Revised Sociosexual Orientation Inventory

1. Abstract

The Revised Sociosexual Orientation Inventory (SOI-R) is an established nine-item psychometric instrument designed to assess individual differences in sociosexual orientation—the degree to which an individual is willing to engage in uncommitted, casual sexual relationships without emotional attachment or long-term investment. Developed by Lars Penke and Jens B. Asendorpf (2008) as a substantial psychometric revision of the original unidimensional Sociosexual Orientation Inventory (SOI; Simpson & Gangestad, 1991), the SOI-R addresses critical structural and methodological shortcomings of its predecessor, including low internal consistency, severely skewed item distributions, problematic open-ended response formats, and inappropriate item formulations for single individuals.

The instrument operationalizes sociosexuality not as a single global construct, but as a hierarchical, tripartite structure consisting of three distinct yet correlated facets: Sociosexual Behavior (lifetime and recent casual sexual partners), Sociosexual Attitude (evaluative endorsement of uncommitted sex), and Sociosexual Desire (spontaneous sexual arousal and fantasies directed toward individuals outside a committed romantic relationship). Each facet is measured using three items, administered via either a 5-point or a 9-point Likert-type response format. Across large-scale international samples comprising over 8,500 respondents, the SOI-R demonstrates robust psychometric properties: high internal consistency (Cronbach's $\alpha = .78$ to $.87$ across facets and sexes), strong one-year test-retest reliability ($r = .68$ to $.86$, with theoretical variations in female desire linked to relationship transitions), robust configural and metric measurement invariance across gender, and exceptional criterion, convergent, and discriminant validity in evolutionary, personality, and social psychological research.

2. Keywords

Sociosexual Orientation, SOI-R, Mating Strategies, Evolutionary Psychology, Sexual Desire, Sexual Attitudes, Casual Sex, Measurement Invariance, Psychometrics, Pair-Bonding

3. Authors

The Revised Sociosexual Orientation Inventory was formulated and validated by:

• Lars Penke, Ph.D. — Professor of Biological Personality Psychology and Psychological Assessment, Georg Elias Müller Institute of Psychology, Georg-August-Universität Göttingen, Germany; formerly at the Centre for Cognitive Ageing and Cognitive Epidemiology, Department of Psychology, University of Edinburgh, United Kingdom. (Correspondence: [email protected]).
• Jens B. Asendorpf, Ph.D. — Professor Emeritus of Personality Psychology, Institute of Psychology, Humboldt University of Berlin, Berlin, Germany.

4. Purpose

The primary purpose of the Revised Sociosexual Orientation Inventory (SOI-R) is to provide a psychometrically sound, theoretically articulated, and structurally nuanced measure of individual variability in human mating strategies. In the landscape of evolutionary psychology, human reproductive behavior is characterized by strategic flexibility, wherein individuals navigate trade-offs between short-term mating (opportunistic copulation, low parental investment, high partner turnover) and long-term mating (high commitment, substantial pair-bonding, extensive bi-parental investment). Sociosexual orientation captures this continuum, contrasting "restricted" sociosexuality (requiring deep emotional closeness, commitment, and mutual exclusivity before sex) with "unrestricted" sociosexuality (comfort with and pursuit of sex without love, commitment, or long-term prospects).

While the original 7-item SOI developed by Jeffry A. Simpson and Steven W. Gangestad (1991) catalyzed nearly two decades of research across social, personality, and anthropological domains, it possessed profound psychometric liabilities:

• Open-Ended Frequency Items: The original instrument employed open-ended numerical questions regarding past partners and expected future partners, creating extreme positive skew, sensitivity to outlier distortion, and susceptibility to social desirability or boastful exaggerations.
• Multifactorial Latent Space Mislabeled as Unidimensional: Factor-analytic reappraisals repeatedly revealed that the original SOI was not unidimensional, causing composite scores to conflate divergent psychological and behavioral processes.
• Item Content Flaws: Specific items, such as Item 4 assessing sex in one's "current relationship," systematically disqualified or biased responses from unpartnered/single individuals.
• Arbitrary Weighting Algorithms: Scoring the original scale required complex standardizations and variable weightings that produced inconsistent empirical outcomes across studies.

The SOI-R was engineered to resolve these psychometric limitations by standardizing response formats on rating scales, eliminating open-ended estimates, removing partner-status prerequisites, and explicitly delineating three psychological facets: Behavior, Attitude, and Desire. In empirical research, the tool serves to disentangle biological, sociological, and psychological drivers of mate selection, flirtation, infidelities, jealousy, and sexual health behaviors. In clinical and counseling settings, the SOI-R offers objective insights into relational discord, discrepancies in sexual desire between partners, risk factors for sexually transmitted infections (STIs), and personality alignments within romantic dyads.

