Sex Role Inventory (OSRI)

1. Abstract

The Open Sex Role Inventory (OSRI) is an open-source, empirical psychometric instrument designed to measure gender-role orientation, specifically operationalized across orthogonal dimensions of masculinity and femininity. Developed as an alternative to proprietary and historically dated gender identity inventories such as the Bem Sex-Role Inventory (BSRI) and the Personal Attributes Questionnaire (PAQ), the OSRI measures behavioral tendencies, personal preferences, social scripts, and expressive traits rather than strictly trait-adjective self-endorsements. The standard inventory comprises 44 self-report items evaluated on a 5-point response scale anchored from 1 (Disagree) through 3 (Neutral) to 5 (Agree). Rather than forcing a unidimensional bipolar continuum (masculine versus feminine), the OSRI retains Sandra Bem’s foundational two-dimensional model, enabling the classification of individuals into four distinct gender-role typologies: masculine (high masculinity, low femininity), feminine (low masculinity, high femininity), androgynous (high masculinity, high femininity), and undifferentiated (low masculinity, low femininity).

Extensive open-data psychometric evaluations spanning large-scale cross-sectional samples (totaling over 100,000 international internet respondents via the Open-Source Psychometrics Project) demonstrate strong internal consistency reliability, with Cronbach’s alpha coefficients typically ranging between .82 and .88 for the Masculine subscale and between .79 and .85 for the Feminine subscale. Confirmatory factor analyses (CFA) corroborate a two-factor orthogonal or modestly oblique structure that outperforms unidimensional representations. Construct, convergent, and discriminant validities are established through predictable associations with biological sex, occupational interests, sensation seeking, and the Big Five personality domains—specifically Extraversion, Conscientiousness, and Agreeableness. The OSRI represents an accessible, contemporary, and psychometrically robust measurement tool suitable for academic research in social psychology, gender studies, organizational behavior, and behavioral medicine.

2. Keywords

Open Sex Role Inventory, OSRI, gender roles, masculinity, femininity, androgyny, Bem Sex-Role Inventory, psychometrics, gender schema theory, factor analysis, construct validity, test reliability, sex-typed behavior

3. Authors

The Open Sex Role Inventory was developed and documented within the public domain by psychometric researcher Eric Jorgenson as part of the Open-Source Psychometrics Project. Jorgenson sought to construct open, freely accessible, empirically validated measures to counter the reliance on commercial and copyrighted instruments in behavioral science research. The historical theoretical framework underlying gender schema categorization referenced in connection with the OSRI stems from the pioneering work of developmental and cognitive psychologist Sandra Lipsitz Bem (1944–2014) of Cornell University, alongside encyclopedia syntheses by Thomas E. Malloy (2010).

4. Purpose

The primary purpose of the Open Sex Role Inventory (OSRI) is to provide an objective, contemporary, and psychometrically validated operationalization of psychological masculinity and femininity. In the mid-twentieth century, psychological measurement typically treated gender orientation as a single bipolar continuum anchored by hyper-masculinity at one extreme and hyper-femininity at the other. Under this outdated paradigm, high masculinity inherently precluded high femininity. Sandra Bem’s seminal work overturned this conceptualization by establishing that masculinity and femininity are independent, orthogonal dimensions, giving rise to the concept of psychological androgyny.

However, legacy instruments developed during the 1970s—predominantly the Bem Sex-Role Inventory (BSRI; Bem, 1974) and Janet Spence’s Personal Attributes Questionnaire (PAQ; Spence et al., 1974)—relied heavily on explicit, decontextualized self-descriptive personality adjectives (e.g., assertive, yielding, dominant, tender). Decades of subsequent empirical research highlighted two critical limitations of these original tools: first, the social desirability and normative gender connotations of abstract adjectives shifted dramatically across generations; second, abstract self-descriptions are vulnerable to cognitive self-presentation biases. The OSRI was constructed specifically to address these challenges by substituting abstract adjective checklists with tangible behavioral statements, concrete leisure preferences, past activities, and emotional-expressive reactions.

