Examining the Psychometric Properties of the Sexual Initiation Strategies Scale (SISS): A New Self-Report Measure of Sexual Aggression Perpetration History

ABSTRACT

Psychometrically-sound measures of sexual aggression are essential for identifying risk factors for perpetration and examining the behavioral impact of prevention interventions. This project reports on the psychometric evaluation of a new measure of sexual perpetration—the Sexual Initiation Strategies Scale (SISS)—which was designed to correct for weaknesses in prior measures. In Study 1, scores on the SISS demonstrated reasonable convergent and divergent validity and test-rest reliability among a sample of mixed gender college students (N = 575). Participants were more likely to endorse having engaged in sexually aggressive behavior on the SISS than on two other measures of sexual aggression. In Study 2, rates of endorsement on the SISS also were higher than on the most commonly-used measure of perpetration in a community sample of men and women (N = 551) in which gay, lesbian, and bisexual participants were oversampled. In general, the SISS seemed to perform well as a measure of sexual aggression perpetration in both men and women and in those reporting both same- and other-gender sexual aggression. Advantages and disadvantages of the SISS as compared to other measures are discussed.

Acknowledgment

The author would like to thank the following individuals who provided feedback related to item development: Sarah Buday, Ana L. M. Jaramillo-Sierra, Kristen Jozkowski, Sara Kern, Monica Perez Trujillo, and Emily Strang. The author also thanks Errin Fornicola and Joseph McBride for their help with data collection and the anonymous reviewers for their helpful feedback on earlier drafts of this paper.

Disclosure Statement

No potential conflict of interest was reported by the authors.

Supplementary Material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/00224499.2022.2130856

Notes

¹ Data were mean imputed for two people on the Sexual Narcissism Scale, 14 people on the Updated Illinois Rape Myth Scale, zero people on the Male Rape Myth Scale, zero people on the Adversarial Heterosexual Beliefs Scale, and one person on the BIDR subscales. We also ran all analyses excluding participants with any missing data on a pairwise basis, and all results were nearly identical, except that with the non-imputed variable, the SISS did demonstrate incremental validity over the SSS in predicting rape myth acceptance.

² Because the SISS was always presented first, it is possible that the higher rates of reporting on the SISS as compared to the other measures were artifacts of order effects; for example, answering the SISS may have caused participants to reevaluate their own behavior and endorse less aggression on the later measures. To evaluate that possibility, exploratory comparisons were conducted across similar items on the SISS and the SSS and SES-SFP. Endorsements on the SSS “asking them repeatedly” item was compared with endorsements on the SISS “asking repeatedly” item; interestingly, endorsement was actually higher on the SSS (6.1%) than SISS (2.8%), p < .001, φ = .47. A composite anger item was created for the SES-SFP (combining use of anger to obtain oral, anal, or vaginal sex) and was compared to the “acting angry, upset, or withdrawn … ” item on the SISS. Endorsement of anger was similar on the SES-SFP (3.8%) and the SISS (3.2%), p = .70, φ = .15. Thus, higher rates of reporting on the SISS as compared to the other measures did not appear to be caused merely by order effects.

³ It is not possible to directly compare number of incidents of sexual perpetration on the SES-SFP and the SISS. The SISS asks about use of tactics, but not the number of times that they were used. The SES-SFP does allow participants to indicate whether each act and tactic combination occurred 0, 1, 2, or 3 or more times; however, even on the SES-SFP, it is not possible to calculate a precise incident rate because multiple acts/tactics could occur during a single incident.

Additional information

Funding

At the time that these data were collected, Zoë Peterson was at University of Missouri-St. Louis. This project was funded in part by a grant from the College of Arts and Sciences at the University of Missouri-St. Louis.