5. Psychological Construct

The construct of sociosexuality, originally coined by Alfred Kinsey and colleagues (Kinsey, Pomeroy, & Martin, 1948) to describe individual distributions of sexual permissiveness and promiscuity, was formally integrated into contemporary personality science as an individual difference construct representing an individual's threshold for sexual engagement in the absence of investment or romantic commitment. The SOI-R conceptualizes sociosexual orientation as a three-dimensional construct:

1. Sociosexual Behavior

The Behavior facet assesses an individual's actual overt history of uncommitted sexual encounters. It quantifies recent sexual partner turnover (within the past 12 months), lifetime incidence of isolated single-encounter sexual interactions ("one-night stands"), and the number of partners with whom an individual engaged in sexual intercourse without desiring a long-term commitment. Crucially, the Behavior facet reflects realized mating opportunities and actions, which depend not only on psychological inclination but also on local sex ratios, physical attractiveness, self-perceived mate value, and sociological opportunity structures.

2. Sociosexual Attitude

The Attitude facet measures an individual's evaluative beliefs, moral dispositions, and cognitive comfort regarding uncommitted sexual intercourse. It taps into explicit ideological acceptance of sexual contact without emotional love, comfort with casual sexual relations across varying partners, and the subjective requirement (or lack thereof) of long-term relationship commitment prior to consenting to sexual intercourse. Unlike behavior, which is bounded by opportunity and external barriers, attitudes capture conscious normative endorsement and psychological permissiveness.

3. Sociosexual Desire

The Desire facet taps into the subjective, motivational, and appetitive underpinnings of sociosexuality. It evaluates the spontaneous frequency of sexual daydreams, fantasies, and visceral physiological arousal elicited by people encountered in everyday life with whom no romantic attachment or committed bond exists. By separating covert appetitive motivation from overt behavioral execution, the Desire facet isolates intrinsic sexual drive directed toward novel or uncommitted targets—a dimension largely neglected by the original SOI.

Crucially, Penke and Asendorpf (2008) demonstrated that these three dimensions exhibit distinct nomological networks. For instance, while men and women display virtually identical levels of actual casual sexual behavior when opportunities are constrained or equalized, men report substantially higher levels of sociosexual desire ($d = 0.86$), demonstrating that the components operate under divergent evolutionary, hormonal, and psychological selective pressures.

6. Theoretical Framework

The architecture of the SOI-R is grounded in contemporary human mating theory, synthesizing Robert Trivers' (1972) Parental Investment Theory, David Buss and David Schmitt's (1993) Sexual Strategies Theory (SST), and Steven Gangestad and Jeffry Simpson's (2000) Strategic Pluralism Model (SPM).

Trivers' Parental Investment Theory posits that the sex that obligatorily invests more biological resources into offspring (typically human females, due to internal gestation and lactation) will act as the more discriminating, selective mate, whereas the sex investing less (typically males) will compete intrasexually for mating access. Consequently, ancestral men could substantially increase their lifetime reproductive success through opportunistic, low-investment copulations with multiple partners. For ancestral women, however, indiscriminate uncommitted sex carried asymmetric risks, including energetic costs of pregnancy, lactation, paternal abandonment, and sexually transmitted diseases, offset only by distinct adaptive benefits such as securing superior genetic quality ("good genes"), immediate resource acquisition, or mate-switching opportunities.

Building on these foundational mechanics, the Strategic Pluralism Model suggests that neither men nor women evolved a uniform, invariant mating strategy. Instead, human evolution favored facultative plasticity: both sexes possess menu-driven short-term and long-term mating adaptations. An individual's calibrated balance between restricted (monogamous, pair-bonding) and unrestricted (promiscuous, multi-partner) strategies represents an adaptive reaction norm influenced by genetic predispositions, fluctuating physical environments, pathogen prevalence, local operational sex ratios, and personal phenotypic value (e.g., facial symmetry, muscularity, physical attractiveness, social status).

By decomposing sociosexuality into Behavior, Attitude, and Desire, the SOI-R provides the operational granularity necessary to test these evolutionary models. For example, parental investment theory predicts that psychological desire for sexual novelty should show deep sex divergence even when cultural norms, access to contraception, or societal penalties equalize overt behaviors between genders. The theoretical independence of these components is vital: an individual may hold favorable sociosexual attitudes due to progressive ideological values while experiencing low intrinsic sociosexual desire, or conversely, an individual might possess intense unrestricted desire while strictly restraining their behavior due to religious convictions, fear of disease, or lack of romantic opportunities.

7. Validity

The SOI-R has undergone extensive empirical validation across multiple psychometric investigations, establishing robust construct, convergent, discriminant, and predictive validity.

