Attitude Function Inventory (AFI)

Abstract

The Attitude Function Inventory (AFI) is a psychometric instrument developed by social psychologist Gregory M. Herek to assess the motivational bases and psychological functions underlying individuals’ attitudes toward minority social groups, specifically gay men and lesbians. While traditional attitudinal measures quantify the valence or extremity of an attitude (ranging from favorable to unfavorable), the AFI operationalizes a functional approach to attitudes, identifying why an attitude is held and what psychological needs it fulfills for the individual. The instrument captures four distinct motivational functions rooted in neofunctional attitude theory: (a) Experiential-Schematic, which derives attitudes from direct, personal interpersonal contact and interactions with gay persons; (b) Social-Expressive, which serves needs for social affiliation, belonging, and conforming to the norms of respected reference groups; (c) Defensive (or ego-defensive), which serves to manage internal anxiety, discomfort, psychological conflict, or perceived threats to self-identity; and (d) Value-Expressive, which allows the individual to assert core personal values, justice principles, or foundational moral and ideological worldviews.

The standard self-report version of the AFI comprises 10 core items scored on a 9-point continuum anchored by 1 (Very True of Me) and 9 (Not at all True of Me). Across empirical studies, internal consistency estimates vary across the four subscales, with Cronbach’s alpha values typically ranging from α = .67 to .82 for the Experiential-Schematic dimension, α = .41 to .62 for the Social-Expressive dimension, α = .72 to .82 for the Defensive dimension, and α = .53 to .61 for the Value-Expressive dimension. Construct and discriminant validity have been substantiated through factor-analytic investigations and differential associations with the Attitudes Toward Lesbians and Gay Men (ATLG) scale, measures of right-wing authoritarianism, religious fundamentalism, and objective behavioral markers of intergroup contact. The AFI serves as a foundational assessment tool in prejudice reduction research, persuasive communication design, and socio-cognitive research exploring the motivational mechanisms of sexual stigma.

Keywords

Attitude Function Inventory, AFI, Gregory M. Herek, functional attitude theory, sexual prejudice, experiential-schematic function, ego-defensive function, value-expressive function, social-expressive function, attitude functions, homophobia measurement, psychometrics

Authors

The Attitude Function Inventory was formulated and validated by Gregory M. Herek, Ph.D., Professor Emeritus of Psychology at the University of California, Davis. Dr. Herek is internationally recognized as a leading authority on sexual prejudice, hate crimes, heterosexism, and the psychological health of sexual minorities. Throughout his academic career, Herek integrated principles of cognitive social psychology with clinical and personality theories to demystify the psychological architecture of anti-gay bias. His theoretical and psychometric work was conducted in affiliation with the Department of Psychology at UC Davis, Yale University, and the City University of New York, supported by grants from the National Institute of Mental Health (NIMH).

Purpose

For decades, social psychological inquiry into intergroup bias was dominated by instruments designed to quantify the direction and intensity of prejudice—delineating whether an individual held positive, neutral, or negative attitudes. However, as Herek (1986a, 1987) argued, treating attitudes solely as unidimensional evaluative summaries fails to capture their psychological utility. Two individuals may possess identical negative scores on an explicit prejudice measure, yet their attitudes may stem from entirely incompatible psychological needs: one may seek to defend against latent psychological conflict or threats to traditional gender roles (an ego-defensive function), while the other may simply be conforming to the prevailing social norms of their immediate community (a social-expressive function).

The primary purpose of the Attitude Function Inventory is to operationalize and measure these underlying functional dynamics. By determining the specific motivational basis of an individual’s attitude toward sexual minorities, the AFI addresses several critical theoretical and practical objectives:

• Diagnostic Differentiation in Social Psychology: The AFI allows researchers to disaggregate broad evaluative constructs (such as “homophobia” or sexual prejudice) into functionally discrete cognitive-motivational categories, illuminating the psychological heterogeneity among individuals who exhibit similar overt evaluative stances.
• Targeted Persuasion and Intervention Design (The Matching Hypothesis): Neofunctional theory posits that attitude change interventions are most effective when the persuasive appeal matches the functional basis of the underlying attitude. The AFI enables researchers and educators to empirically test the “matching hypothesis”—for example, evaluating whether value-expressive attitudes are more receptive to moral and civil-liberties messaging, while defensive attitudes require anxiety-reduction and self-affirmation techniques.
• Intergroup Contact Research: The inventory provides an empirical method to measure the psychological integration of intergroup contact. Rather than merely documenting the frequency of contact, the AFI’s experiential-schematic scale evaluates whether personal contact has been cognitively organized into an active schema guiding the person’s evaluations.
• Public Policy and Sociological Analysis: In applied survey research, understanding the distribution of attitude functions across diverse demographic and ideological cohorts informs public education campaigns regarding civil rights, anti-discrimination legislation, and hate crime prevention.