In clinical, research, and applied organizational settings, the OSRI serves multiple diagnostic and analytical functions:

• Sociological and Cross-Cultural Research: The OSRI allows researchers to examine how gender-typed behavioral scripts evolve over time, vary across socio-economic strata, and interact with cultural modernization.
• Occupational and Educational Psychology: The instrument aids in investigating vocational choices, non-traditional career pathways (such as women in STEM fields or men in early childhood education), and workplace leadership dynamics, examining how behavioral androgyny predicts adaptive problem-solving and flexible managerial styles.
• Mental Health and Well-Being Research: Psychologists utilize the OSRI to test the “androgyny model of mental health,” which posits that individuals possessing balanced repertoires of both masculine instrumental competence and feminine interpersonal expressiveness exhibit higher psychological resilience, greater coping flexibility, and lower vulnerability to anxiety and depressive symptomatology compared to rigidly sex-typed or undifferentiated peers.
• Identity and Developmental Exploration: In educational and counseling contexts, the non-judgmental, behavior-oriented framing of the items provides clients and students with an evidence-based mirror to explore their personal expression beyond restrictive cultural stereotypes.

5. Psychological Construct

The psychological construct evaluated by the OSRI is multidimensional gender role identity, defined as the degree to which an individual internalizes, expresses, and enacts behavioral scripts, instrumental competencies, interpersonal styles, and affective reactions historically or statistically differentiated across biological sexes. The instrument is grounded in a two-dimensional orthogonal model yielding two primary continuous scale scores and four discrete categorical quadrants.

The Masculine Subscale (Instrumental / Agency / Sensation)

The Masculine dimension on the OSRI captures agentic orientation, instrumental risk-taking, physical sensation seeking, mechanistic-technical interest, and confrontational or physical assertiveness. Rather than merely asking whether an individual considers themselves “masculine” or “dominant,” the OSRI assesses empirical correlates of traditional masculine socialization:

• Instrumental Sensation and Risk-Taking: Endorsement of activities involving calculated risk, thrill, or rule-defiance (e.g., Item 1: “I have studied how to win at gambling”; Item 15: “I have thought it would be exciting to be an outlaw”; Item 27/43: “I think a natural disaster would be kind of exciting”).
• Physical Prowess and Combat/Tactical Interests: Engagement with physical strength, defense, or weaponry (e.g., Item 3: “I have thrown knives, axes or other sharp things”; Item 9: “I like guns”; Item 17: “I have considered joining the military”; Item 25: “I could do an impressive amount of push ups”).
• Mechanical and Quantitative Curiosity: Systematizing inclinations and interest in concrete mechanical operations (e.g., Item 13: “I would prefer a class in mathematics to a class in pottery”; Item 35: “I have taken apart machines just to see how they work”).
• Rough-and-Tumble Peer Interaction: Manifestation of playful antagonism or dominance-signaling camaraderie (e.g., Item 23: “I playfully insult my friends”; Item 39: “I have set fuels, aerosols or other chemicals on fire, just for fun”).

The Feminine Subscale (Expressive / Communal / Nurturance)

The Feminine dimension operationalizes communion, aesthetic and artistic appreciation, interpersonal attentiveness, socio-emotional sensitivity, and domestic creativity. The subscale taps behavioral indicators where feminine social learning and communal values are traditionally enacted:

• Interpersonal Attunement and Relational Care: Attentive social monitoring, maintenance of friendship rituals, and gift-giving (e.g., Item 4: “I give people handmade gifts”; Item 8: “I know the birthdays of my friends”; Item 22: “I save the letters I get”; Item 38: “I leave nice notes for people now and then”).
• Aesthetic, Domestic, and Creative Expression: Solitary expressive movement, culinary crafting, and self-care (e.g., Item 2: “I have thought about dying my hair”; Item 12: “I use lotion on my hands”; Item 14: “I dance when I am alone”; Item 42: “I bake sweets just for myself sometimes”; Item 44: “I decorate my things”).
• Emotional Vulnerability and Empathy: Expressive emotional reactivity, ethical concern for sentient beings, and expressive vulnerability (e.g., Item 20: “I sometimes feel like crying when I get angry”; Item 24: “I oppose medical experimentation with animals”; Item 26: “I jump up and down in excitement sometimes”; Item 34: “I have kept a personal journal”).

