Surveying Denominations and Congregations: An Epilogue

Abstract

The four preceding papers describe research that illustrates the problems and prospects of survey research conducted on denominations and congregations. As a whole, this set of papers underscores the present limitations of surveys for assessing beliefs and practices, and the local congregational environments which shape these phenomena. Virtually all of the researchers, with the possible exception of Cornwall and Cunningham (who appear to have the singular advantage of working with relatively complete, detailed worldwide population listings for LDS members), are handicapped somewhat in their sampling procedures. Both those studying "mainline" Protestant denominations (Roozen and Carroll, and Hadaway), as well as those studying Catholics (Leege and Welch), suffer from the absence of fully enumerated populations and nationally centralized lists. And this problem would seem to be greatly compounded for those who attempt to characterize religious subgroups that cross-cut denominations (e.g., Evangelicals and Pentecostals). All of the researchers also acknowledge the difficulties involved in identifying and correcting for biases introduced by nonrespondents. This can pose a serious problem, especially when policy decisions about denominational outreach programs are made based on information drawn predominantly from relatively active members. It seems unlikely that such decisions would accurately reflect the needs or interests of more marginal members. In addition, the papers reveal problems that confront some denominational surveys more than others. For example, Hadaway indicates some of the unique problems that derive from the characteristic congregational polity structure of Southern Baptist churches. Moreover, the lack of strong, uniformly accepted criteria for defining a church's membership in the Southern Baptist Convention may create additional difficulties for sample design in both denominational studies of individual members and congregations. Although it probably affects all denominations to some degree, "parish-shopping" appears to be at least a bit more salient for surveys of Catholic parishioners and their parishes. Leege and Welch suggest this phenomenon can be particularly troublesome for studies that focus on the parish as a social context, and they alert others to these potential difficulties. It is also interesting to note that three of the research teams surveying notably different denominations (viz., Catholics, LDS, and Presbyterian congregations), report similar national response rates for members, averaging close to 60%. This pattern of convergence confirms the utility of multiple wave procedures. But it also implies that survey researchers may have to redefine their standards for "acceptable" response rates, as the problem of beleaguered, over-surveyed respondents becomes more pervasive and apparent. Although the authors have been both critical and realistic in discussing the attendant problems of denominational surveys, one should not be unduly pessimis-