Using Internet-Derived Data to Measure Religion: Understanding How Google Can Provide Insight into Cross-National Religious Differences

Abstract

Internet and social media data provide new sources of information for examining social issues, but their potential for scholars interested in religion remains unclear. Focusing on cross-national religion data, we test the validity of measures drawn from Google and Twitter against well-known existing data. We find that Google Trend (GT) searches for the dominant religions' major holidays, along with "Buddhism,"can be validated against traditional sources. We also find that GT and traditional measures account for similar amounts of variation, and the GT measures do not differ substantially from established ones for explaining several cross-national outcomes (e.g., fertility, circumcision, and alcohol use), as well as new ones (e.g., interest in religious buildings and sex). The Twitter measures do not perform as well. Our study provides insight into best practices for generating and using these measures, and offers evidence that internet-generated data can replicate existing measures that are less accessible and more expensive.