For evangelicals to keep offsetting losses in future generations, they will need to become more racially diverse. The unaffiliated have had a much more dramatic trajectory, starting at just 1 in 20 of GSS respondents back in The nones experienced big jumps in and , and have added, on average, Their continued growth has been well-documented and at this point, expected.
The difference between these groups is not statistically significant at this point; but I see no reason to think that the nones will level off anytime soon. If the nones maintain their growth while evangelicals stagnate, it is statistically inevitable that those of no religious faith will be the largest group in America in the next five years.
Last year also brought a small uptick in the number of people who identify as mainline Protestants —their first increase in nearly 20 years. They have suffered tremendous declines over the last 40 years, going from 30 percent of the population in to just 10 percent in But in , that share bumped to It would unwise to extrapolate too much from a single-year increase.
Evangelicals declined 1. It is too early to say that these evangelical losses were mainline gains; however, if this trend continues or possibly accelerates in the next two years, it would give some credence to the claim that Trump has led to evangelical defections. The GSS also shows that mainline Protestants have continued to move toward the left of the political spectrum, while evangelicals maintain their shift to the right.
This continuing polarization may lead more moderate evangelicals to seek out a church that shares their political worldview. Ryan P. Pew Research Center now uses as the last birth year for Millennials in our work. President Michael Dimock explains why. About Pew Research Center Pew Research Center is a nonpartisan fact tank that informs the public about the issues, attitudes and trends shaping the world.
It conducts public opinion polling, demographic research, media content analysis and other empirical social science research. Pew Research Center does not take policy positions. It is a subsidiary of The Pew Charitable Trusts. Newsletters Donate My Account. Research Topics. Billy Graham preaching in As the country remembers Graham, here are five facts about American evangelical Protestants.
PRRI has been collecting and aggregating data on religious diversity in America continuously since The combined dataset contains interviews with , Americans across all 50 states from to This dataset was merged with American Community Survey ACS data on 3, counties across all 50 states to produce, for the first time, a rigorous estimate of the religious demographics of every county in the U.
Because there is no measure of religious affiliation on the ACS or any other U. To measure religious diversity in the United States, this report uses an index developed to measure variations in the concentration of global religious populations.
The index is calculated so that a score of 1 signifies complete diversity—every religious group is of equal size—and a score of 0 indicates a complete lack of diversity and one religious group comprises the entire population of a given county [7]. Results for all demographic, religious affiliation, and political affiliation questions were based on 50, bilingual telephone interviews including 35, cell phone interviews conducted between January 7, and December 20, by professional interviewers under the direction of SSRS.
National results for questions on specific issues are based on a subset of 10, telephone interviews including 6, cell phone interviews conducted in 10 weeks spread across the year. State level results for questions on specific issues are based on the national subset plus an oversample of 10 key states in 30 additional weeks, resulting in 23, interviews across all 50 states plus the District of Columbia including 16, cell phone interviews.
Throughout , at least 1, interviews were completed each week, with about interviews conducted among respondents on their cell phones. Each week, interviewing occurred over a five- or six-day period, starting Tuesdays or Wednesdays and going through Sunday or Monday.
The selection of respondents within households was accomplished by randomly requesting to speak with the youngest adult male or female currently living in the household.
Data collection was based on stratified, single-stage, random-digit-dialing RDD of landline telephone households and randomly generated cell phone numbers. The sample was designed to represent the total U. The landline and cell phone samples were provided by Marketing Systems Group. The weighting was accomplished in two separate stages. In the second stage, sample demographics were balanced to match target population parameters for gender, age, education, race and Hispanic ethnicity, region U.
Census definitions , population density, and telephone usage. The population density parameter was derived from Census data. All other weighting parameters were derived from an analysis of the U. The sample weighting was accomplished using iterative proportional fitting IFP , a process that simultaneously balances the distributions of all variables. Weights are trimmed so that they do not exceed 4. The use of these weights in statistical analysis ensures that the demographic characteristics of the sample closely approximate the demographic characteristics of the target populations.
Table A1 shows the sample sizes for all 50 states for the sample. Table A2 shows sample sizes for religious affiliations in the report. The design effect of 1. In addition to sampling error, surveys may also be subject to error or bias due to question wording, context, and order effects.
The complete dataset contains , interviews. The survey was made possible by generous support from the Arcus Foundation, the E. Weighting processes were identical to that described above for the data using the most recently available National Health Interview Survey and Current Population Survey.
Table A1 shows the sample sizes for all 50 states for the overall dataset. NORC has developed and implemented small area estimation models on a number of key projects for government and social science surveys. The estimates have varying measurement error based on the size of the county and the number of surveys completed in each county. Smaller counties have larger measurement error. The small area modeling approach for this project modeled weighted survey estimates of each religious affiliation by county and year.
Since the distribution of religious denominations are correlated across time, the small area model captures county-level correlations across time via a time-series component to improve estimates for the most current year. Furthermore, the model incorporates external auxiliary data that is predictive of the distribution of religious denominations at the county-level from the 5-year American Community Survey.
The best set of predictors for each religious denomination was selected using a variable selection approach known as LASSO. Finally, model-based estimates for each religious denomination were generated for all counties in the U. Respondents who self-identify as white, non-Hispanic, Protestant and do not identify as born-again or evangelical are categorized as white mainline Protestants.
These groups are combined due to small sample sizes that prevent individual analysis. Laine, The Antitrust Bulletin, June 22,
0コメント