Development and Validation of the Trait Sexual Motivation Scale (TSMS)

Abstract

Sexual motivation, the interest in sexual activity, affects people’s thinking, feeling, and behavior. Common scales used to assess sexual motivation suffer from drawbacks that limit their validity and applicability. We therefore developed and validated the Trait Sexual Motivation Scale (TSMS), a brief, theory-driven self-report scale, over the course of four preregistered studies (N_total = 2,083). Results indicated good model fit, high internal consistency and stability of the second-order (i.e., trait sexual motivation) and first-order (i.e., cognition, affect, behavior) factor scores, and scalar measurement invariance for gender and relationship status. The TSMS correlated as expected with sexual and non-sexual constructs and predicted sexual outcomes cross-sectionally and prospectively in everyday life. Overall, the TSMS emerged as an economical, reliable, and valid measure of sexual motivation.

Acknowledgments

We thank Paula Laute and Tobias Ludwig for their valuable assistance in creating the online surveys and preparing the manuscript.

Data availability statement

The data that support the findings of this study are openly available on the Open Science Framework at https://osf.io/ux9nk/.

Open Scholarship

This article has earned the Center for Open Science badges for Open Data, Open Materials and Preregistered through Open Practices Disclosure. The data and materials are openly accessible at https://osf.io/ux9nk/.

Disclosure statement

The authors report there are no competing interests to declare.

Notes

1 We value all gender identities. Because academic (and social) discussions about possible gendered expressions of sexual motivation usually contrast male and female sexuality, we follow previous research and focus on individuals who self-identify as male and female.

1 Dynamic cutoffs are not yet available for higher-order models. We thus determined dynamic cutoffs for an isomorphic one-level model with correlated latent factors “Cognition”, “Affect”, and “Behavior”.

2 For reference, group differences based on manifest scores were t(656) = 11.57, d = 0.90, p < .001 for gender, and t(656) = 0.94, d = 0.07, p = .346 for relationship status.

3 We had pre-registered latent analyses using structural equal modeling, but one model showed insufficient fit. All other models fitted the data well and conclusions were identical to those drawn based on the manifest models.

4 To demonstrate convergent and incremental validity, we used alternative measures of sexual motivation based on a theoretical conceptualization closely related to the one proposed here. If one were to distinguish between excitatory and inhibitory processes as proposed by the dual control model of sexual behavior, variants of the SES/SIS scales would have been natural candidates (Carpenter, et al., 2011; Janssen et al., 2002).

5 We assessed only sociosexual attitudes because the validity and interpretation of items measuring sociosexual desire (e.g., "How often do you have fantasies about having sex with someone with whom you do not have a committed romantic relationship?") and behavior (e.g., With how many different partners have you had sex within the past 12 months?") are likely to depend strongly on a person's relationship status (Lippa, 2009). Although sociosexual desire is thought to be a specific form of general sexual desire (Penke & Asendorpf, 2008), it is impossible to empirically disentangle these concepts for single people, whose desire by definition cannot refer to a relationship partner. For those in sexually exclusive relationships, behavioral items are likely to be of limited information because, unlike singles, having multiple sex partners requires sexual infidelity.

6 We did not preregister comparing the TSMS with the SDI due to concerns about the SDI's validity for singles (Vallejo-Medina et al., 2020).

7 For technical reasons, some response periods exceeded 60 minutes. Rerunning the analyses based on responses collected within 60 minutes only did not alter any of the conclusions.