In his discussion of Kinsey, for example, Serwatka writes:
Unfortunately, despite some of the strengths of his surveys and interviews, Kinsey’s sampling techniques left much to be desired and have justifiably [27] generated a number of criticisms about the accuracy of his numbers. The male participants Kinsey and his colleagues used in their study were predominantly white, they voluntarily signed up to be in a study on sexual attitudes and activities, and they included a higher percentage of college students than would be expected as well as an overrepresentation of volunteers who had been in the prison system.It takes a certain talent to pack so much misinformation into one sentence. One at a time:
The male participants Kinsey and his colleagues used in their study were predominantly white ...In fact the participants whose sexual histories were analyzed in Sexual Behavior in the Human Male were all white. Kinsey interviewed African-Americans, but not enough for representative samples, so their histories were not used in the book. Besides, American society is predominantly white and was even more so in the 1930s and 1940s when Kinsey was collecting data, so the question for a representative sample would be whether blacks were underrepresented.
... they voluntarily signed up to be in a study on sexual attitudes and activities ...I might read this clause more charitably and less critically if Wersatka hadn't already shown that he's misinformed about Kinsey's work. True, the informants participated voluntarily, in that they weren't compelled, but that is true of the participants in other, later surveys that Wersatka cites uncritically. It's hard to imagine how you'd get participants involuntarily.
... and they included a higher percentage of college students than would be expected as well as an overrepresentation of volunteers who had been in the prison system.True, Kinsey began by getting the histories of college students from the Marriage class he was teaching. But he knew he needed a wider range of subjects, and traveled the country in quest of them. He buttonholed just about everyone he met, and gave lectures which concluded with an appeal for 100% participation by the audiences -- which he often got, often in churches and American Legion chapters.
The complaint about too many convicts in the sample is a perennial, presumably based on the assumption that too many prisoners would stuff the ballot box, as it were, with hot man-to-man action. The trouble is that it isn't true, and has been known not to be true for a long time. As Martin Duberman wrote in a review of a 1997 biography of Kinsey:
... Paul Gebhard (one of Kinsey's co-authors and his successor as director of the Kinsey Institute for Sex Research -- he retired in 1982), himself reacting to criticism leveled against the two volumes, spent years "cleaning" the Kinsey data of its purported contaminants -- removing, for example, all material derived from prison populations in the basic sample. In 1979, Gebhard, with Alan Johnson, published The Kinsey Data, and -- to his own surprise -- found that Kinsey's original estimates held: Instead of Kinsey's 37 percent, Gebhard and Johnson came up with 36.4 percent; the 10 percent figure (with prison inmates excluded) came to 9.9 percent for white, college-educated males and 12.7 percent for those with less education.So, the numbers dropped by less than one percent. Duberman added:
And as for the call for a "random sample," a team of independent statisticians studying Kinsey's procedures had concluded as far back as 1953 that the unique problems inherent in sex research precluded the possibility of obtaining a true random sample, and that Kinsey's interviewing technique had been "extraordinarily skillful." They characterized Kinsey's work overall as "a monumental endeavor."Kinsey's original goal was 100,000 histories; he eventually got 18,000, and used about 12,000 of them in the two Sexual Behavior volumes. I get the impression that Serwatka believes Kinsey used every history he took, but Kinsey knew better than that, had a statistician on staff, and built his final samples with care. The result was imperfect, of course, but not for the reasons Serwatka claims.
So, what's going on here? I've gotten as far as Serwatka's discussion of "genetic" influences on sexual orientation, and he presents the contemporary debate about the issue as one between raving queer theorists and fundamentalist Christians on one side, and thoughtful scientists on the other. But he's got a good many of his facts wrong (I may write more about this in a day or two), and the gay and lesbian scientists who've criticized the "born gay" research on scientific grounds are conspicuously absent from his bibliography. I intend to try to finish the book, but Serwatka's hearty, preacher's chalk-talk style is painful to read. I'll soldier on, however.