Anti-gay Harvest pastors’ bizarre iPhone analogy

Pastors Jason Powell & Chad Williams from Greg Laurie’s Harvest Christian Fellowship, California

Advertisements

2 thoughts on “Anti-gay Harvest pastors’ bizarre iPhone analogy

  1. From HERE

    All the data I’ve seen supporting the hyper-promiscuous median gay male claim has been junk science. It often refers to real studies — but to studies of groups that we have no reason to think are representative of the median gay male. The Wells-Petry Exclusion book, for instance, mostly relies on studies of gay males who are being treated for sexually transmitted diseases; there’s lots of reason to think that those people have many more sexual partners than the median gay male. The Masters, Johnson & Kolodny college textbook relies on a study that (1) was limited to the San Francisco Bay area, and (2) involved a self-selected sample, not a randomly chosen sample; Schmidt’s Straight & Narrow, a book that criticizes homosexuality from a Christian perspective likewise heavily relies on this study, as well as on another survey (Jay & Young) of self-selected respondents. (Schmidt also cites other articles, which I’m getting the library to pull — if I find that some of the data there is of higher quality, I’ll certainly report it.)

    I actually think this would make an interesting story for mainstream journalists to cover (if they haven’t already). I suspect that many thoughtful, generally knowledgeable people believe the myth; I know that I had, based on some stuff that I recall vaguely hearing from the mainstream media in the 1980s. The myth has been spread both by people that oppose homosexuality, and by people who seem to have no such agenda (see Masters, Johnson & Kolodny). As I argue above, the myth is, I think, politically quite salient. But it does appear to be a myth.

    [Eugene Volokh, 3:07 PM]
    STRAIGHT & NARROW: A fellow academic whose work I generally much admire also e-mailed me about the hyper-promiscious median gay male theory, and pointed me both to Bell & Weinberg’s Homosexualities and a book by Thomas Schmidt called Straight & Narrow (1995). The Schmidt book is a criticism of homosexuality from an evangelical Christian perspective, and chapter 6 purports to be a summary of a wide range of statistical evidence.

    The evidence that I checked, though, seems to be highly unsound. For instance, here’s its core claim about the number of sexual partners among homosexuals:
    We can quantify the phenomenon of homosexual promiscuity, especially among males, more specifically. The numbers are astounding. Bell and Weinberg found that 74 percent of male homosexuals reported having more than one hundred partners during their lifetime, 41 percent more than five hundred partners, 28 percent more than one thousand partners. . . .
    Note that the claim is about homosexuals generally — “homosexual promiscuity,” “found that 74 percent of male homosexuals,” and such.

    What Schmidt doesn’t mentioned is that, as I quoted in my earlier post, Bell & Weinberg specifically said:
    It should be pointed out that reaching any consensus about the exact number of homosexual men or women exhibiting this or that characteristic is not an aim of the present study. The nonrepresentative nature of other investigators’ samples as well as of our own precludes any generalization about the incidence of a particular phenomenon even to persons living in the locale where the interviews were conducted, much less homosexuals in general.
    And why do Bell & Weinberg warn precisely against what Schmidt is doing — “generaliz[ing] about the incidence of a particular phenomenon . . . [among] homosexuals in general”? Because Bell & Weinberg was (1) a survey only of gays living in the San Francisco Bay area, and (2) wasn’t even a random sample, but was rather a self-selected sample, drawn from those gays who heard about the study through various means — “such as public advertising, bars, personal contacts, gay baths, organizations, mailing lists, and public places” — and who then chose to respond. Statisticians routinely warn against the huge unreliability of such self-selected samples.

    This isn’t mentioned, however, in Schmidt’s discussion of Bell & Weinberg. Likewise, the one other study that I have so far checked of those Schmidt relies on (I plan to ask the library for the other studies), a “large survey conducted by K. Jay and A. Young,” was likewise a self-selected survey, though a national one. Schmidt claims that
    Another large survey conducted by K. Jay and A. Young found that only 7 percent of male homosexuals had been in a relationship that had lasted more than ten years; 38 percent had never been in a relationship that lasted more than one year; 55 percent had never been in a relationship that lasted more than two years. (p. 107)
    But the first page of the Kay & Young chapter from Schmidt draws his data says “As with all the statistics, it is difficult to know if the subsample suggests an accurate figure for the gay male population generally.” You bet it is, because the sample is self-selected — but Schmidt doesn’t seem to be detained by this, doesn’t acknowledge the survey’s limitations, and says that the survey “found” information about “male homosexuals,” with no qualifiers or caveats. (By way of comparison, the Hite Report on Female Sexuality reported that over 70% of women who were married 5 or more years had been unfaithful; but to my knowledge this number is now rejected, precisely because the Hite Report involved a self-selected, rather than random, sample of respondents.)

    So, as usual, don’t believe everything you read — especially on this subject, apparently.

    Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s