Personal Attributes Questionnaire

Abstract

The Personal Attributes Questionnaire (PAQ) is a seminal psychometric instrument developed by Janet Taylor Spence, Robert L. Helmreich, and Joy Stapp (1974, 1975) to assess self-ascribed personality characteristics traditionally associated with masculinity and femininity. Originally designed as a 55-item inventory and subsequently refined into an empirically robust 24-item short form, the PAQ challenged the historic psychometric assumption that masculinity and femininity reside at opposite poles of a single unidimensional continuum. Instead, the instrument conceptualizes these dimensions as two orthogonal, socially desirable trait clusters: Instrumentality (traditionally masculine traits emphasizing agency, self-assertion, goal-directed competence, and independence) and Expressiveness (traditionally feminine traits emphasizing communion, interpersonal warmth, emotional sensitivity, and nurturance). The 24-item version comprises three distinct 8-item subscales: the Masculinity (M) scale, the Femininity (F) scale, and the Masculinity-Femininity (M-F) scale. Items are rated along a 5-point bipolar semantic differential scale anchored by opposing descriptive statements (scored from 0 to 4).

Psychometric evaluations across diverse university, adolescent, and adult populations demonstrate that the PAQ possesses solid internal consistency reliability, with Cronbachu2019s alpha coefficients typically ranging from .78 to .85 across subscales, and correlations between the 24-item short form and the original 55-item scale exceeding .91. Extensive construct, convergent, and discriminant validity analyses confirm that the M and F scales capture distinct behavioral dispositions that correlate predictably with generalized self-esteem, interpersonal competence, stress coping, and psychological well-being, while remaining conceptually distinct from explicit gender-role attitudes or biological sex. By offering a psychometrically rigorous operationalization of David Bakanu2019s agency and communion paradigms, the PAQ continues to serve as an indispensable measurement tool in social, personality, clinical, developmental, and organizational psychology.

Keywords

Personal Attributes Questionnaire, Janet Taylor Spence, Instrumentality, Expressiveness, Agency and Communion, Gender-Role Orientation, Psychological Androgyny, Masculinity, Femininity, Psychometrics, Self-Concept, Sex-Role Stereotypes

Authors

The Personal Attributes Questionnaire was developed by a team of prominent personality and social psychologists affiliated with the Department of Psychology at the University of Texas at Austin:

• Janet Taylor Spence, Ph.D. (1923u20132015): A pioneer in psychological assessment, anxiety research (creator of the Taylor Manifest Anxiety Scale), and the psychology of gender. Dr. Spence served as President of the American Psychological Association (APA, 1984) and was the founding President of the Association for Psychological Science (APS, 1988u20131989). Her foundational theoretical work redefined how social sciences operationalize gender identity, sex-role orientations, and achievement motivation.
• Robert L. Helmreich, Ph.D. (1937u20132012): Professor of Psychology at the University of Texas at Austin and a world-renowned authority on human factors, crew resource management (CRM), group performance under extreme environments (e.g., aerospace and aviation safety), and personality assessment.
• Joy Stapp, Ph.D.: Research psychologist and statistician at the University of Texas at Austin and the American Psychological Association, recognized for her extensive methodological and psychometric contributions to gender-role assessment, social attribution, and demographic research in behavioral science.

Purpose

The primary purpose of the Personal Attributes Questionnaire is to measure self-conceptions of gender-typed personal traits along independent, socially valued dimensions, specifically separating self-assertive instrumental characteristics from interpersonal expressive characteristics. Throughout the mid-twentieth century, standard psychological instruments (such as the Masculinity-Femininity scales of the Minnesota Multiphasic Personality Inventory [MMPI] and the California Psychological Inventory [CPI]) operated under the implicit psychometric axiom that masculinity and femininity were mutually exclusive, bipolar endpoints of a single continuum. Under this legacy framework, an increase in masculine traits inherently signaled a deficit in feminine traits, pathologizing individuals who exhibited multidimensional or gender-atypical behavioral repertoires.

Spence, Helmreich, and Stapp (1974, 1975) formulated the PAQ to provide a psychometrically sound, theoretically coherent alternative capable of measuring these traits independently. Rather than evaluating adherence to socio-political gender norms or institutionalized gender roles, the PAQ specifically assesses internal self-descriptorsu2014personality traits that university men and women empirically perceive as socially desirable across both sexes, but which normative cultural stereotypes attribute to one sex more than the other. The theoretical rationale rests on disentangling personality traits (instrumentality and expressiveness) from gender-role attitudes (such as views evaluated by the Attitudes toward Women Scale [AWS]), gender-typed behaviors, and biological sex.