Construct & Convergent Validity

In the primary psychometric validation study conducted by Penke and Asendorpf (2008) across two large community and student samples ($N_1 = 8,522$; $N_2 = 2,708$), the global SOI-R score demonstrated near-perfect convergent alignment with the original 7-item SOI ($r = .90$ to $.95$), confirming that the revised scale preserves the broader construct captured by Simpson and Gangestad (1991) while removing structural noise. Convergent validity is evidenced by significant positive correlations with:

• Sensation Seeking and Impulsivity: Moderately correlated with behavioral and attitudinal openness ($r = .25$ to $.38$).
• Extraversion: Positively associated with sociosexual openness, driven particularly by assertiveness and social potency.
• The Dark Triad: Positive associations with Machiavellianism, psychopathy, and narcissism, which frequently predict exploitative or opportunistic mating orientations.
• Sex Drive / Libido: Strong convergent alignment, specifically localized to the Desire facet ($r > .60$).

Discriminant Validity

Discriminant validity is supported by negligible or negative correlations with unrelated personality constructs. Sociosexuality scores exhibit near-zero correlations with General Mental Ability / Cognitive IQ ($r = -.02$ to $.04$) and Emotional Stability / Neuroticism ($r = -.05$ to $.03$). Furthermore, Agreeableness and Conscientiousness exhibit modest negative associations with uncommitted sociosexuality ($r = -.15$ to $-.25$), illustrating that sociosexuality is not a proxy for general antisociality, but reflects a specialized psychological domain.

Predictive and Behavioral Validity

The distinctive predictive utility of the tripartite model was demonstrated through behavioral laboratory experiments, speed-dating paradigms, and longitudinal designs:

• Flirtatious Behavior: In video-coded interactions with attractive opposite-sex confederates, men's sociosexual Desire predicted direct flirting and assertive displays, whereas women's actual flirting was predicted jointly by their Desire and their perceived physical attractiveness.
• Longitudinal Partner Accrual: In prospective 12-month follow-up evaluations, only the Behavior and Desire facets—not the explicit Attitude facet—statistically predicted the actual number of new sexual partners accrued and the probability of romantic relationship dissolution.
• Infidelity: Unrestricted sociosexual desire significantly predicted extradyadic romantic and sexual betrayals within established couples, demonstrating empirical power over generic relationship satisfaction metrics.

8. Reliability

The SOI-R demonstrates consistently high reliability across diverse demographic strata, sexual orientations, and cultural populations.

Internal Consistency

In large-scale empirical studies (e.g., Penke & Asendorpf, 2008; Penke, 2006), Cronbach's alpha ($lpha$) coefficients for the full scale and its individual subscales have demonstrated high psychometric adequacy across both 5-point and 9-point response formats:

Test-Retest Stability

One-year longitudinal test-retest correlations ($r$) confirm substantial temporal stability for Behavior ($r = .83$ to $.86$) and Attitude ($r = .73$ to $.79$). Retest stability for the Desire facet is moderate for men ($r = .68$) but noticeably lower for women ($r = .39$). Substantive analyses revealed that this lower female retest stability is not psychometric error, but rather an authentic developmental reaction norm: women systematically suppress extra-pair sociosexual desires upon forming a committed romantic relationship and elevate these desires when transitioning out of a relationship, highlighting female desire as dynamically contingent upon pair-bond status.

9. Factor Analysis

The factorial structure of the SOI-R was established through rigorous exploratory and confirmatory factor analytic procedures.

Exploratory Factor Analysis (EFA)

During scale development, Penke (2006) extracted items from an initial candidate pool of 47 items spanning multiple domains of sexuality, motivation, and relationship behavior. Principal axis factoring with oblique rotation unambiguously supported a three-factor solution based on scree tests, eigenvalue criteria ($> 1.0$), and parallel analysis. The three best-performing items per factor were selected based on high primary loadings ($> .65$) and negligible cross-loadings ($< .20$).

Confirmatory Factor Analysis (CFA)

Penke and Asendorpf (2008) tested competing latent structural models using maximum likelihood structural equation modeling across thousands of participants. A three-factor first-order model representing Behavior, Attitude, and Desire demonstrated superior goodness-of-fit compared to a unidimensional model:

• Unidimensional Model Fit: Inadequate fit indices (e.g., $\chi^2/df > 18.0$, $\text{CFI} < .82$, $\text{RMSEA} > .11$).
• Three-Factor Correlated Model Fit: Outstanding empirical fit ($\text{CFI} > .97$, $\text{TLI} > .96$, $\text{RMSEA} < .05$, $\text{SRMR} < .03$). All standardized factor loadings were robust, ranging between $.68$ and $.89$.

Latent inter-factor correlations between the three subscales were low to moderate ($r = .17$ to $.55$), supporting structural distinctiveness. The correlation between the Attitude and Behavior facets was significantly stronger among women than among men, indicating that women's sexual actions are more strictly bounded by their subjective attitudes. Multigroup CFA confirmed configural and metric invariance across sexes, ensuring that differences in latent factor means reflect genuine psychological distinctions rather than differential item functioning.

