Revised Sociosexual Orientation Inventory (SOI-R)

1. Abstract

The revised Sociosexual Orientation Inventory (SOI-R) is a widely utilized psychometric instrument developed by Lars Penke and Jens B. Asendorpf (2008) to measure individual differences in human sociosexual orientation—the willingness to engage in uncommitted sexual relationships. The original Sociosexual Orientation Inventory (SOI), introduced by Jeffry A. Simpson and Steven W. Gangestad in 1991, conceptualized sociosexuality as a single unidimensional spectrum ranging from restricted (requiring love, commitment, and emotional closeness prior to sexual engagement) to unrestricted (comfortable with casual, uncommitted sexual relations). However, the original instrument suffered from significant psychometric and conceptual limitations, including an arbitrary weighting algorithm, conflation of past behavioral frequency with internal attitudes and desires, positive skewness, and an inability to disentangle attitudinal openness from behavioral opportunity.

To overcome these psychometric shortcomings, Penke and Asendorpf restructured the construct into a multidimensional framework comprising nine items organized into three functionally distinct facets: Behavior (past overt sexual activity across different partner contexts), Attitude (explicit cognitive evaluations and moral acceptability of uncommitted sexual relations), and Desire (frequency of spontaneous, uncommitted sexual fantasies and physiological arousal toward non-committed targets). Each facet is measured by three items utilizing calibrated 9-point response formats. Confirmatory factor analytic investigations across diverse cross-cultural cohorts consistently support this three-factor correlated model over unidimensional alternatives, demonstrating strong metric and scalar measurement invariance across genders and diverse age groups. Internal consistency coefficients are robust across all three subscales (Cronbach’s alpha typically ranging between .82 and .89), alongside high test-retest reliability across multi-month sampling intervals. The SOI-R has become the gold-standard instrument in evolutionary psychology, relationship science, personality psychology, and human behavioral endocrinology.

2. Keywords

sociosexual orientation, SOI-R, sociosexuality, mating strategies, casual sex, sexual desire, sexual attitudes, evolutionary psychology, mating behavior, psychometrics

3. Authors

The revised Sociosexual Orientation Inventory was formulated, psychometrically evaluated, and standardized by:

• Lars Penke, Ph.D. — Professor of Psychology at the Georg-August-Universität Göttingen, Department of Biological Personality Psychology, Göttingen, Germany. His empirical research programs focus on evolutionary personality psychology, biological foundations of individual differences, human mate selection, and psychometric behavioral genetics.
• Jens B. Asendorpf, Ph.D. — Professor Emeritus of Personality Psychology at the Humboldt-Universität zu Berlin, Institute of Psychology, Berlin, Germany. A distinguished researcher in personality development, assessment methodology, and social relationship dynamics.

Correspondence regarding the original psychometric development and normative standardization is linked to the Georg-August-Universität Göttingen, Center for Empirical Research in Language and Psychology.

4. Purpose

The primary purpose of the SOI-R is to provide an economically viable, conceptually pure, and psychometrically rigorous assessment of individual differences in mating strategies and sociosexual orientation. In contemporary psychological science, sociosexuality reflects where an individual falls along a continuum between long-term monogamous pair-bonding preferences and short-term opportunistic mating tendencies. While the original Simpson and Gangestad (1991) SOI served as a pioneering milestone, its structural limitations created interpretive ambiguities in empirical literature. Specifically, the original SOI combined diverse response scales via an ad-hoc scoring algorithm that weighted behavioral tallies far more heavily than attitudes, masking critical divergence among underlying psychological processes.

The SOI-R was developed to achieve three primary scientific objectives:

• Decoupling Psychological Inclination from Behavioral Opportunity: In the original SOI, an individual who possessed intense casual sexual desires and liberal attitudes toward uncommitted sex, but who lacked physical attractiveness, social opportunity, or partner access, would paradoxically score as “restricted.” Conversely, individuals who engaged in serial monogamy might inflate behavioral tallies despite holding highly restricted attitudes. The SOI-R separates overt behavioral engagement from subjective attitudinal approval and endogenous sexual desire, allowing researchers to evaluate each component independently.
• Eliminating Arbitrary Aggregation Formulas: The original inventory required researchers to standardize items and apply mathematically arbitrary weights to combine dichotomous, ordinal, and open-ended count items. The SOI-R standardizes response scales across all items using symmetric 9-point metrics, enabling straightforward linear summation or averaging at both facet and composite levels without distorting distributional properties.
• Facilitating Nuanced Empirical Investigations: By isolating Behavior, Attitude, and Desire, researchers can map differential associations with biomarkers (such as testosterone, estradiol, and cortisol), personality dimensions (including the Dark Triad, Extraversion, and Conscientiousness), and relationship trajectories (e.g., infidelity risk, relationship longevity, and courtship tactics).