Psychological Construct

The Attitude Function Inventory is built upon the construct of attitude functionality—the conceptual principle that individuals maintain cognitive, affective, and behavioral orientations toward attitude objects because doing so achieves specific intrapsychic or interpersonal goals. Rather than viewing attitudes as passive reflections of objective information, the functional perspective conceptualizes them as active, utilitarian tools. The AFI delineates four primary constructs within this domain:

1. Experiential-Schematic Function

The experiential-schematic function captures attitudes that originate directly from personal, concrete interactions with members of the target group. When an attitude serves this function, the attitude object is mentally represented through cognitive schemas structured around real-life episodes, interpersonal relationships, and behavioral exchanges. For instance, an individual whose brother, close colleague, or longtime neighbor is gay organizes their attitudes around those specific human interactions rather than abstract cultural stereotypes or secondhand dogmas. This function reflects the operationalization of the contact hypothesis, serving an informational and schematic efficiency purpose where individual experiences guide broader social generalizations.

2. Social-Expressive Function

The social-expressive function relates to an individual’s need for social belonging, integration, and social adjustment within valued peer groups, families, and communities. Attitudes functioning in this capacity are held because expressing them fosters interpersonal acceptance, avoids ostracism, or affirms alignment with respected reference groups. The individual adopts the attitudes perceived as dominant or normative among significant others. In this context, an unfavorable or favorable opinion about sexual minorities is maintained primarily because it acts as social currency, signaling solidarity with family expectations, religious bodies, or peer networks.

3. Defensive (Ego-Defensive) Function

Rooted in psychodynamic and social defense mechanisms, the defensive function posits that attitudes can act as psychological shields against internal anxiety, unconscious intrapsychic conflict, feelings of vulnerability, or threats to personal self-esteem. In the context of attitudes toward gay men and lesbians, Herek (1986b) observed that defensive attitudes often emerge from anxieties surrounding rigid gender roles, heterosexual masculinity, or visceral feelings of disgust and revulsion. Holding negative attitudes allows the individual to project unacceptable impulses outward, suppress threatening thoughts, and preserve an unblemished self-concept by establishing rigid boundaries against the perceived “other.”

4. Value-Expressive Function

The value-expressive function allows individuals to assert and validate their foundational self-concept, core personal values, philosophical tenets, and moral convictions. Rather than defending against anxiety or conforming to peers, the attitude is an outward vehicle for communicating “who I am” and “what I stand for.” This function operates bi-directionally: an individual with positive attitudes may derive them from an overarching commitment to universal civil liberties, social justice, and constitutional equality, whereas an individual with negative attitudes may base them on explicit, deeply internalized religious or moral doctrines regarding divine order and natural law.

Theoretical Framework

The Attitude Function Inventory is grounded in the tradition of functional attitude theory, originally pioneered in mid-20th-century social psychology by Daniel Katz (1960) and M. Brewster Smith, Jerome S. Bruner, and Robert W. White (1956). Early theorists posited that human attitudes cannot be comprehended solely by analyzing cognitive processing capacity or conditioning paradigms; they must be understood within the teleological framework of human motivation.

Historical Antecedents: Katz and Smith, Bruner, & White

Katz (1960) proposed four classical functions of attitudes: instrumental/utilitarian (maximizing rewards, minimizing punishments), ego-defensive (protecting against internal conflict and external threats), value-expressive (expressing central values and self-identity), and knowledge (providing structure, consistency, and meaning to a chaotic universe). Similarly, Smith et al. (1956) identified object appraisal, social adjustment, and externalization. Although intuitively compelling, early functional theories suffered from significant psychometric challenges: researchers struggled to reliably measure functions independently of attitude valence, leading to a temporary decline in empirical functional research during the cognitive revolution of the 1970s.

Herek’s Neofunctional Synthesis

In the mid-1980s, Gregory Herek revived and refined this paradigm by formulating a neofunctional model of attitudes (Herek, 1986a, 1987). Herek introduced critical theoretical distinctions to overcome previous empirical roadblocks:

• Separation of Content and Function: Herek demonstrated that an attitude’s evaluative direction (positive vs. negative) is conceptually independent of its psychological function. A positive attitude can be value-expressive (based on egalitarianism) or social-expressive (based on conforming to liberal peer norms); a negative attitude can be defensive (warding off gender insecurity) or value-expressive (rooted in conservative theological beliefs).
• Typology of Functional Domains: Herek synthesized earlier typologies into three broad functional categories: evaluative functions (appraising the object’s inherent utility, further refined into experiential and instrumental needs), expressive functions (social-expressive and value-expressive), and defensive functions (externalizing internal threats).
• The AFI Operationalization: The Attitude Function Inventory was specifically constructed to provide an empirical methodology for measuring these functional motivations via direct self-report, demonstrating that individuals possess sufficient metacognitive awareness to report the primary reasons underlying their social evaluations.

