2023 marks the release of several datasets and analytical tools capable of generating new insights into religious identity and change. The proposed panel presents four of these datasets, all of which take account of multiple dimensions of religiosity and incorporate multiple waves of data collection. (1) DIM-R+ is the longed-for harmonization of numerous multi-wave, multi-national surveys (ISSP, WVS, EVS, ESS). (2) RICH-USA is a systematic harmonization of numerous surveys on religious and identity and change in the USA. (3) RICH-India is a systematic harmonization of numerous surveys on religious and identity and change in India. (4) ARDEMIS is a system for visualizing the impact of measurement assumptions on religious demographic projections. These publicly available datasets and tools have unusually broad potential for advancing understanding of religious identity and change and thus will be of strong interest to scholars of religion who make use of demographic religion data in their research.
The Dimensions of Religiosity Dataset (DIM-R+) is a harmonization of all of the currently available waves of the European Social Survey (ESS), the World Values Survey (WVS), the European Values Study (EVS), and the International Social Survey Program (ISSP) that includes data from more than a million individuals. In addition to harmonizing existing variables, DIM-R+ includes theoretically well-grounded syntheses of variables for key measures of religiosity: religious affiliation, participation in public religious practices, participation in private religious practices, self-declared religiosity, and supernatural belief. The usefulness of DIM-R+ can be appreciated by using it to improve the robustness and generality of David Voas’ fuzzy fidelity theory, which has been a major development in our understanding of the secularization process.
Measures of religion have been available from US polls and surveys since the 1940s, and yet, the characteristics and trajectory of religious change in the country is a topic of intense debate. The Religious Identity and Change in the USA (RICH-USA) harmonizes existing measures of religion present in historical polls and more contemporary surveys, permitting a more comprehensive view of religion in the US and how it has changed over time. Like DIM-R+, RICH-USA includes measures of religion along multiple dimensions including identity, public religious service attendance, private practices, the personal importance of religion, and supernatural worldview. The creation of RICH-USA is an unprecedented effort to harmonize eight decades of religion data to connect past and contemporary research on religion in the US.
The Religious Identity and Change in India dataset (RICH-India) combines an historical dataset for religiosity in India with more recent quantitative data based on the Indian national census and other datasets. This paper describes how RICH-India was constructed, includes information on how the codebook was created and the proxies used to measure historic religiosity, and raises questions about possible methods for linking and relating the two parts of the dataset. No comprehensive quantitative sources exist that measure religiosity in India before the early 1990s. Using a categorical coding scheme, we generated a historical dataset by performing a content analysis of ethnographic profiles in the Anthropological Survey of India’s People of India project publications and other earlier twentieth-century sources containing qualitative statements on religious beliefs and practices. A quantitization process translates coded qualitative data into quantitative data.
Religious population projections are politically sensitive click-bait. Methodological self-awareness calls for understanding to what extent measurement assumptions affect the population projections in which they are employed. Traditional demographic methods do not support the required sensitivity analyses. But emerging demographic methods from computational social simulation possess the flexibility to determine precisely how sensitive population projections are on numerous measurement assumptions, in all possible combinations, simultaneously. The Assumption-Relative Religious Demographics Information System (ARDEMIS) is a computer-simulation-based system for estimating and visualizing the impact of measurement assumptions on religious demographic projections. Just as we learn more from the distribution of a statistic than from its mean, so we learn far more from population projections produced in the ARDEMIS way than we do from a projection presented with no sense of its likelihood or stability in the face of alternative measurement assumptions.