Measuring the Multidimensional Construct of Pornography: A Long and Short Version of the Pornography Usage Measure

With the increase in the availability and usage of pornography, the research on the effects of pornography has also increased. This research has uncovered several controversies in the field regarding how pornography usage influences attitudes, sexual behaviors, and relationships. However, many of the measures of pornography are problematic as there is often little reliability and validity information for them and it is not clear that participants in these research studies are referring to the same types of materials when they answer pornography usage questions. Consequently, many of the research findings are suspect and it is crucial to develop reliable and valid scales to measure general pornography usage. In this study, we present both a 20-item and a 7-item version of the Pornography Usage Measure (PUM) that is based on extensive previous research on what types of materials individuals consider pornographic and that indicates pornography is a multidimensional construct. We evaluated the reliability and validity of both versions by using an MTurk sample of 934 males and 705 females (N =1639 total) to conduct confirmatory factor analyses, item response analyses, and structural equation model analyses. These analyses demonstrated that there were adequate reliability and early evidence for content, construct, concurrent, and predictive validity for both versions of the PUM. This measure could improve the quality of future research on pornography by providing more consistency between different studies about what is being measured when individuals indicate their pornography usage patterns.