Validity

Extensive psychometric investigations have evaluated the construct, convergent, discriminant, and predictive validity of the Attitude Function Inventory across diverse undergraduate, community, and national samples.

Construct and Factorial Validity

Construct validity was initially established through exploratory and confirmatory factor analyses conducted by Herek (1987). When the 10 core items are evaluated, items consistently load onto their hypothesized latent factors—Experiential-Schematic, Social-Expressive, Defensive, and Value-Expressive. The extraction of four distinct dimensions confirmed that individuals distinguish between experiential sources of attitudes, peer-normative influences, defensive/avoidant emotional reactions, and overarching moral-philosophical value frameworks.

Convergent and Discriminant Validity

Convergent and discriminant validity have been confirmed through differential correlational patterns with established psychological measures:

• Attitudes Toward Lesbians and Gay Men (ATLG): The AFI subscales display theoretically expected patterns with Herek’s ATLG scale. The Defensive subscale correlates strongly and positively with elevated ATLG prejudice scores (often r > .60), confirming that visceral discomfort and cognitive avoidance are key drivers of anti-gay hostility. Conversely, the Experiential-Schematic subscale shows robust inverse correlations with ATLG scores when respondents report positive personal contact, confirming that personal intimacy with gay individuals is associated with lower prejudice.
• Authoritarianism and Dogmatism: Research demonstrates that the Social-Expressive and Defensive subscales correlate moderately with Right-Wing Authoritarianism (RWA) and religious fundamentalism, reflecting reliance on external authority structures and defense of traditional normative boundaries.
• Intergroup Contact Measures: The Experiential-Schematic subscale demonstrates strong convergent validity with objective behavioral measures of contact, such as the number of known gay friends, family members, or frequency of contact (Herek, 1988), while remaining distinct from purely ideological constructs.

Predictive and Experimental Validity (The Matching Hypothesis)

The most rigorous validation of the AFI stems from experimental studies testing persuasive communication matching. Social psychologists have utilized the AFI to pre-screen participants and subsequently expose them to persuasive messages framed in either value-expressive (e.g., highlighting civil liberties and justice) or social-expressive (e.g., highlighting social consensus and respected peer norms) terms. In accordance with functional theory, participants demonstrate significantly greater attitude shift when the framing of the persuasive communication matches their predominant AFI subscale profile, supporting the functional construct validity of the inventory.

Reliability

The psychometric evaluation of the Attitude Function Inventory indicates variable internal consistency coefficients across its four subscales, a characteristic documented in Herek’s foundational work and subsequent independent replications (e.g., Tureau, 2003):

• Experiential-Schematic Subscale (Items 1, 2, 3, 4): Demonstrates good internal consistency across diverse populations, with Cronbach’s alpha coefficients routinely falling between α = .67 and α = .82. The multi-item assessment of personal interaction history provides a cohesive measurement model.
• Social-Expressive Subscale (Items 5, 6): Yields modest reliability estimates, typically ranging from α = .41 to α = .62. This lower reliability is largely an artifact of scale length (consisting of only two primary items) and the nuanced distinction between peer approval and familial social influence. While adequate for group-level research, caution is recommended when using this subscale for individual-level diagnostic profiling.
• Defensive Subscale (Items 7, 8): Exhibits solid reliability coefficients, with Cronbach’s alpha values typically ranging between α = .72 and α = .82. The high inter-item correlation between cognitive avoidance (not wanting to think about homosexuality) and visceral affective discomfort reflects a unified ego-defensive reaction.
• Value-Expressive Subscale (Items 9, 10): Displays moderate internal consistency, with alpha coefficients historically ranging between α = .53 and α = .61. This moderate coefficient reflects the breadth of the underlying value construct, which incorporates both secular-humanitarian civil liberties concerns (Item 9) and traditional moral/normative imperatives (Item 10). When researchers separate egalitarian values from religious/moral values into distinct sub-facets, reliability estimates improve.

Test-retest stability coefficients assessed over 4- to 6-week intervals show satisfactory temporal stability (r values between .70 and .84), demonstrating that functional profiles represent enduring motivational orientations rather than transient situational states.

Factor Analysis

The latent structural integrity of the Attitude Function Inventory has been evaluated through exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) across several validation studies.

