Is the Link Between Pornography Use and Relational Happiness Really More About Masturbation? Results From Two National Surveys

Abstract

Numerous studies have observed a persistent and most often negative association between pornography use and romantic relationship quality. While various theories have been suggested to explain this association, studies have yet to empirically examine whether the observed link between pornography consumption and relationship outcomes has more to do with solo masturbation than actually watching pornography. The current study drew on two nationally representative data sets with nearly identical measures to test whether taking masturbation practice into account reduces or nullifies the association between pornography use and relational happiness. Controls are included for sex frequency and satisfaction, depressive symptoms, and other relevant correlates. Results from both the 2012 New Family Structures Study (NFSS; N = 1,977) and the 2014 Relationships in America survey (RIA; Ng = 10,106) show that masturbation is negatively associated with relational happiness for men and women, while pornography use is either unassociated or becomes unassociated with relational happiness once masturbation is included. Indeed, evidence points to a slight positive association between pornography use and relational happiness once masturbation and gender differences are accounted for. Findings suggest that future studies on this topic should include measures of masturbation practice along with pornography use and that modifications to theories connecting pornography use to relationship outcomes should be considered.

Acknowledgments

The author wishes to thank Paul J. Wright for feedback on the content of this article and Mark Regnerus for the use of the 2014 RIA survey data.

Notes

¹ Because the NFSS includes only adults ages 18 to 39 while the RIA includes adults ages 18 to 60, the full models described were also estimated with the RIA data restricted to ages 18 to 39. The results were identical in substance and statistical significance to what I present in Tables 3–6 (available upon request), and thus I use the full sample to maximize available data.

² Because these variables are moderately associated with pornography use, masturbation, and relational happiness in both surveys (see correlations in Table 2), models were also estimated without these variables (available upon request) to discern whether they could be mediating an effect. While each contributed to the variance explained, removing these variables did not change the substantive findings described in the Results section.