Cross-Survey Analysis to Enumerate Religious Minorities in the United States
*Elizabeth Tighe, Brandeis University
Keywords: enumeration, religious minorities, cross-survey analysis
A defining characteristic of the United States is its religious diversity. An additional characteristic is the separation of church and state, which precludes the US government from collecting data on the religious identification of citizens. To study religion in the US, researchers have relied on congregation studies or single surveys with nationally representative samples such as the American Religious Identification Survey (Kosmin & Keysar, 2008) and the US Religious Landscape Survey (Pew Forum, 2008). Reliance on single surveys can be problematic, particularly for estimation of smaller religious denominations. Many include too few respondents to be able to describe small groups reliably. Further, any single survey can contain systematic errors arising from questionnaire design, sampling, sponsorship, and “house” effects. Although targeted sampling can be done to obtain over-samples of particular denominations, obtaining over-samples of multiple small groups simultaneously for purposes of comparative analysis can be prohibitive to social researchers, as can the potential biases introduced by methods to obtain over-samples (e.g., those more strongly affiliated more likely to be identified through targeting). To address these challenges, we draw on methods of cross-survey analysis (e.g., Park, Gelman & Bafumi, 2004) to systematically combine data across a sample of over 150 independent surveys. Hierarchical Bayesian analysis is used to account for clustering of respondents within surveys and to examine potential biases across surveys. Estimates are post-stratified on basic demographics such as age, sex, race, educational attainment and geographic dispersion. The results from this analysis provide a methodological framework for studying other small groups and can serve as an important source of data for targeted surveys by providing a reliable resource for post-stratification adjustments based on religious identification.