In clinical and counseling psychology, the PAQ has been extensively applied to assess individual differences in self-esteem, interpersonal conflict resolution, marital adjustment, and affective resilience. In industrial-organizational psychology, researchers utilize the instrument to investigate leadership styles, workplace self-efficacy, managerial competence, and organizational citizenship behaviors. Furthermore, in developmental, health, and social research, the scale facilitates the empirical investigation of how gendered self-schemata influence coping mechanisms under chronic stress, health-seeking behaviors, compassion, and reactions to gender-role conflict.

Psychological Construct

The PAQ operationalizes personality self-concept across three differentiated dimensions, derived from rigorous empirical item generation and social desirability ratings:

1. The Masculinity (M) Scale: Instrumentality

The Masculinity scale captures an individualu2019s self-perceived instrumentality or agency. Items comprising this scale represent personality characteristics that are deemed socially desirable for both men and women in contemporary society, yet are stereotypically attributed to, and normative within, adult males. Instrumentality reflects a self-assertive, goal-oriented, and autonomous orientation toward the external environment. Core facets include:

• Independence and Decisiveness: The capacity to make autonomous judgments, formulate plans without undue reliance on external validation, and maintain self-direction (e.g., “Can make decisions easily”, “Not at all independent … Very independent”).
• Perseverance and Resilience: Psychological tenacity in the face of obstacles, sustained task commitment, and affective stability under environmental strain (e.g., “Never gives up easily”, “Stands up well under pressure”).
• Self-Confidence and Assertiveness: A pervasive conviction in oneu2019s personal efficacy, competence, and ability to influence social dynamics constructively (e.g., “Very self-confident”, “Feels very superior”).

2. The Femininity (F) Scale: Expressiveness

The Femininity scale captures self-perceived expressiveness or communion. Like the M scale, the items on the F scale represent characteristics judged to be socially desirable in both men and women, but which are stereotypically attributed in higher degree to adult females. Expressiveness reflects an affective, interpersonal orientation focused on connection, relational harmony, empathy, and social responsiveness. Core facets include:

• Empathy and Emotional Sensitivity: Attunement to the affective states, distress, and subjective experiences of other individuals (e.g., “Very aware of feelings of others”, “Very understanding of others”).
• Interpersonal Warmth and Kindness: Altruistic concern, relational openness, and emotional approachability (e.g., “Very warm in relations with others”, “Very kind”).
• Prosocial Devotion: A willingness to provide support, nurturance, and personal investment into the welfare of social groups or loved ones (e.g., “Able to devote self completely to others”, “Very helpful to others”).

3. The Masculinity-Femininity (M-F) Scale

The Masculinity-Femininity scale represents an exploratory, distinct psychometric dimension composed of items whose perceived social desirability differs according to biological sex. Unlike the M and F scalesu2014where traits are considered universally positive for both gendersu2014items on the M-F scale describe traits that are socially condoned or rewarded in men but socially discouraged in women, or vice versa (e.g., aggression, vulnerability, emotionality). High scores on this scale denote a constellation of characteristics traditionally sanctioned in males (e.g., aggressiveness, competitiveness, emotional stoicism), whereas low scores reflect traits historically tolerated or expected primarily in females (e.g., emotional vulnerability, passivity, crying easily).

Theoretical Framework

The conceptual architecture of the Personal Attributes Questionnaire is anchored in the theoretical convergence of structural sociological theory and psychological metatheory regarding human personality dimensions.

Bakanu2019s Duality of Human Existence: Agency and Communion

Philosopher and psychologist David Bakan (1966) posited that all living human organisms must navigate two fundamental, pervasive modalities of existence: Agency and Communion. Agency manifests in the drive toward individual differentiation, mastery, self-protection, assertion, and control over the environment. Communion, conversely, manifests in the drive toward union, belonging, relational intimacy, cooperation, and the integration of the self into a larger collective. Spence and colleagues mapped the cultural constructs of masculinity and femininity directly onto these universal modalities, proposing that what common vernacular termed “masculinity” was essentially personality instrumentality (agency), while “femininity” reflected personality expressiveness (communion).