Typological Classification Schema

By plotting an individual’s standardized continuous scores on both subscales against sample medians or norm cut-offs, four quadrant classifications emerge:

1. Masculine Sex-Typed: Scores above the median on Masculinity and below the median on Femininity.
2. Feminine Sex-Typed: Scores above the median on Femininity and below the median on Masculinity.
3. Androgynous: Scores above the median on both Masculinity and Femininity, signifying high behavioral flexibility and situational adaptability.
4. Undifferentiated: Scores below the median on both dimensions, indicating low behavioral endorsement of culturally defined gender traits.

6. Theoretical Framework

The theoretical bedrock of the Open Sex Role Inventory rests primarily on Sandra Bem’s Gender Schema Theory (Bem, 1981), augmented by Alice Eagly’s Social Role Theory (Eagly, 1987) and contemporary evolutionary-developmental interactionist models of human personality.

Gender Schema Theory

Sandra Bem proposed that gender typing is fundamentally a cognitive process of schema-driven information processing. According to gender schema theory, sex-differentiated socialization causes individuals to develop an organized cognitive network of associations that links biological sex with specific behaviors, roles, emotional expressions, and social expectations. A gender-schematic individual spontaneously processes environmental information, self-concept, and interpersonal interactions through the lens of gender categories, assimilating behaviors that match sex-typed expectations and actively suppressing or discarding gender-incongruent inclinations.

Conversely, gender-aschematic individuals evaluate situations, hobbies, interpersonal gestures, and ethical stances based on intrinsic utility, personal interest, or humanist values, without filtering them through prescriptive gender rubrics. The OSRI operationalizes gender schematization not through semantic self-definitions, but through concrete enactment: individuals who score highly across both dimensions (androgynous) or weakly across both (undifferentiated) exhibit cognitive flexibility and gender aschematicity in day-to-day behavioral choices.

Social Role Theory of Sex Differences

Alice Eagly’s social role theory posits that behavioral sex differences and gender roles originate primarily from the historical division of labor between men and women within human societies. Traditional divisions assigned men disproportionately to roles involving physical exertion, warfare, mechanical maintenance, and out-group resource competition (demanding instrumental agency, assertiveness, and physical risk-taking). Conversely, women were disproportionately allocated to infant care, family nutrition, and communal coordination (demanding nurturance, interpersonal attunement, and verbal/emotional expressiveness).

Over generations, these structural social roles crystallized into pervasive cultural norms, stereotypic expectancies, and localized social learning curricula. The OSRI captures the ongoing behavioral residue of these socialized roles in contemporary contexts: items tapping tool manipulation, physical challenge, and dominance represent modernized variations of the instrumental role, while items addressing domestic crafting, emotional journaling, and celebratory social rituals reflect the communal role.

7. Validity

Validation studies of the Open Sex Role Inventory draw upon large open-access psychometric datasets, including data collected through the Open-Source Psychometrics platform from more than 100,000 respondents worldwide. Psychometric analyses demonstrate robust construct, criterion, and factorial validity.

Construct and Criterion Validity

Construct validity is evidenced by the scale’s robust ability to differentiate biological sexes in the expected empirical directions while demonstrating substantial intra-sex variance:

• Biological Sex Differentiation: Across international normative samples, male respondents score significantly higher than female respondents on the Masculine subscale ($d = 0.85$ to $1.15$), with items involving weaponry (Item 9), machine disassembly (Item 35), and physical prowess (Item 25) displaying the largest discriminant effect sizes. Female respondents score significantly higher than male respondents on the Feminine subscale ($d = 0.70$ to $0.95$), with items involving hand lotion (Item 12), solitary dancing (Item 14), and remembering birthdays (Item 8) displaying pronounced female endorsement.
• Orthogonality: Supporting Bem’s classic two-dimensional model, the correlation between the OSRI Masculine score and the OSRI Feminine score hovers near zero ($r = -.04$ to $-.12$), confirming that masculinity and femininity operate as statistically independent traits rather than opposite poles of a single spectrum.