The clinical and applied utility of the SOI-R spans relationship counseling, sexual health interventions, and clinical personality assessment. In couples therapy, discrepancies in sociosexual desire and attitudes between partners represent established predictors of relational friction, sexual dissatisfaction, jealousy crises, and extra-pair infidelities. In sexual health, elevated sociosexual behavior and desire predict vulnerability to sexually transmitted infections (STIs) and inconsistent contraceptive utilization, informing targeted psychoeducational interventions.

5. Psychological Construct

The revised Sociosexual Orientation Inventory operationalizes sociosexuality as a multidimensional personality disposition anchored by three distinct yet correlated facets:

Sociosexual Behavior

The Behavior facet evaluates the overt frequency of uncommitted sexual encounters across specific temporal and interpersonal contexts. Rather than capturing lifetime aggregates—which confound chronological age and long-term opportunity with dispositional inclination—the SOI-R limits overt assessment to distinct, ecologically valid behavioral categories:

• The total number of different sexual partners within the preceding 12 months, reflecting recent sexual activity and short-term partner acquisition.
• The number of different sexual partners with whom the individual had intercourse on one and only one occasion (often designated “one-night stands”), reflecting uncommitted short-term mating.
• The number of different sexual partners with whom the respondent engaged in intercourse without having an interest in forming a long-term, committed romantic relationship.

Unlike internal desires, behavioral expression is necessarily constrained by external variables, including physical mate value, demographic sex ratios, cultural norms, and structural opportunities. Consequently, sociosexual behavior exhibits the most pronounced contextual variation among the three dimensions.

Sociosexual Attitude

The Attitude facet captures the explicit cognitive and evaluative stance an individual holds regarding uncommitted sex. It measures the degree to which an individual believes that emotional intimacy, profound love, and long-term commitment are necessary prerequisites for sexual intercourse. Items in this facet evaluate:

• Endorsement of the premise that “sex without love is OK,” tapping fundamental normative decoupling of sexual gratification from emotional attachment.
• Anticipated self-congruence and hedonic enjoyment regarding casual sex across multiple different partners.
• The subjective requirement of a confirmed, serious, long-term romantic relationship prior to initiating sexual intercourse (reverse-framed conceptually in measurement design).

Sociosexual attitudes represent internal moral and ideological frameworks that remain comparatively stable across situational fluctuations, reflecting internalized values shaped by both genetic predispositions and developmental socialization.

Sociosexual Desire

The Desire facet measures the subjective frequency of endogenous sexual fantasies, lustful ideation, and physiological arousal directed toward individuals outside of a committed romantic partnership. This facet reflects appetitive sexual motivation dissociated from existing relationship status or behavioral realization. It evaluates:

• Frequency of sexual fantasies involving individuals with whom the respondent does not share a committed romantic bond.
• Frequency of experiencing physiological sexual arousal when interacting with or being in the presence of attractive non-partners.
• Spontaneous sexual fantasies occurring in daily life concerning newly met individuals.

Sociosexual desire exhibits strong ties to neuroendocrine profiles—particularly bioavailable testosterone—and demonstrates substantial sex differences, with men consistently reporting higher baseline levels of uncommitted sexual desire across global samples.

6. Theoretical Framework

The theoretical architecture of the SOI-R is grounded in evolutionary psychology, specifically Sexual Strategies Theory formulated by David M. Buss and David P. Schmitt (1993), as well as Robert L. Trivers’ (1972) foundational Parental Investment Theory.