Exploratory Factor Analyses

In Herek’s (1987) seminal psychometric evaluation, principal axis factoring with both orthogonal (Varimax) and oblique (Promax/Oblimin) rotations was conducted on the 10-item inventory. The scree test and eigenvalue-greater-than-one criterion routinely supported a four-factor solution, corresponding directly to the theoretical neofunctional framework:

• Factor 1 (Experiential-Schematic): Items 1, 2, 3, and 4 exhibit prominent factor loadings, typically ranging from .58 to .84, with negligible cross-loadings onto other factors. These items capture attitudes grounded in concrete relational history and anticipated future interactions.
• Factor 2 (Defensive): Items 7 and 8 load robustly onto a distinct defensive factor, with primary loadings frequently exceeding .75. These items share minimal shared variance with experiential items, reflecting their origin in intrapsychic discomfort rather than relational contact.
• Factor 3 (Social-Expressive): Items 5 and 6 define this factor, displaying factor loadings between .60 and .79. These items reflect social conformity and reliance on respected others’ appraisals.
• Factor 4 (Value-Expressive): Items 9 and 10 load onto this final factor (loadings ranging from .52 to .71), capturing overarching ethical and civil-rights principles.

Confirmatory Factor Analyses and Model Fit

Subsequent structural equation modeling (SEM) and confirmatory factor analyses (e.g., Tureau, 2003) have tested alternative structural configurations, including a single-factor general function model, a two-factor model (internal/individual vs. external/social), and the hypothesized four-factor model. The four-factor oblique model consistently demonstrates superior fit to empirical data, yielding acceptable goodness-of-fit indices (e.g., Comparative Fit Index [CFI] > .92; Tucker-Lewis Index [TLI] > .90; Root Mean Square Error of Approximation [RMSEA] ≤ .06; Standardized Root Mean Square Residual [SRMR] ≤ .05). Oblique factor models show that while the four dimensions are distinct, modest inter-factor correlations exist (particularly between Defensive and Social-Expressive dimensions in conservative samples), aligning with theoretical expectations of shared socio-normative conservatism.

Instrument / Measurement Tool

• Full Instrument Name: Attitude Function Inventory (AFI)
• Original Author: Gregory M. Herek, Ph.D.
• Publication Era: 1986–1987
• Test Type: Self-report psychometric rating inventory
• Administration Format: Paper-and-pencil or computerized self-administered questionnaire
• Item Count: 10 core items
• Target Population: Adults, adolescents, and college student populations; primarily utilized in social psychological, sociological, and political attitude research
• Administration Time: Approximately 3 to 5 minutes
• Response Scale: 9-point Likert-type continuum, anchored at 1 (“Very True of Me”) and 9 (“Not at all True of Me”) (or reversed depending on computational direction)
• Subscale Structural Breakdown:
  • Experiential-Schematic: Items 1, 2, 3, 4 (contact-based and interaction-driven attitudes)
  • Social-Expressive: Items 5, 6 (social alignment, approval, reference-group conformity)
  • Defensive: Items 7, 8 (ego-defensive, avoidance, anxiety, revulsion)
  • Value-Expressive: Items 9, 10 (civil liberties, moral beliefs, personal values)
• Scoring and Computational Procedures:
  • Depending on researcher preference, items are either retained in their original 1–9 direction (where lower scores indicate higher functional endorsement) or reverse-coded (such that 9 indicates maximal endorsement: 1 = Not at all True of Me, 9 = Very True of Me).
  • Subscale scores are typically computed as the mean (or sum) of the constituent items for that dimension.
  • Because individuals often utilize multiple attitude functions simultaneously, the AFI can be analyzed as a multi-dimensional profile (evaluating relative elevation across all four subscales) or by categorizing participants according to their primary/dominant function.

Permissions & Fee and Test Year

The Attitude Function Inventory was developed and published in the mid-1980s, with foundational validation papers published in 1986 and 1987 in the Journal of Social Issues and Social Psychology Quarterly. The scale is considered an open-access academic measurement instrument intended for scholarly, scientific, and educational research. No commercial license fees or formal royalties are required for its non-commercial academic administration.

Scholars and researchers utilizing the AFI are expected to provide proper bibliographic attribution to Dr. Gregory M. Herek in all derived publications, dissertations, and research reports in accordance with standard APA ethical guidelines. Researchers adapting the inventory for attitudes toward other social groups or contexts must document any structural or lexical modifications in their methodologies.

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_02

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.tb00228.x

Herek, G. M. (1986b). On heterosexual masculinity: Some psychical consequences of the social construction of gender and sexuality. 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. 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/10.1086/338409

Katz, D. (1960). The functional approach to the study of attitudes. Public Opinion Quarterly, 24(2), 163–204. https://doi.org/10.1086/266945

Smith, M. B., Bruner, J. S., & White, R. W. (1956). Opinions and personality. John Wiley & Sons.

Tureau, Z. L. (2003). College student identity and attitudes toward gays and lesbians (Doctoral dissertation, University of North Texas). UNT Digital Library. http://digital.library.unt.edu/ark:/67531/metadc4286/

Items of the Scale