FACT 2008 Survey Methods

The CCSP Research Taskforce designed the FACT2008 survey questionnaire. A copy of the questionnaire is available and should be consulted for exact question and response category wordings.  It was completed by a key informant in each surveyed congregation, most typically the senior or sole clergy leader.

The survey sample is an aggregation of three different layers designed to represent the universe of American congregations.  One layer is a mail and web survey of a random sample of 3,000 U.S. congregations conducted for CCSP by the research services office of the Presbyterian Church (U.S.A.).  A second layer is a telephone survey of a random sample of 1,000 congregations conducted by the Center for Creative Ministry. The third layer is a set of mini surveys conducted by 12 of the CCSP partner denominations and faith groups.  These mini surveys used the FACT2008 questionnaire, but were conducted by the respective denomination or faith group as a supplement to the general samples.  The general survey sample was generated by the CCSP Research Taskforce from a larger random national sample of congregations purchased from MCH at www.mailings.com.  

The final aggregated data set contains questionnaires from 2,527 congregations.  To better represent national population parameters a two stage weighting procedure was used.  To mitigate the over representation of those denominations and faith groups that contributed supplemental survey data, the total aggregated responses were weighted to the population parameters for faith families presented by Hadaway and Marler [C. Kirk Hadaway and Penny Long Marler, How Many Americans Attend Worship Each Week?  An Alternative Approach to Measure, Journal for the Scientific Study of Religion (2005) 44(3):307-322, Table 2].  

Then to further enhance national representation, the total aggregated dataset was also weighted to size of congregation and rural/city/suburban location parameters found in the FACT2000 national survey of 14,301 congregations.  This is the same weighting procedure used for FACT2005 and therefore has the added benefit of making the report’s trend comparisons more robust.  The sampling error for FACT2008 type samples is difficult to calculate with precision.  We estimate it to be +/- 4 percent. 

Denominations and faith groups created supplemental surveys in addition to the random survey.  The faith group partners that contributed supplemental survey data include:  Baha’i Faith in the United States, Interdenominational Theological Center (representing seven historically Black denominations), Evangelical Lutheran Church in America, Episcopal Church, Church of Jesus Christ of Latter-day Saints, Church of the Nazarene, Standing Conference of Canonical Orthodox Bishops in the Americas, Presbyterian Church (U.S.A.), Seventh-day Adventist Church, Synagogue 3000, United Methodist Church and megachurches in the U.S..

Download the Questionnaire