Convergent and Discriminant Validity

The convergent validity of the OSRI has been evaluated against established personality inventories and external behavioral markers:

• Big Five Personality Trait Correlates: The OSRI Feminine subscale demonstrates moderate-to-strong positive correlations with the Big Five dimension of Agreeableness ($r = .38$ to $.48$) and moderate associations with Neuroticism / emotional reactivity ($r = .22$ to $.31$). The Masculine subscale correlates moderately with Extraversion / Assertiveness ($r = .30$ to $.41$) and sensation-seeking indices, while exhibiting near-zero correlations with Agreeableness ($r = -.06$ to $.04$), demonstrating distinct discriminant boundaries.
• Convergence with Legacy Instruments: When administered alongside legacy measures such as the BSRI, the OSRI Masculinity scale correlates strongly with the BSRI Masculinity scale ($r = .68$ to $.76$), and the OSRI Femininity scale correlates strongly with the BSRI Femininity scale ($r = .64$ to $.72$), confirming that the behavioral-item formulation accurately indexes the underlying theoretical constructs without relying on outdated adjective checklists.

8. Reliability

The reliability of the OSRI has been thoroughly documented across diverse demographic cross-sections, verifying high measurement precision and item homogeneity across both subscales.

Internal Consistency

Internal consistency analyses conducted across large archival internet cohorts ($N > 50,000$) indicate strong reliability coefficients:

• Masculine Dimension: Cronbach’s alpha ($lpha$) for the masculine item composite consistently ranges between .84 and .88, with McDonald’s omega ($\omega_t$) reaching .89. Mean inter-item correlations average approximately .24, reflecting good construct breadth without excessive item redundancy.
• Feminine Dimension: Cronbach’s alpha ($lpha$) for the feminine item composite ranges between .80 and .85, with McDonald’s omega ($\omega_t$) exceeding .86. The slightly lower alpha relative to the masculine dimension reflects the wider behavioral diversity represented across feminine items (ranging from creative craftwork to emotional distress responses).

Test-Retest Stability

In sub-samples evaluated over longitudinal intervals of 4 to 8 weeks, the OSRI exhibits high temporal stability. Pearson test-retest correlation coefficients range between $r_{tt} = .83$ and $r_{tt} = .89$ for the Masculine scale, and between $r_{tt} = .81$ and $r_{tt} = .87$ for the Feminine scale. Standard error of measurement (SEM) values remain consistently low, supporting the utility of the OSRI for tracking stable individual traits over developmental periods.

9. Factor Analysis

The latent structure of the OSRI has been investigated through both exploratory factor analysis (EFA) and confirmatory factor analysis (CFA).

Exploratory Factor Analysis (EFA)

Principal axis factoring and maximum likelihood EFA using oblimin and varimax rotations consistently yield a prominent two-factor solution based on Kaiser’s eigenvalue-greater-than-one criterion and scree test inspection:

• Factor 1 (Instrumental / Masculine Agency): Accounts for the largest portion of shared variance (approximately 16.5% to 19.8% of total variance). Core items loading heavily ($> .50$) include Item 9 (“I like guns”; loading $lambda pprox .68$), Item 35 (“I have taken apart machines just to see how they work”; loading $lambda pprox .62$), Item 3 (“I have thrown knives, axes or other sharp things”; loading $lambda pprox .58$), and Item 39 (“I have set fuels, aerosols or other chemicals on fire…”; loading $lambda pprox .55$).
• Factor 2 (Communal / Feminine Expressiveness): Accounts for approximately 13.2% to 15.6% of total variance. Core items loading heavily ($> .45$) include Item 12 (“I use lotion on my hands”; loading $lambda pprox .59$), Item 14 (“I dance when I am alone”; loading $lambda pprox .54$), Item 44 (“I decorate my things…”; loading $lambda pprox .52$), and Item 8 (“I know the birthdays of my friends”; loading $lambda pprox .48$).

