to legal protections for LGBTQ+ individuals, opposition to same-sex marriage, support for bans on same-sex adoption, reluctance to hire or work alongside openly gay colleagues, and willingness to tolerate homophobic speech or hate crimes. In laboratory settings, high scorers on the ATG display elevated physiological indicators of stress and avoidance (e.g., increased galvanic skin response, greater physical seating distance) when interacting with an openly gay confederate.

Reliability

The psychometric reliability of the ATLG has been demonstrated across diverse cultural, demographic, and linguistic samples. Internal consistency and temporal stability metrics consistently exceed conventional psychometric standards.

Internal Consistency

In Herek’s original 1984 validation study involving university undergraduates, the full 20-item ATLG demonstrated an exceptional internal consistency with a Cronbach's alpha of $\alpha = .90$. When evaluated independently, the subscales demonstrated strong reliability:

• Attitudes Toward Lesbians (ATL): Cronbach's alpha coefficients typically range from $\alpha = .77$ to $\alpha = .89$ across diverse undergraduate and adult community cohorts.
• Attitudes Toward Gay Men (ATG): Cronbach's alpha coefficients regularly span from $\alpha = .89$ to $\alpha = .94$, reflecting exceptionally high internal coherence among items assessing hostility toward gay men.

Subsequent psychometric evaluations across national probability samples (Herek, 1988, 2002) confirmed that the high reliability of the ATLG is sustained across broader demographic segments, including older adults, varying education levels, and diverse regional populations.

Temporal Stability (Test-Retest Reliability)

Test-retest reliability evaluations over intervals ranging from two to four weeks have yielded stability coefficients between $r = .80$ and $r = .90$ for the total ATLG, indicating that the scale captures stable, enduring socio-political and affective attitudes rather than transient situational moods. Even across longer periods of several months in non-interventional control groups, test-retest correlations remain exceptionally high ($r > .75$), demonstrating robust longitudinal measurement stability.

Factor Analysis

The dimensional architecture of the ATLG has been scrutinized through both Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA) across dozens of empirical studies.

Initial Exploratory Factor Analyses

In Herek’s foundational 1984 investigation, principal components analysis with varimax rotation clearly revealed a robust two-factor solution accounting for a substantial proportion of the total variance (over 50%). Factor 1 unambiguously comprised items targeted at gay men (ATG, Items 11–20), while Factor 2 was defined by items targeted at lesbians (ATL, Items 1–10). Primary factor loadings for intended items typically exceeded .55, with minimal cross-loadings onto the non-target factor (rarely exceeding .25). The correlation between the two factors is typically moderate to strong ($r = .50$ to $r = .70$), confirming that while attitudes toward lesbians and gay men share significant common variance under the superordinate construct of sexual prejudice, they constitute distinct psychological dimensions that warrant independent scoring.

Confirmatory Factor Analyses and Structural Sub-dimensions

Subsequent CFAs across multicultural samples have evaluated alternative structural models:

1. Unidimensional Model: A single global “heterosexism” factor. This model routinely yields inadequate goodness-of-fit indices (e.g., $\text{CFI} < .80$, $\text{RMSEA} > .10$), confirming that collapsing across target genders obscures meaningful variance.
2. Two-Factor Correlated Model: The standard ATL and ATG bifurcation. This model consistently achieves superior fit across diverse populations ($\text{CFI} > .92$, $\text{TLI} > .90$, $\text{RMSEA} < .06$).
3. Hierarchical / Multidimensional Sub-factor Models: Several structural analyses have identified secondary factors nested within both the ATL and ATG subscales. Specifically, each subscale tends to bifurcate into: (a) a Condemnation/Tolerance factor (focusing on sin, sickness, moral decay, and emotional disgust), and (b) a Social/Civil Rights factor (focusing on employment discrimination, legal protections, and adoption rights).

Measurement invariance analyses across gender groups confirm metric and scalar invariance, demonstrating that the ATLG measures identical latent constructs in both male and female respondents, permitting direct, unbiased cross-gender comparisons.

Instrument / Measurement Tool

The standard ATLG is a paper-and-pencil or computerized self-report psychometric instrument. Below is the structural profile and scoring architecture of the instrument:

• Test Type: Multidimensional self-report attitude scale.
• Target Population: Heterosexual adolescents and adults (standardized primarily on individuals aged 18 and older).
• Administration Format: Individual or group administration; available via paper-and-pencil or online psychometric surveys.
• Administration Time: Approximately 5 to 10 minutes.
• Item Count: 20 items total.
  • Attitudes Toward Lesbians (ATL) Subscale: Items 1 through 10.
  • Attitudes Toward Gay Men (ATG) Subscale: Items 11 through 20.
• Response Scale: 9-point Likert scale:
  • 1 = Strongly Agree
  • 2 = Moderately Agree
  • 3 = Slightly Agree
  • 4 = Neutral / Unsure (or Mildly Agree depending on variant)
  • 5 = Neutral / Unsure
  • 6 = Mildly Disagree
  • 7 = Slightly Disagree
  • 8 = Moderately Disagree
  • 9 = Strongly Disagree

  (Note: In the original scoring key, numerical values are assigned such that higher scores reflect greater sexual prejudice / negative attitudes toward gay people).

