History of American Religion, 1770-1930

Data Archive > U.S. Church Membership Data > State-Level > Summary

Gurrentz, B. (2020, March 4). History of American Religion, 1770-1930.
The History of American Religion, 1770-1930 dataset allows users to examine historical change in religious and social measures in the United States from 1770 to 1930 using state-level historical data. Through the aid of U.S. Censuses, ICPSR, the American Presidency Project, and prominent religion scholars like Roger Finke and Rodney Stark, the file includes more than 290 historical variables that measure the religious adherence rates, population estimates, voting behavior, employment, and other social indicators over time.
Data File
Cases: 50
Variables: 272
Weight Variable: None
Data Collection
Date Collected: 1770-1930
Collection Procedures
For in-depth information on historical data collection and measurement, please note the citation information on variable labels and consult the reference list of sources below:

Department of Commerce and Labor, Bureau of the Census. Census of Religious Bodies, 1926, Part I: Summary and Detailed Tables. Washington, D.C.: Government Printing Office, 1930.

Finke, Roger and Rodney Stark. 1986. "Turning Pews into People: Estimating 19th Century Church Membership." Journal for the Scientific Study of Religion 25 (2): 180-192.

Haines, Michael R., and Inter-university Consortium for Political and Social Research. Historical, Demographic, Economic, and Social Data: The United States, 1790-2002. Ann Arbor, MI: Inter-university Consortium for Political and Social Research [distributor], 2010-05-21. https://doi.org/10.3886/ICPSR02896.v3

Stark, Rodney. America 1920-1980. MicroCase Corporation.

Stark, Rodney and Roger Finke. 1988. "American Religion in 1776: A Statistical Portrait." Sociological Analysis 49 (1): 39-51.

Statistical Abstracts of the United States, Series. U.S. Census Bureau. Retrieved 8/12/19 from: https://www.census.gov/library/publications/time-series/statistical_abstracts.html

Woolley, John T. 1999. The American Presidency Project. Santa Barbara, Calif.: University of California. Web Source. Retrieved 4/5/19 from: https://www.presidency.ucsb.edu/index.php
Principal Investigators
Benjamin Gurrentz, The Association of Religion Data Archives

Note on Adherence Rates
The adherence rate variables (adrt_[YEAR]) are calculated based on Roger Finke and Rodney Stark's 1986 article entitled "Turning Pews into People: Estimating 19th Century Church Membership." The 1850, 1860, and 1870 estimates are highly reliable predictive regression estimates of adherence based on church organizations, seats, and property measures in the U.S. Census. This is because religious adherence estimates were not available during this time period, but were later collected from 1890 into the early 20th century. Moreover, all adherence rates adjust for the fact that some religious organizations count children as religious adherents (e.g., Roman Catholics and Lutherans) while others do not (e.g., Baptist or Methodists). For denominations that do not include children, adherence rates were adjusted to include them using the following formula: Adherents = Reported members * (Total population/(Total population - Children Under 13)).
Note on Population Estimates
Most of the population estimates in the dataset come from Michael Haines and ICPSR's historical dataset: Historical, Demographic, Economic, and Social Data: The United States, 1790-2002. These estimates derive from U.S. Census estimates and other resources, but you may find discrepancies with some historical sources during periods in which new territories are incorporated as states. The ICPSR dataset tends to use population estimates at the time of collection even if certain areas later become recognized as separate states. For example, West Virginia was originally part of Virginia prior to the 1861, so population estimates for that area are coded as Virginia up until incorporation.
Bookmark and Share