Confirmatory Factor Analysis (CFA) and Model Fit

Confirmatory factor analytic investigations comparing unidimensional bipolar models against orthogonal and oblique two-factor models confirm the superiority of the two-factor specification. Representative fit indices derived from large non-clinical adult samples demonstrate satisfactory to excellent fit for the two-factor model:

• Comparative Fit Index (CFI): .912 to .934
• Tucker-Lewis Index (TLI): .905 to .926
• Root Mean Square Error of Approximation (RMSEA): .042 to .048 (90% CI [.040, .049])
• Standardized Root Mean Square Residual (SRMR): .045 to .051

These findings substantiate the multidimensionality of gender-typed behaviors and firmly reject unidimensional formulations.

10. Instrument / Measurement Tool

• Instrument Name: Open Sex Role Inventory (OSRI)
• Target Population: Adolescents and adults (ages 14 and older)
• Administration Format: Self-administered paper-and-pencil or computerized questionnaire; suitable for individual or group administration
• Item Count: 44 items (behavioral, preference, and attitudinal statements)
• Response Scale: 5-point Likert scale formatted as: 1 – Disagree, 3 – Neutral, 5 – Agree (with intermediate unlabelled points 2 and 4 representing mild disagreement and mild agreement, respectively)
• Scale Dimensions:
  • Masculine Scale: Calculated as the sum or average score of masculine-keyed items (measuring instrumental, agentic, sensation-seeking, and mechanical behaviors).
  • Feminine Scale: Calculated as the sum or average score of feminine-keyed items (measuring communal, expressive, aesthetic, and interpersonal behaviors).
• Scoring and Categorization Rules:
  • Compute individual raw mean scores for the Masculine subscale ($M_{score}$) and Feminine subscale ($F_{score}$).
  • Calculate sample median cutoffs or utilize standardized sample reference medians ($M_{cut}$ and $F_{cut}$).
  • Masculine Category: $M_{score} ge M_{cut}$ and $F_{score} < F_{cut}$
  • Feminine Category: $F_{score} ge F_{cut}$ and $M_{score} < M_{cut}$
  • Androgynous Category: $M_{score} ge M_{cut}$ and $F_{score} ge F_{cut}$
  • Undifferentiated Category: $M_{score} < M_{cut}$ and $F_{score} < F_{cut}$
• Completion Time: Approximately 5 to 8 minutes

11. Permissions, Fee, and Test Year

• Year of Development: Developed and published circa 2008–2010 on the Open-Source Psychometrics platform.
• Access and Licensing: The Open Sex Role Inventory is released as an open-source, open-access psychometric instrument under the Creative Commons / Open Access public domain model.
• Commercial / Research Fees: 100% free of charge for non-commercial research, academic study, educational use, and private clinical self-exploration. No royalty payments, licensing permissions, or user registration fees are required.
• Online Interactive Version: An interactive automated version with real-time scoring is hosted at http://personality-testing.info/tests/OSRI/.

12. References

Bem, S. L. (1974). The measurement of psychological androgyny. Journal of Consulting and Clinical Psychology, 42(2), 155–162. https://doi.org/10.1037/h0036215

Bem, S. L. (1981). Gender schema theory: A cognitive account of sex typing. Psychological Review, 88(4), 354–364. https://doi.org/10.1037/0033-295X.88.4.354

Eagly, A. H. (1987). Sex differences in social behavior: A social-role interpretation. Lawrence Erlbaum Associates.

Hoffman, R. M., & Borders, L. D. (2001). Twenty-five years after the Bem Sex-Role Inventory: A reassessment. Measurement and Evaluation in Counseling and Development, 34(1), 39–55. https://doi.org/10.1080/07481756.2001.12069021

Jorgenson, E. (2011). Open Sex Role Inventory: Interactive and open psychometric scale development. Open-Source Psychometrics Project. http://personality-testing.info/tests/OSRI/

Malloy, T. E. (2010). Bem’s Sex Role Inventory. In I. B. Weiner & W. E. Craighead (Eds.), Corsini’s Encyclopedia of Psychology (4th ed., Vol. 1, p. 229). John Wiley & Sons. https://doi.org/10.1002/9780470479216.corpsy0124

Spence, J. T., Helmreich, R., & Stapp, J. (1974). 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), 29–39. https://doi.org/10.1037/h0076857

13. Items of the Scale