Parsons and Bales: Functional Role Differentiation

In mid-century sociological functionalism, Talcott Parsons and Robert F. Bales (1955) posited that social groups require two complementary types of leadership and functional roles to survive: instrumental roles (task-oriented, problem-solving, environmental adaptation) and expressive roles (internal maintenance, emotional equilibrium, group solidarity). Spence and Helmreich argued that while social roles might historically have exhibited gender-typed division of labor, individual human beings possess psychological capacities for both instrumental competence and expressive warmth simultaneously. They explicitly decoupled functional personality traits from fixed societal sex roles.

The Androgyny Paradigm and Spenceu2019s Critical Clarification

During the early 1970s, the paradigm of psychological androgynyu2014championed concurrently by Sandra Bem (Bem Sex-Role Inventory, BSRI, 1974) and Spence, Helmreich, and Stapp (1975)u2014revolutionized gender scholarship. Under the four-category typology derived from median splits of independent M and F scores, individuals can be categorized into four distinct profiles:

• Androgynous: High Instrumentality, High Expressiveness
• Masculine-Typed: High Instrumentality, Low Expressiveness
• Feminine-Typed: Low Instrumentality, High Expressiveness
• Undifferentiated: Low Instrumentality, Low Expressiveness

Crucially, Spence maintained a more conservative, psychometrically rigorous interpretation of these scores than Bem. While Bem initially argued that androgyny represented a global cognitive filter (“gender schema theory”), Spence consistently demonstrated that the PAQ measures a specific, bounded subset of instrumental and expressive personality traits. Spence cautioned researchers against assuming that high scores on the PAQu2019s M and F scales automatically indexed generalized “sex-role identity”, feminist attitudes, sexual orientation, or adherence to domestic roles. This conceptual precision established the PAQ as an empirically stable, construct-valid measure of trait agency and communion.

Validity

The psychometric validity of the PAQ has been substantiated through dozens of empirical investigations across five decades, examining construct, convergent, discriminant, and predictive paradigms.

Construct and Factorial Validity

Construct validity was established during initial scale construction through extensive normative evaluation across high school, university, and community adult cohorts (Spence & Helmreich, 1978). In a foundational validation study involving over 715 university students, male and female participants demonstrated statistically significant mean differences on virtually every item in the predicted direction: men scored significantly higher on Masculinity and Masculinity-Femininity items, while women scored significantly higher on Femininity items ($p < .001$). Factor analytic confirmation verified the structural independence of the M and F subscales, supporting the core theoretical postulate that instrumentality and expressiveness operate as orthogonal constructs.

Convergent Validity

The convergent validity of the PAQ is supported by robust correlations with alternative gender-trait instruments and generalized personality measures:

• Bem Sex-Role Inventory (BSRI): Correlations between the PAQ Masculinity scale and the BSRI Masculinity scale typically range between $r = .75$ and $r = .85$; correlations between the PAQ Femininity scale and BSRI Femininity scale fall between $r = .65$ and $r = .78$, confirming high conceptual concordance despite differences in scale anchors and item selection.
• Self-Esteem and Psychological Efficacy: The M scale correlates substantially with the Rosenberg Self-Esteem Scale (RSES) ($r pprox .50u2013.62$) and the Texas Social Behavior Inventory (TSBI; $r pprox .60u2013.70$), demonstrating that self-ascribed instrumentality is strongly associated with social confidence, perceived competence, and generalized self-worth. Expressiveness (F) also demonstrates positive, moderate correlations with self-esteem ($r pprox .25u2013.35$), with androgynous individuals consistently demonstrating the highest self-esteem profiles.

Discriminant Validity

The PAQ demonstrates excellent discriminant validity when examined against sociopolitical and ideological measures. Specifically, correlations between PAQ scores (both M and F) and the Attitudes toward Women Scale (AWS) are virtually zero ($r < .10$), confirming that possessing an instrumental or expressive personality does not determine or reflect one’s political views regarding legal, economic, or marital rights of women. Furthermore, the scales exhibit minimal vulnerability to generalized social desirability response bias when controlling for the substantive desirability inherent to the traits themselves.

Predictive and Criterion Validity

Empirical studies consistently confirm the PAQu2019s capacity to predict psychological and behavioral outcomes:

• Mental Health and Coping: High instrumentality (M) negatively predicts depressive symptomatology, trait anxiety, and learned helplessness, functioning as an internal buffer against environmental stressors. Expressiveness (F) uniquely predicts relational quality, empathy, social support utilization, and marital satisfaction.
• Conflict Resolution: Research in adolescent and adult conflict management (e.g., Keener, 2010) demonstrates that expressive traits predict cooperative and integrative negotiation strategies, whereas instrumental traits predict direct assertiveness and problem-focused resolution.
• Self-Compassion and Caregiving: Studies examining clinical and subclinical populations (e.g., Tatum, 2012) reveal that the combination of high M and high F (androgyny) optimizes both self-directed compassion and other-oriented caregiving efficacy.

