SOI-R vs. the Original SOI: Why the Revised Inventory Replaced It

June 16, 2026 · 3 min read

There are two well-known instruments for measuring sociosexual orientation, and they are easy to confuse: the original SOI (Simpson & Gangestad, 1991) and the SOI-R (Penke & Asendorpf, 2008). This site uses the SOI-R. Here is what changed and why the revised version became the standard in modern research.

The original SOI (1991)

Simpson and Gangestad introduced the Sociosexual Orientation Inventory in 1991. It combined a handful of items — including number of partners, number of one-time partners, anticipated future partners, frequency of fantasy about others, and attitudes toward casual sex — into a single composite score. That one number was meant to place a person somewhere on the restricted–unrestricted dimension.

The original SOI was hugely influential, but it had two well-documented limitations:

1. It mixed distinct components into one score. Past behavior, current attitudes, and ongoing desire were summed together, so two people with very different profiles could end up with the same number.
2. Its scoring was statistically awkward. The behavior counts were open-ended and highly skewed, and the items used different formats, which complicated weighting and made the composite hard to interpret.

The SOI-R (2008)

Penke and Asendorpf rebuilt the instrument as nine items measuring three separate but related facets:

• Behavior — past sociosexual behavior (number of casual / one-time partners).
• Attitude — evaluations of uncommitted sex.
• Desire — frequency of sexual desire and fantasy directed outside a committed relationship.

Each facet gets its own score, and they can also be averaged into an overall sociosexuality score. Critically, the three facets are correlated but distinct — they are not redundant — which is exactly why reporting them separately captures information the original single score threw away.

Why researchers switched

The SOI-R offered several concrete advantages that led to its widespread adoption:

• Cleaner measurement. All nine items use consistent 9-point response scales, giving the instrument better psychometric properties.
• More information. The three-facet structure reveals which aspect of sociosexuality is high or low — for instance, high desire but restricted behavior.
• Validated structure. The facet model was supported across large samples and has held up in cross-cultural and follow-up research.

Which one should you use?

For almost all purposes today, the SOI-R is the better choice: it is the current standard, it is easy to administer, and it gives a richer picture than a single number. That is why the test on this site uses the SOI-R, and why the statistics page breaks results down facet by facet.

For a step-by-step walkthrough of how the SOI-R is calculated, see SOI-R scoring explained.

References

Penke, L., & Asendorpf, J. B. (2008). Beyond global sociosexual orientations. Journal of Personality and Social Psychology, 95, 1113–1135.

Simpson, J. A., & Gangestad, S. W. (1991). Individual differences in sociosexuality. Journal of Personality and Social Psychology, 60, 870–883.