• Scoring and Directionality:
  • Directionality Convention: Standard psychometric convention for the ATLG codes higher total scores as indicating greater sexual prejudice (more negative attitudes).
  • Negatively Worded / Prejudicial Items (Keyed Directly):
    • ATL Items: 1, 3, 5, 6, 8, 9, 10.
    • ATG Items: 12, 13, 14, 16, 18, 19.

    If respondents rate these items with 1 (“Strongly Agree”) on a 1-to-9 scale where 1 indicates agreement, their values are reverse-transformed (or scored as 9 for strong agreement) so that endorsement of anti-gay statements corresponds to maximum prejudice.

  • Positively Worded / Egalitarian Items (Reverse Scored):
    • ATL Items: 2, 4, 7.
    • ATG Items: 11, 15, 17, 20.

    Strong agreement with these items indicates non-prejudicial, egalitarian attitudes, and scores are inverted accordingly.

• Subscale and Composite Scoring:
  • ATL Score: Sum or mean of transformed Items 1 through 10 (Range: 10 to 90 for sum scoring on a 9-point scale).
  • ATG Score: Sum or mean of transformed Items 11 through 20 (Range: 10 to 90 for sum scoring on a 9-point scale).
  • Total ATLG Score: Sum of ATL and ATG scores (Range: 20 to 180). Alternatively, researchers commonly calculate an average item score ranging from 1 to 9.

Permissions & Fee and Test Year

The Attitude toward Lesbians and Gay Men Scale was formally published by Dr. Gregory M. Herek in 1984 in the Journal of Homosexuality. Dr. Herek dedicated the ATLG to non-commercial academic research and scholarly inquiry. Qualified researchers, university students, and non-profit educational institutions are permitted to reproduce and administer the scale without monetary royalty fees, provided that appropriate scholarly attribution is cited and the items are not modified without documentation. Commercial use, inclusion in fee-charging diagnostic platforms, or distribution in for-profit proprietary environments requires formal permission from the copyright holder and publisher. Additional historical documentation, psychometric scoring algorithms, and international short-form variants are maintained via university archives and Dr. Herek’s official academic research repositories.

References

• Herek, G. M. (1984). Attitudes toward lesbians and gay men: A factor-analytic study. Journal of Homosexuality, 10(1–2), 39–51. https://doi.org/10.1300/J082v10n01_03
• Herek, G. M. (1984). Beyond “homophobia”: A social psychological perspective on attitudes toward lesbians and gay men. Journal of Homosexuality, 10(1–2), 1–21. https://doi.org/10.1300/J082v10n01_01
• Herek, G. M. (1986a). The instrumentality of attitudes: Toward a neofunctional theory. Journal of Social Issues, 42(2), 99–114. https://doi.org/10.1111/j.1540-4560.1986.tb00227.x
• Herek, G. M. (1986b). On heterosexual masculinity: Some psychodynamics of double and triple jeopardy. American Behavioral Scientist, 29(5), 563–577. https://doi.org/10.1177/000276486029005005
• Herek, G. M. (1987). Can functions be measured? A new perspective on the functional approach to attitudes. Social Psychology Quarterly, 50(4), 285–303. https://doi.org/10.2307/2786814
• Herek, G. M. (1988). Heterosexuals’ attitudes toward lesbians and gay men: Correlates and gender differences. The Journal of Sex Research, 25(4), 451–477. https://doi.org/10.1080/00224498809551476
• Herek, G. M. (1990). The context of anti-gay violence: Notes on cultural and psychological heterosexism. Journal of Interpersonal Violence, 5(3), 316–333. https://doi.org/10.1177/088626090005003006
• Herek, G. M. (2002). Gender gaps in public opinion about lesbians and gay men. Public Opinion Quarterly, 66(1), 40–66. https://doi.org/10.1086/338629
• Tureau, Z. L. (2003). College Student Identity and Attitudes Toward Gays and Lesbians (Doctoral dissertation, University of North Texas). UNT Digital Library. https://digital.library.unt.edu/ark:/67531/metadc4286/

Items of the Scale