Michael Kammen on public surveys
THE AVERAGED AMERICAN:
Surveys, Citizens, and the Making of a Mass Public.
By Sarah E. Igo. Harvard Univ. Press. 398 pp. $35
Reviewed by Michael Kammen
How to count Americans accurately has been a contentious question ever since the first federal census was undertaken in 1790. A century ago, foundations and commissions began to support more focused surveys, usually with an eye to policy, such as tenement housing reform in New York City. During the 1920s and ’30s, with the development of quantitative methods in the social sciences, new sorts of ambitious, intensive surveys emerged. Social science was coming of age at the same time as Americans’ sense of themselves as a mass public, and Sarah Igo argues that the new statistics helped shape this national identity.
Igo, who teaches history at the University of Pennsylvania, examines three influential case studies of this new social research. Robert and Helen Lynd lived for many months in Muncie, Indiana, as they scrutinized everything from attendance at women’s clubs to library usage to produce their Middletown studies, published in 1929 and 1937. George Gallup and Elmo Roper began polling the opinions of the American public in 1935. And Alfred Kinsey and his staff conducted thousands of personal interviews with people about their sexual histories to publish reports on the sexual behavior of American men and women, in 1948 and 1953, respectively.
These landmark investigations were widely praised at the time, even as critics noted their flaws. The Lynds excluded African Americans, for example. Gallup and other pollsters wrongly predicted that Dewey would trump Truman in the presidential election of 1948. Prominent statisticians faulted Kinsey’s sampling techniques, and moralists resisted certain of his findings, such as surprisingly high rates of homosexual contact for men and premarital sex for women.
As a historian, Igo is particularly attuned to the changes over time that these studies signaled. She points out, for example, that the Middletown volumes differed from previous case studies in that they were not designed to analyze and solve a social problem. The Lynds’ objective was simply to aggregate detailed information about the lifestyles and preferences of “normal” Americans. As one enthusiastic clergyman told his congregation at the time, “For once we have had the searchlight of social science turned upon a typical American town. . . . We’ve had so many studies of the abnormal. We’ve heard so much about the defective, delinquent, and dependent.”
So much for the Jukes and the Kallikaks. Tell us about people like us—the mainstream. To that end, one of Alfred Kinsey’s most aggressively pursued goals was to expand Americans’ sense of what qualified as “normal” sex. Whatever the defects of his research suggested by later studies and by biographers who have questioned his objectivity, in many respects Kinsey succeeded in this aim, as Miriam Reumann has shown in her fine book American Sexual Character: Sex, Gender, and National Identity in the Kinsey Reports (2005).
Public-opinion pollsters’ methods also represented a break from the past. Instead of conducting intensive community surveys like those that made the Lynds famous, Gallup and Roper developed statistical techniques that permitted a small cross-section of citizens from different regions, classes, and races “to stand in for the whole. Their scope was national rather than local, their subjects no longer rooted in a specific, if generalizable, geographic place.”
By the mid-20th century, Igo says, a large portion of the American public liked and trusted what social science could tell them. Although her book does not suffer from a lack of context, one might have hoped for still more to substantiate this claim. Some of the Lynds’ best-known findings, for example, weren’t exactly revelations. They may have been struck by the “pecuniary civilization” in Middletown, but Americans’ penchant for commercial opportunism had been a central theme of Alexis de Tocqueville’s Democracy in America a century earlier. Nor was ethnographic research new to the public; National Geographic had been publishing middlebrow ethnographies for decades.
Based on a thorough reading of the voluminous fan mail the investigators received as well as extensive reviews and feature stories about them and their work, Igo makes a persuasive case that all three sorts of surveys enhanced Americans’ understanding of who they were. The most striking thing during the debates about Kinsey’s reports, Igo concludes, “was the fact that Americans were more eager than ever before to become research subjects—ready to conceive of themselves as case histories in an aggregate bank of survey data.”
Of course, we have no idea how many of these eager subjects were exhibitionists and how many others refused to participate in polls or interviews of various kinds. As we learn from counted,” numerous others mistrusted the conventional sample size of 1,500 and still others remained skeptical of the process of sampling itself. W. H. Auden’s admonition to the 1946 graduating class at Harvard was symptomatic of widespread doubts in the country: “Thou shalt not sit/ With statisticians nor commit/ A social science.”
One of Igo’s major conclusions is that “modern survey methods helped to forge a mass public.” Americans could now learn what Mary and John Q. Public liked and disliked, and consequently gain an enlarged sense of the diverse views held by a rapidly expanding populace. I am inclined toward a different interpretation. A great deal of scholarship has been produced that suggests that coming to terms with and interpreting an increasingly vast and impersonal public required modern survey methods—whether the exhaustive analysis of a community or the labor-intensive process of conducting interviews with thousands of people from all walks of life.
The Lynds may very well have been hoping to identify the mores of average (white) Americans, but the point of polls by Gallup and Roper and interviews by Kinsey and his staff was to delineate a range of differences in beliefs and practices. What comes across in The Averaged American is not a series of medians and means but patterns of segmentation and divergence. The diversity of 1960s and ’70s America that Igo notes in her epilogue was not new—it was only more pronounced and visible than it had been a generation earlier.
That is not to say that Igo’s notion of “averaged” Americans isn’t valid, but perhaps it applies to a different body of literature than the important but particular works she cites. Especially during the 1950s, survey-based books and articles appeared that defined the average American family as a nuclear unit with 2.5 children, or told readers that persons of a certain height should weigh between 115 and 125 pounds. Americans who did not match the newly revealed norms (or averages) for cars and television sets per family may very well have felt anxiety about their aberrations. But the work of Gallup, Roper, and, especially, Kinsey argues against the grain of “averaged Americans.” However one feels about multiculturalism as an American mantra, diversity has been with us for quite some while.