“I’m Feeling Lucky”: Identifying Novel Correlates of Sexual Infidelity via Google Search Databases

Brandon Wagner, University of North Carolina at Chapel Hill

Today, survey research increasingly confronts declining response rates and limited financial support. Developing new and reliable complementary data sources is of paramount importance, particularly for the study of sexual behaviors. Using digitally derived data generated by respondent’s normal activities offers a potential solution; this data is not dependent on recruiting subjects, is relatively low cost and may even allow better measures of respondent action and intention than self-report. In this paper, I test the potential of search engine databases, namely Google, for sexual research. I locate potential correlates of perceived sexual infidelity, a topic problematic for measurement in most research designs. Testing the quality of this data source, I validate the uncovered correlations both theoretically and empirically using individual data from the National Longitudinal Study of Adolescent Health, Wave 4. In addition to identifying correlates of perceived infidelity, findings highlight the potential of search data as a research tool.