Reliability

The Personal Attributes Questionnaire has consistently exhibited solid reliability across diverse psychometric investigations:

Internal Consistency

In the seminal standardization cohorts (Spence & Helmreich, 1978), Cronbachu2019s alpha coefficients for the 24-item short form demonstrated robust internal consistency across independent samples:

• Masculinity (M) Scale: $\alpha = .85$
• Femininity (F) Scale: $\alpha = .82$
• Masculinity-Femininity (M-F) Scale: $\alpha = .78$

Subsequent psychometric investigations across college, adolescent, and clinical cohorts have corroborated these values, with alpha coefficients for the M scale typically hovering between $.80$ and $.86$, and the F scale ranging between $.78$ and $.84$. The M-F scale, reflecting a broader constellation of evaluatively divergent items, typically yields alphas between $.72$ and $.79$.

Short-Form to Long-Form Equivalence

The psychometric reduction from the original 55-item inventory to the 24-item operational form was executed with minimal loss of measurement fidelity. Bivariate Pearson correlations between the 8-item subscales and their parent 55-item counterparts confirmed exceptional equivalence:

• Short M to Original M: $r = .93$
• Short F to Original F: $r = .93$
• Short M-F to Original M-F: $r = .91$

Test-Retest Stability

Stability coefficients evaluated across test-retest intervals ranging from four to thirteen weeks demonstrate temporal consistency, with coefficients generally exceeding $r_{tt} = .80$ for the M and F subscales among adult cohorts. The stability of the scores supports the theoretical premise that the PAQ measures enduring personality traits rather than fluctuating emotional states or situational attitudes.

Factor Analysis

The internal structural integrity of the PAQ has been rigorously vetted through both exploratory factor analysis (EFA) and confirmatory factor analysis (CFA).

Exploratory Factor Structure

Early principal components and common factor analyses with varimax (orthogonal) rotations performed by Spence, Helmreich, and Stapp consistently yielded two dominant, non-overlapping dimensions corresponding to Instrumentality and Expressiveness:

• Factor 1 (Expressiveness / Communion): Dominated by robust loadings ($lambda > .45$) from items such as “Very warm in relations with others”, “Very understanding of others”, “Able to devote self completely to others”, “Very helpful to others”, and “Very gentle”. These items load negligibly on the instrumental dimension.
• Factor 2 (Instrumentality / Agency): Dominated by robust loadings ($lambda > .45$) from items including “Never gives up easily”, “Very self-confident”, “Can make decisions easily”, “Stands up well under pressure”, and “Very active”. Cross-loadings onto the expressive dimension remain consistently trivial ($< .20$).

Confirmatory Factor Analysis and Structural Independence

Contemporary CFA investigations using modern estimation techniques (e.g., maximum likelihood, robust ML) support the two-factor orthogonal model for the M and F scales. Structural equation models specifying two latent factorsu2014Agency (M) and Communion (F)u2014consistently demonstrate acceptable to superior fit indices ($CFI > .90$, $TLI > .90$, $RMSEA < .06$) when evaluated across separate gender groups. Crucially, the latent correlation between the Instrumental and Expressive latent factors typically hovers near zero (varying from $r = -.05$ to $r = .12$), empirically proving the orthogonal, independent structural organization of the two core trait complexes.

Items belonging to the exploratory M-F subscale often exhibit bifurcated structural behavior, with certain items aligning with instrumental dominance/toughness and others aligning negatively with emotional reactivity. Consequently, most contemporary personality and clinical investigations focus primarily on the two independent M and F scales to model agency and communion.