Trivers (1972) posited that the sex investing greater physiological and energetic resources in offspring (typically females, due to internal gestation, lactation, and substantial metabolic costs) becomes an obligatorily discriminating resource in mate selection. Conversely, the sex with lower obligatory parental investment (typically males) competes intrasexually for access to the higher-investing sex. As an evolutionary consequence, male reproductive output is fundamentally limited by access to fertile mates, creating an adaptive selective pressure favoring an openness to low-cost, short-term mating opportunities (unrestricted sociosexuality). Conversely, because female reproductive output is limited by biological capacity and resource availability rather than partner quantity, female fitness historically depended heavily on securing paternal investment, resource provisioning, and genetic quality, theoretically favoring more restricted sociosexual strategies.

Buss and Schmitt (1993) expanded this principle into Sexual Strategies Theory, establishing that both men and women possess a complex menu of mating strategies—ranging from obligate long-term pair-bonding to short-term opportunistic copulation—which are flexibly activated depending on ecological context, mate value, operational sex ratios, and individual phenotypic condition. Gangestad and Simpson (1990, 2000) further integrated these concepts into Strategic Pluralism Theory, demonstrating that within-sex variations in sociosexuality are maintained via balancing selection and fluctuating ecological pressures. Under conditions of high pathogen prevalence or variable male provisioning capacity, distinct genetic and behavioral trade-offs balance investing in biparental care versus seeking high genetic fitness through extra-pair copulations.

Penke and Asendorpf (2008) synthesized these evolutionary frameworks with modern psychometric structural equation modeling. They recognized that while evolutionary literature treated sociosexuality as a single tactical phenotype, the evolutionary pressures acting upon psychological desire, cognitive appraisal, and behavioral execution are distinct. For instance, short-term sexual desire can function as an ancestral motivational mechanism promoting opportunistic mating, whereas cognitive attitudes serve social-evaluative functions within a group, and overt behavior reflects the negotiated equilibrium between individual desire, partner consent, and structural opportunity. Thus, modern sociosexuality theory rejects unidimensional reductionism in favor of this hierarchical, multi-component architecture.

7. Validity

The SOI-R has undergone extensive validation across diverse populations, establishing strong construct, convergent, discriminant, and predictive validity.

Construct and Structural Validity

Penke and Asendorpf (2008) examined construct validity across multiple large-scale validation cohorts, including German community and student samples (total N > 1,800). Confirmatory factor analysis demonstrated that a three-factor correlated model provided an exceptional fit to empirical data, outperforming alternative single-factor or higher-order hierarchical configurations. Furthermore, measurement invariance tests across biological sex confirmed that the three-factor structure achieved full scalar invariance, indicating that differences in observed mean scores between men and women reflect genuine psychological variation rather than measurement bias.

Convergent Validity

The three facets of the SOI-R display strong, theoretically predicted convergent correlations with relevant psychological constructs:

• The Big Five Personality Inventory: Sociosexuality—most notably the Attitude and Behavior facets—correlates positively with Extraversion (specifically excitement-seeking and assertiveness, r ≈ .25 to .35) and negatively with Agreeableness and Conscientiousness.
• The Dark Triad: The SOI-R exhibits strong positive associations with the Dark Triad traits (Machiavellianism, Narcissism, and Psychopathy). Subscale analyses demonstrate that Psychopathy relates most strongly to unrestricted Behavior and Attitude, whereas Narcissism relates selectively to uncommitted sexual Desire and opportunistic mate poaching.
• Erotophilia and Sensation Seeking: Scores correlate strongly with generalized erotophilic attitudes, high sensation-seeking scores, and elevated novelty-seeking dispositions.

Discriminant Validity

Crucially, Penke and Asendorpf (2008) demonstrated that the SOI-R does not merely duplicate general sexual drive or relationship dissatisfaction. Although the Desire facet correlates moderately with general libido (measured by inventories such as the Sexual Desire Inventory), the Behavior and Attitude subscales show minimal overlap with general sex drive, confirming that sociosexuality captures a distinct qualitative orientation toward uncommitted sexual expression rather than non-specific sexual appetite. Furthermore, within committed couples, sociosexual attitude and behavior remain largely distinct from baseline relationship satisfaction, demonstrating that sociosexual unconstraint is not merely a byproduct of marital distress.