10. Instrument / Measurement Tool

• Instrument Type: Self-report psychometric rating inventory.
• Administration Format: Suitable for paper-and-pencil questionnaires, digital web-based surveys, and computerized laboratory tasks.
• Item Count: 9 items organized into three 3-item subscales:
  • Behavior: Items 1, 2, and 3.
  • Attitude: Items 4, 5, and 6.
  • Desire: Items 7, 8, and 9.
• Response Formats: Two alternative validated response formats are available:
  • 5-Point Scale: Highly recommended for general, clinical, and non-student adult populations due to ease of category discrimination.
  • 9-Point Scale: Designed for backward compatibility and aggregation with original SOI datasets.
• Estimated Completion Time: 1 to 2 minutes.
• Scoring and Aggregation:
  • Items 1 through 3 are scored 1 to 5 (or 1 to 9).
  • Items 4 and 5 are scored 1 to 5 (or 1 to 9).
  • Item 6 is reverse-keyed: On a 5-point scale, $1 \rightarrow 5, 2 \rightarrow 4, 3 \rightarrow 3, 4 \rightarrow 2, 5 \rightarrow 1$ (on a 9-point scale, score as $10 – X$).
  • Items 7 through 9 are scored 1 to 5 (or 1 to 9).
  • Facet Scores: Calculated by summing or averaging the three corresponding items for each subscale (Behavior = Items 1+2+3; Attitude = Items 4+5+6R; Desire = Items 7+8+9).
  • Total Global Score: Calculated by summing or averaging all nine items.
• Analytical Recommendation: Due to marked sex differences in Desire and Attitude, scores should be standardized within sex, analyzed separately for men and women, or have sex entered as a statistical covariate.

11. Permissions & Fee and Test Year

• Year of Initial Publication: 2008 (Scale construction documented in Penke, 2006; published in Journal of Personality and Social Psychology).
• Copyright & Licensing: The SOI-R is placed in the public domain for academic and non-commercial scientific research. No licensing fees or royalty payments are required for empirical research applications.
• Available Translations: The inventory has been adapted and validated in numerous languages, including Chinese, Dutch, English, French, German, Icelandic, Italian, Malaysian, Spanish, and Swedish.
• Official Repository: Scale instruments, translations, and demographic norm tables are openly maintained via Lars Penke's academic website (www.larspenke.eu/soi-r/).

12. References

• Asendorpf, J. B., & Penke, L. (2005). A mature evolutionary psychology demands careful conclusions about sex differences. Behavioral and Brain Sciences, 28(2), 275–276. https://doi.org/10.1017/S0140525X05230058
• 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. (2000). The evolution of human mating: Trade-offs and strategic pluralism. Behavioral and Brain Sciences, 23(4), 573–597. https://doi.org/10.1017/S0140525X0000337X
• Kinsey, A. C., Pomeroy, W. B., & Martin, C. E. (1948). Sexual behavior in the human male. W.B. Saunders.
• Kinsey, A. C., Pomeroy, W. B., Martin, C. E., & Gebhard, P. H. (1953). Sexual behavior in the human female. W.B. Saunders.
• Penke, L. (2006). Development of the revised Sociosexual Orientation Inventory (SOI-R) [Unpublished manuscript]. Institute of Psychology, Humboldt University of Berlin.
• 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
• Schmitt, D. P. (2005). Sociosexuality from Argentina to Zimbabwe: A 48-nation study of sex, culture, and strategies of human mating. Behavioral and Brain Sciences, 28(2), 247–275. https://doi.org/10.1017/S0140525X05000051
• 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
• Simpson, J. A., Wilson, C. L., & Winterheld, H. A. (2004). Sociosexuality and romantic relationships. In J. H. Harvey, A. Wenzel, & S. Sprecher (Eds.), The handbook of sexuality in close relationships (pp. 87–111). Lawrence Erlbaum Associates.
• Townsend, J. M., Kline, J., & Wasserman, T. H. (1995). Low-investment copulation: Sex differences in motivations and emotional reactions. Ethology and Sociobiology, 16(1), 25–51. https://doi.org/10.1016/0162-3095(94)00067-M
• 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-Atherton.
• Voracek, M. (2005). Shortcomings of the Sociosexual Orientation Inventory: Can psychometrics inform evolutionary psychology? Behavioral and Brain Sciences, 28(2), 296–297. https://doi.org/10.1017/S0140525X05400054
• Webster, G. D., & Bryan, A. (2007). Sociosexual attitudes and behaviors: Why two factors are better than one. Journal of Research in Personality, 41(4), 917–922. https://doi.org/10.1016/j.jrp.2006.08.007

13. Items of the Scale (Questionnaire)