Instrument / Measurement Tool

• Instrument Name: Personal Attributes Questionnaire (PAQ) (also referenced historically in expanded format as the Extended Personal Attributes Questionnaire, EPAQ)
• Authors: Janet Taylor Spence, Robert L. Helmreich, and Joy Stapp
• Year of Formal Publication: 1974 (original 55-item scale); 1975 (24-item short form)
• Target Population: Adolescents and adults (ages 14+)
• Administration Format: Self-report paper-and-pencil or interactive digital computer-based questionnaire
• Item Count: 24 items in the standardized short form (8 items per subscale)
• Subscales (8 items each):
  • Masculinity (M) Scale: Measures socially desirable instrumentality and agency.
  • Femininity (F) Scale: Measures socially desirable expressiveness and communion.
  • Masculinity-Femininity (M-F) Scale: Measures gender-differentiated social traits.
• Response Scale: 5-point bipolar semantic differential scale (scored 0 to 4, anchored by opposing descriptive phrases at the extreme ends: A = 0, B = 1, C = 2, D = 3, E = 4).
• Scoring and Directionality:
  • Items are scored from 0 to 4 (or 1 to 5 depending on coding convention). High scores always reflect a high level of the construct indicated by the subscale name.
  • Reversed items must be recoded prior to computing aggregate subscale sums so that the pole denoting high instrumentality or high expressiveness receives the highest numerical value.
  • Masculinity (M) Scale Items: 4, 5, 13, 15, 16, 17, 18, 19
  • Femininity (F) Scale Items: 3, 7, 8, 9, 10, 14, 21, 23
  • Masculinity-Femininity (M-F) Scale Items: 1, 2, 6, 11, 12, 20, 22, 24
• Categorization Typology: Through a median-split methodology based on normative sample medians, individuals can be classified into four psychological categories: Androgynous (High M, High F), Masculine-Typed (High M, Low F), Feminine-Typed (Low M, High F), and Undifferentiated (Low M, Low F). Contemporary researchers also utilize continuous regression/interaction models without dichotomization.
• Completion Time: Approximately 5 to 10 minutes.

Permissions & Fee and Test Year

The Personal Attributes Questionnaire was first published in 1974 via the JSAS Catalog of Selected Documents in Psychology (Spence, Helmreich, & Stapp, 1974) and subsequent open scientific articles in the Journal of Personality and Social Psychology (1975) and their monograph (1978). The authors placed the instrument into the public scientific domain to foster non-commercial academic, psychological, and clinical research. No commercial licensing fee or proprietary publisher permission is required for non-profit academic or therapeutic use, provided that appropriate scholarly attribution is accorded to Dr. Janet T. Spence and colleagues.

References

• Bakan, D. (1966). The duality of human existence: An essay on psychology and religion. Rand McNally.
• Bem, S. L. (1974). The measurement of psychological androgyny. Journal of Consulting and Clinical Psychology, 42(2), 155u2013162. https://doi.org/10.1037/h0036215
• Coleman, H. L. (2009). The personality traits of instrumentality and expressiveness in relation to microcomputer playfulness (Doctoral dissertation, University of Texas at Austin). UT Electronic Theses and Dissertations.
• Gerhardstein, K. R. (2010). Attitudes toward transsexual people: Effects of gender and appearance (Doctoral dissertation, Indiana State University). ProQuest Dissertations Publishing.
• Keener, E. (2010). Boysu2019 and girlsu2019 strategies for managing peer conflict in adolescence (Doctoral dissertation, West Virginia University). The Research Repository @ WVU.
• Parsons, T., & Bales, R. F. (1955). Family, socialization and interaction process. Free Press.
• Persky, D. R. (2010). The ethical orientation of advertising executives: The relationship among ethical climate, sex role identity, and selected demographic and job variables (Doctoral dissertation, New York University). ProQuest Dissertations Publishing.
• Spence, J. T., & Helmreich, R. L. (1978). Masculinity and femininity: Their psychological dimensions, correlates, and antecedents. University of Texas Press.
• Spence, J. T., & Helmreich, R. L. (1979). The many faces of androgyny: A reply to Locksley and Colten. Journal of Personality and Social Psychology, 37(6), 1032u20131046. https://doi.org/10.1037/0022-3514.37.6.1032
• Spence, J. T., Helmreich, R. L., & Stapp, J. (1974). The Personal Attributes Questionnaire: A measure of sex-role stereotypes and masculinity-femininity. JSAS Catalog of Selected Documents in Psychology, 4, 43 (Ms. No. 617).
• Spence, J. T., Helmreich, R. L., & Stapp, J. (1975). Ratings of self and peers on sex role attributes and their relation to self-esteem and conceptions of masculinity and femininity. Journal of Personality and Social Psychology, 32(1), 29u201339. https://doi.org/10.1037/h0076857
• Tatum, K. J. (2012). Adherence to gender roles as a predictor of compassion and self-compassion in women and men (Doctoral dissertation, Baylor University). Baylor University Libraries.

Items of the Scale