--- - |2 Abstract A popular measure of religious identification poses several problems for researchers because of its lack of conceptual and historical grounding. The scheme truncates continuity along the well‐established sect‐church continuum, and poses problems for both theory and methodological operationalization. We reiterate the importance of key sociological and historical foundations for classifying religious groups, provide an empirical assessment of a more exacting scheme compared to the newly dominant classification, and supply a coding system for the General Social Survey which could also be applied to other studies. We conclude by discussing the usefulness of our categorization for advancing social science research on religious identification. - 'Journal for the Scientific Study of Religion, EarlyView. '