|
ABSTRACT
Sociologists of religion often overlook the role of demography. An exception to this rule is found in the work of Pippa Norris and Ronald Inglehart, who link religious decline to human development and the demographic transition. However, their individual-level thesis is based on bivariate trends, with multivariate analysis limited to the aggregate level. In this article, I test their thesis at the individual level using data from the World Values Surveys across a wide range of countries. Analysis of aggregate trends shows that measures of human development that appear significant in bivariate correlations do not survive multivariate, time-series scrutiny. Moreover, I deploy multilevel analysis to explain why aggregate trends provide a misleading picture of how rising national education and income levels affect individuals’ religious beliefs. The results cast the developmentalist version of the secularization thesis into doubt. Instead, I suggest that religious belief becomes deregulated and increasingly varied in modern societies as religiosity takes on a self-conscious, rather than taken-for-granted, character. The demographic advantage that religious populations have suggests that the future of secularization, far from confirming a secular teleology, remains indeterminate. |