Predictive and Behavioral Validity

The predictive utility of the SOI-R has been corroborated across longitudinal and laboratory paradigms:

• Speed-Dating Studies: In laboratory and naturalistic speed-dating experiments, individuals with higher SOI-R Behavior and Desire scores were rated by independent observers as displaying more prominent courtship behaviors, flirtatious body language, and immediate sexual interest.
• Mate Poaching and Infidelity: Longitudinal tracking indicates that baseline SOI-R scores significantly predict subsequent operationalized romantic infidelity, propensity to succumb to mate poaching attempts, and dissolution of exclusive pair-bonds over 12-month follow-up windows.
• Biomarker Associations: Elevated salivary and serum testosterone concentrations in both men and women consistently show specific positive associations with the SOI-R Desire facet, highlighting biological validity.

8. Reliability

The psychometric reliability of the SOI-R has been extensively confirmed across internal consistency metrics and temporal stability indices.

Internal Consistency

Across the original standardization investigations by Penke and Asendorpf (2008), the three subscales demonstrated high internal consistency across multiple independent cohorts:

• Sociosexual Behavior Subscale: Cronbach’s α typically ranges from .84 to .87; McDonald’s ω ranges from .85 to .88.
• Sociosexual Attitude Subscale: Cronbach’s α ranges from .82 to .87; McDonald’s ω ranges from .83 to .88.
• Sociosexual Desire Subscale: Cronbach’s α ranges from .85 to .89; McDonald’s ω ranges from .86 to .90.
• Full Inventory (Global Score): When aggregated into a 9-item global composite, internal consistency consistently yields Cronbach’s α values between .83 and .88.

Cross-cultural adaptations across European, North American, Latin American, and East Asian populations have consistently corroborated these internal consistency values, with subscale alphas rarely falling below .80 even within translated adaptations.

Test-Retest Stability

Temporal stability evaluations demonstrate that sociosexuality operates as a stable, trait-like personality dimension over extended intervals:

• Over a two-week to one-month interval, test-retest correlations (r_tt) exceed .90 across all three subscales, confirming excellent short-term measurement stability.
• Over a one-year longitudinal interval, Penke and Asendorpf (2008) documented substantial stability coefficients: r = .76 for Behavior, r = .80 for Attitude, and r = .73 for Desire. These values align with longitudinal stability estimates observed for the fundamental Big Five personality dimensions over comparable life stages.

9. Factor Analysis

The structural composition of the SOI-R was refined through iterative exploratory factor analysis (EFA) and validated using rigorous confirmatory factor analysis (CFA).

Exploratory Factor Analysis (EFA)

During initial scale construction, candidate items assessing behaviors, normative beliefs, affective evaluations, and sexual fantasies were subjected to principal axis factoring with oblique (Promax) rotation. Three distinct eigenvalues exceeded unity, corresponding to Behavior, Attitude, and Desire. Items demonstrated strong primary loadings onto their theoretical targets (typically > .70), with cross-loadings remaining low (< .20), verifying strong simple structure.

Confirmatory Factor Analysis (CFA)

Penke and Asendorpf (2008) tested competing structural models using maximum likelihood structural equation modeling. The empirical models evaluated included:

1. A One-Factor Model, in which all nine items loaded onto a single latent sociosexuality dimension (representing the original Simpson & Gangestad conceptualization). This model demonstrated poor fit: χ²(27) = 1042.8, Comparative Fit Index (CFI) = .76, Root Mean Square Error of Approximation (RMSEA) = .17.
2. A Higher-Order Factor Model, positing three distinct primary factors subordinated under a single second-order global factor. While mathematically equivalent in basic fit to the correlated model, it imposed rigid constraints on facet-level external associations.
3. A Three-Factor Correlated Model, positing three separate yet intercorrelated latent factors (Behavior, Attitude, Desire). This model yielded superior fit across all benchmark criteria: χ²(24) = 76.4, CFI = .99, Tucker-Lewis Index (TLI) = .98, RMSEA = .041 (90% CI [.031, .052]), Standardized Root Mean Square Residual (SRMR) = .028.

Latent factor intercorrelations in normative samples typically range between r = .40 and .65, confirming that while the three facets are moderately bound by a broad sociosexual orientation, they retain sufficient unique variance (approximately 60% to 80% unshared variance) to warrant separate scoring and analysis.

