Pornography, Sexual Insecurity, and Orgasm Difficulty

ABSTRACT

While recent meta-analyses have provided answers to a number of historically contentious debates about correlates of pornography use, several questions remain unanswered. Whether pornography is associated with sexual functioning is one such question. Informed by theorizing on sexual scripting, social comparisons, and sexual objectification, the present study examined the possibility that pornography is related to orgasm difficulty through sexual insecurity (i.e., insecurity about one’s sexual performance and sexual attractiveness). Data were from the National Survey of Porn Use, Relationships, and Sexual Socialization (NSPRSS), a U.S. population-based probability study. There was an indirect effect of pornography consumption frequency on orgasm difficulty through sexual insecurity. Participants who used pornography more frequently reported higher levels of sexual insecurity, and higher levels of sexual insecurity predicted orgasm difficulty. There was also an indirect effect of partner pressure to view pornography on orgasm difficulty. Higher levels of partner pressure to view pornography were associated with higher levels of sexual insecurity, which in turn predicted orgasm difficulty. Results were indistinguishable by gender and maintained after controlling for a number of potential confounds. These findings suggest that some men and women’s personal and (pressured) partnered pornography consumption have the potential to increase orgasm difficulty through bodily and performance insecurity.

Acknowledgments

The authors are grateful to the following for their generous support of our research: Julie Parker Benello, Abigail E. Disney, Natasha and David Dolby, Embrey Family Foundation, The Fledgling Fund, Ruth Ann Harnisch and The Harnisch Foundation, Chandra Jessee, Suzanne Lerner, Cristina Ljungberg, Ann Lovell, Nion McEvoy, Regina K. Scully, Artemis Rising Foundation, Lindsey Taylor Wood and Jacki Zehner. We are also grateful to Jill Bauer, Ronna Gradus, and Rashida Jones for their participation in survey development, including their review and feedback on survey drafts.

Notes

1. Assuming a conventional two-tailed null hypothesis test (alpha error probability .05) and .80 statistical power, the sample size required to detect a correlation of .10 (the lower-bound of Cohen’s (1992) “small” effect size range) is 783.

2. These percentages are based on 1919 responses, as ten participants for the present analysis (0.5%) declined to answer this question.

3. This was the only measure to depart from conventional skewness guidelines (i.e., to have skew > 2) (Byrne, 2010; Hair et al., 2010). A logarithmic transformation was conducted to reduce its positive skew and analyses were re-conducted using this transformed variable. Results using the transformed variable paralleled results using the variable in its original form.