Polling and survey data are part of the lifeblood of American political sport and ideological ammo on the nation's culture-war fronts. Politicians of all stripes use polling to their advantage. The media feeds on “polls show” and “the survey says” findings that, in turn, are integrated into news cycles and become mantras of other assessments of American life. Polls and surveys also play a powerful role in constructing, influencing—and distorting—how Americans understand religion.
Inventing American Religion offers a historical narrative of how the nation's religious behavior came to be measured as yet another American social and cultural project. While there were various late nineteenth- and early twentieth-century efforts at canvassing for religion-related data, it was George Gallup who, in 1935, designed the first national poll about American religion based on innovative scientific sampling methods. A diverse polling industry followed, with Gallup's efforts remaining one of the most important long-term diviners of religion trends.
As a sociologist and one of academe's most prolific scholars of religion, Robert Wuthnow is no stranger to the topic. As Wuthnow tells the story of how American views of religion have been shaped by polls and surveys, his concerns become obvious. These include a standard fare of methodological criticisms relating to questionable poll and survey wording; treating religion as an objectified thing; multiple forms of response bias (time and fatigue factors, question sequencing, etc.); confidence intervals that are no longer meaningful; self-deceptive reporting of beliefs and practices; and bad or inadequate sampling strategies. This latter problem has become especially egregious in the face of plummeting response rates that have led to reported trends based on exceedingly small numbers of actual respondents.
Wuthnow's larger concern here is what gets distorted about our perceptions of religion by way of polling and survey approaches. Both—but especially the former—have often done little more than affirm many of the nation's shared beliefs about its nature and destiny, reflected generalizations pertaining to white middle-class Protestants, or shown religion always to have favorable consequences. Both have also produced factually mistaken assertions. One classically misleading “finding” has been the overestimation of the number of people who actually attend religious services.
Wuthnow also draws attention to the failures of polls and surveys to touch much of the local, personal, and familial nature of religion. Nor have they been reliable measures of the cultural meaning of religious symbolism, or of what actually matters about religion in America's increasingly diverse cultural milieu. The author attributes this latter shortcoming to “inadvertent white norming” surrounding much of what we assume about the topic. Wuthnow also examines some of the tensions between pollsters and survey research by academic scholars of religion who are generally more interested in establishing causal relationships than in reporting information.
Since the 1990s, public confidence in polls and surveys has declined dramatically. Bias, scientific unreliability, and widespread recognition that polls can show nearly anything and be interpreted in multiple ways have produced ever greater skepticism about their merit—in the measure of any social or cultural domain.
Wuthnow's overall take is that while polling and survey information can be helpful in establishing some of the general parameters of religion in American culture, neither is sufficient in itself. Better to scrutinize both, but never rely solely on either. In the context of survey research in particular, ethnographic strategies enhance the possibilities of richer insight.
Inventing American Religion is an important read for social and behavioral science researchers. It is an essential read for scholars and nonscholars alike (especially journalists and religious leaders) concerned with a more accurate and inclusive understanding of the complexities of the American religious landscape. More attention might have been given to why some faith traditions (mainline Protestantism) have been more receptive to polling and surveys—along with social science scrutiny in general—than others (Catholicism). Nevertheless, Wuthnow's study is a solid and engaging exploration of the “invention” of American religion—past and present—by polls and surveys, and the pluses and minuses of each.
