The Role of Religiosity in Adolescents' Compulsive Pornography Use: A Longitudinal Assessment

ABSTRACT

Despite increasing concerns that adolescent men are vulnerable to developing compulsive pornography use, little research has been done in this area. Given recent theorizing and research concerning moral incongruence, we hypothesized that symptoms of compulsive pornography use should generally be associated with higher levels of pornography use and increased growth in male adolescent pornography use over time, but that this pattern would be attenuated among very religious participants. These hypotheses were tested with mixed effects growth models using two independent panel samples of male Croatian adolescents. As expected, adolescent men who reported features of compulsive pornography use tended to exhibited higher levels of pornography use. However, contrary to expectations, increased growth in pornography use was limited to more religious compulsive users. Compared to nonreligious compulsive users, these adolescents started with lower initial levels of pornography use and their use increased over time at a greater rate of change. This study's results are the first to suggest that some adolescent men who report high levels of pornography use tend to exhibit symptoms of compulsive use, which highlights a need for counseling and therapeutic attention. Our findings also have implications for the emerging theory of moral incongruence.

Notes

1 A comparable latent growth model using structural equation modeling (SEM) was also preregistered (see here: https://osf.io/w4ckp/). The results of both approaches were quite similar. We decided to present the linear mixed-modeling approach because the SEM approach required that religiosity be arbitrarily dichotomized to test H3.

2 With controls included, the target association between the frequency of pornography use and compulsive use was confirmed in the Rijeka panel. Although this pattern was not confirmed in the Zagreb panel, it is very likely that the loss of the three-way interaction between time, compulsive pornography use, and religiosity occurred because of a substantial sample loss that resulted from missing data on the control variables rather than the effect of the control variables themselves.

Additional information

Funding

This work has been fully funded by Croatian Science Foundation grant number 9221 awarded to the second author.