U.S. Congregational Life Survey, Wave 2, 2008/2009, Random Sample Attender Survey

Data Archive > U.S. Surveys > Religious Groups > Congregations/Other Organizations > US Congregational Life Survey > Summary


Over 500,000 worshipers in more than 5,000 congregations across America participated in the U.S. Congregational Life Survey (Wave 1 and Wave 2)—making it the largest survey of worshipers in America ever conducted. Three types of surveys were completed in each participating congregation: (a) an attender survey completed by all worshipers age 15 and older who attended worship services during the weekend the survey was given; (b) a congregational profile describing the congregation’s facilities, staff, programs, and worship services completed by one person in the congregation; and (c) a leader survey completed by the pastor, priest, minister, rabbi, or other principal leader. Together the information collected provides a unique three-dimensional look at religious life in America. (From Appendix 1, U.S. Congregational Life Survey Methodology, A Field Guide to U.S. Congregations, Second Edition).

This data file contains data for Random Sample Attenders participating in Wave 2 of the U.S. Congregational Life Survey. (U.S. Congregational Life Survey Wave 2 Random Sample Congregational Profile data, Leader data, and Associate Leader data will be provided in separate data files.)

Data File
Cases: 64,674
Variables: 148
Weight Variable: WT_WHTA, WT_WHTB
Data Collection
Date Collected: Fall 2008 and Spring 2009
Original Survey (Instrument)
USCLS Wave 2 Attender Survey
Funded By
The Lilly Endowment, Inc.
Louisville Institute
Presbyterian Church (U.S.A.)
Collection Procedures
Self-administered surveys
Sampling Procedures
The second sample (Wave 2) was identified and recruited by Harris Interactive. Of 1,330 congregations nominated and verified in the Wave 2 sample, 201 agreed to participate (15% of congregations) and 148 returned completed worshiper surveys (73% of those that agreed). Worshipers in these congregations completed 38,468 surveys. Congregations in the national random sample that participated in 2001 (Wave 1) were also invited by Harris Interactive to take part in Wave 2. In 2008, Harris Interactive verified and located 411 of the 434 congregations that participated in Wave 1. Of these, 145 agreed to participate in Wave 2 (35% of congregations), and 108 returned completed surveys from their worshipers (74% of those that agreed). These congregations returned 26,206 completed attender surveys.

In Wave 1 and Wave 2, denominations were also invited and encouraged to draw a random sample of their congregations. Denominational samples were large enough so that the results are representative of worshipers and congregations in each denomination. This allows denominations to compare their “typical” congregation and worshiper to congregations and worshipers in other denominations. Denominations participating in this oversampling procedure in Wave 1 and Wave 2 were Church of the Nazarene, Evangelical Lutheran Church in America (ELCA), Presbyterian Church (U.S.A.), Seventh-day Adventist Church, United Methodist Church (UMC), and United Church of Christ (UCC). In Wave 2, the Church of God (Cleveland, Tennessee) also participated by surveying a national random sample of their denomination’s congregations. (From Appendix 1, U.S. Congregational Life Survey Methodology, A Field Guide to U.S. Congregations, Second Edition).

Five versions of the attender survey were used: Standard (for all non-Catholic congregations) in English (Form A), Standard in Spanish (Form A-S), Standard in Korean (Form A-K), Catholic in English (Form B), Catholic in Spanish (Form B-S). The major difference between the Catholic and Standard versions is the substitution of “Mass” for “worship service” and “priest” for “pastor.” All Catholic parishes in the Random Sample received Catholic forms. The Standard survey was offered in Korean because of the prevalence of Korean congregations within the Presbyterian Church (U.S.A.).

In Wave 2, every Random Sample congregation received the Standard version (A or B) for 85 percent of the attender forms. The remaining 15 percent of forms were evenly divided among nine other versions—each with the same first three pages as the Standard version (Form A), but with a different back page. These different forms were topical in nature (e.g., giving, spirituality and health, sharing faith). So in a congregation with 100 people, 85 Form A or B surveys were distributed, and one or more of each of nine other versions of the survey were distributed. Every congregation did not necessarily get every one of the nine variations—for example, congregations with fewer than 50 total forms would not have received all of the nine different variations. Therefore, some small congregations might have gotten forms D through G for their 15 percent and another congregation might have gotten forms H through M and so on. The PIs determined the number of each of these nine versions to send to each congregation so that the overall number of responses for each version would be balanced. This procedure ensured that the PIs received enough completed forms of each of the nine different versions to be able to treat them as a random sample of worshipers. In addition, each oversample used a similar procedure to distribute a variety of back page forms to worshipers in that sample. Data from these back page forms are not being archived at this time, but the first three pages of data from these back page forms are included in the Random Sample Attender data being provided. This explanation is included so that those who use the data will understand why some worshipers lack data for items Q47 to Q60 that appeared on the back page of the Standard version of the survey.
Principal Investigators
Cynthia Woolever, Research Services, Presbyterian Church (U.S.A.), co-principal investigator

Deborah Bruce, Associate Research Manager, Research Services, Presbyterian Church (U.S.A.), co-principal investigator

Joelle Kopacz, Research Assistant, Research Services, Presbyterian Church (U.S.A.)

Ida Smith-Williams, Associate for Information, Research Services, Presbyterian Church (U.S.A.)
Related Publications
A Field Guide to U.S. Congregations: Who’s Going Where and Why. Second Edition. 2010. Cynthia Woolever and Deborah Bruce, Westminster John Knox Press.

A Field Guide to U.S. Congregations: Who’s Going Where and Why. 2002. Cynthia Woolever and Deborah Bruce, Westminster John Knox Press.

Beyond the Ordinary: 10 Strengths of U.S. Congregations. 2004. Cynthia Woolever and Deborah Bruce, Westminster John Knox Press.

Places of Promise: Finding Strength in Your Congregation’s Location. 2008. Cynthia Woolever and Deborah Bruce, Westminster John Knox Press.

Leadership That Fits Your Church: What Kind of Pastor for What Kind of Congregation. 2012. Cynthia Woolever and Deborah Bruce, Westminster John Knox Press.

Other reports are listed at: http://www.USCongregations.org
Weighting
As in the 2001 data set, there are two weight variables in the current file. The first set of weights (WT_WHTA) is used for estimating the the percentage of congregations that have a certain characteristic. The second set of weights (WT_WHTB) is used for estimating the percentage of attendees who attend congregations that have a certain characteristic. Use these weights when the attender data is aggregated to the congregational level.
Merging Datasets
The Random Sample Attender data can be linked to the Random Sample Profile data by the CONGREGA variable.