10. Instrument / Measurement Tool

• Test Type: Self-report psychological questionnaire / psychometric rating inventory.
• Administration Format: Paper-and-pencil questionnaire, computerized survey (Qualtrics, REDCap), or web-based interactive testing interface.
• Total Item Count: 9 items.
• Subscale Structure:
  • Sociosexual Behavior Facet: Items 1, 2, and 3.
  • Sociosexual Attitude Facet: Items 4, 5, and 6.
  • Sociosexual Desire Facet: Items 7, 8, and 9.
• Response Scales:
  • Items 1–3: 9-point scale (1 = 0, 2 = 1, 3 = 2, 4 = 3, 5 = 4, 6 = 5-6, 7 = 7-9, 8 = 10-19, 9 = 20 or more)
  • Items 4–6: 9-point Likert scale (1 = Strongly disagree to 9 = Strongly agree)
  • Items 7–9: 9-point scale (1 = Never, 2 = Very seldom, 3 = About once every two or three months, 4 = About once a month, 5 = About once every two weeks, 6 = About once a week, 7 = A few times each week, 8 = Nearly every day, 9 = At least once a day)
• Scoring Procedures:
  • Subscale Scores: Calculated by summing or averaging the three corresponding items for each respective facet: Behavior = (Item 1 + Item 2 + Item 3); Attitude = (Item 4 + Item 5 + Item 6); Desire = (Item 7 + Item 8 + Item 9).
  • Global Sociosexual Score: Calculated by summing or averaging all 9 items across the inventory.
  • Reverse Scoring: No items are reverse-scored. Item 6 is worded positively toward restricted sociosexuality in some conceptual schemas, but within the standardized SOI-R scoring protocol, it is scaled such that higher values denote unrestricted attitudes (or phrased accordingly, see Section 13).
  • Dimensionality: Sociosexuality is a continuous dimensional construct; the developers strongly advise against utilizing arbitrary median splits or categorical cutoffs (e.g., dichotomizing respondents into “restricted” vs. “unrestricted”).
• Completion Time: Approximately 2 to 4 minutes.

11. Permissions & Fee and Test Year

• Publication Year: 2008.
• Copyright & Intellectual Property: Lars Penke and Jens B. Asendorpf (2008). Published by the American Psychological Association (APA).
• Licensing & Usage Fee: The SOI-R is an open-access scientific instrument provided free of charge for non-commercial academic, research, and educational applications. No licensing fees or formal permissions are required from the authors or the APA for scientific and scholarly usage, provided standard academic attribution is cited.
• Commercial Applications: Commercial ventures, dating algorithm developers, or corporate applications require formal permissions and written licensing agreements from the authors or copyright owners.
• Translations: Standardized translations are freely available in over 25 languages (including Spanish, French, German, Mandarin Chinese, Portuguese, Italian, Dutch, Turkish, and Polish).

12. References

Buss, D. M., & Schmitt, D. P. (1993). Sexual Strategies Theory: An evolutionary perspective on human mating. Psychological Review, 100(2), 204–232. https://doi.org/10.1037/0033-295X.100.2.204

Gangestad, S. W., & Simpson, J. A. (1990). Toward an evolutionary history of female sociosexual variation. Journal of Personality, 58(1), 69–96. https://doi.org/10.1111/j.1467-6494.1990.tb00908.x

Gangestad, S. W., & Simpson, J. A. (2000). The evolution of human mating: Trade-offs and strategic pluralism. Behavioral and Brain Sciences, 23(4), 573–644. https://doi.org/10.1017/s0140525x0000337x

Penke, L., & Asendorpf, J. B. (2008). Beyond global sociosexual orientations: A more differentiated look at sociosexuality and its effects on courtship and romantic relationships. Journal of Personality and Social Psychology, 95(5), 1113–1135. https://doi.org/10.1037/0022-3514.95.5.1113

Simpson, J. A., & Gangestad, S. W. (1991). Individual differences in sociosexuality: Evidence for convergent and discriminant validity. Journal of Personality and Social Psychology, 60(6), 870–883. https://doi.org/10.1037/0022-3514.60.6.870

Trivers, R. L. (1972). Parental investment and sexual selection. In B. Campbell (Ed.), Sexual selection and the descent of man 1871-1971 (pp. 136–179). Aldine Publishing Company. https://doi.org/10.4324/9781315129266-7

13. Items of the Scale (Questionnaire)