Columnists are still rehashing Maureen Dowd's Oct. 30 New York Times column "What's a Modern Girl to Do?"
There, Dowd raised a question that plays to the fears of ambitious women everywhere: do men prefer women who keep their opinions and independence to themselves?
Dowd answered with a resounding "yes!" but the data she leans upon doesn't provide much support.
Dowd's column may be best viewed as a promo for her new book Are Men Necessary? (list price $25.95). But intense discussion -- described by some as a "mediagasm" -- has ensued, only because she captures a valid social concern: "[W]as the feminist movement some sort of cruel hoax? Do women get less desirable as they get more successful?"
Despite wide-ranging discussion, no one seems to question or even examine Dowd's supporting data.
She begins with a hat tip toward fellow-NYT journalist John Schwartz, who she claims "made the trend official." In a Dec. 14, 2004 piece titled "Glass Ceiling at Altar as Well as Bedroom," Schwartz wrote, "Men would rather marry their secretaries than their bosses, and evolution may be to blame."
Schwartz cited the same "study" as Dowd: "Relational dominance and mate-selection criteria: Evidence that males attend to female dominance," conducted by Stephanie L. Brown and Brian P. Lewis, and published in the journal of Evolution and Human Behavior in late 2004.
One hundred and twenty male and 208 female undergraduate students "at a large university in the southwestern United States" participated. The "study" states that the students "were randomly assigned to one of six experimental conditions" but does not indicate whether they were randomly selected from the student body as a whole.
Thus, from the beginning, there is a glaring problem with the data.
The population studied (120 males) is too small to be significant. Moreover, the population is unrepresentative of males in general. Indeed, it may not even be representative of the narrow subcategory "undergraduate males on one campus." The "study" provides no way to judge.
The 120 males were asked to rank their attraction to the photograph of a female who was described as either "their supervisor (higher dominance), coworker (equal dominance), or their assistant (lower dominance)." In short, they were asked to speculate about their probable response to a hypothetical situation.
This is another glaring problem with the resulting data. The "study" measures highly subjective and easily influenced reactions. The "easily influenced" aspect becomes important as the "study" clearly states that the researchers anticipated a specific result.
It states, "The following prediction was tested: Males were expected to be most attracted to the female target when she was described as a subordinate, and we expected a male preference for the subordinate target to be most pronounced when males were considering a long-term (high investment) relationship with the target."
This is called "bias." It influences everything from how a researcher phrases questions -- both in terms of language and tone -- to the unspoken cues that communicate what is "expected" of them to the participants. Bias invalidates results, especially when purely theoretical responses are being measured. Not surprisingly, the "study" found "results supported predictions."
What questions led to such predictable data? Three different questions were presented along with a photograph; each question placed the participant in a different power dynamic with the person pictured.
The "higher dominance" scenario was, "Please imagine that you have just taken a job and that Jennifer...is your immediate supervisor. She...is the person you report to on a daily basis. She/he has the responsibility for disciplining absence or poor performance on your part, for rewarding reliable or creative performance..."
Again, not surprisingly, most men did not feel comfortable with becoming romantically involved with someone who could crush their careers should romance turn sour. They did not reject an attraction to powerful women per se; they rejected an attraction to women who wielded great power over their careers. The two categories are not remotely comparable.
The fact that women responded in a more neutral manner is interesting.
But what does it indicate? They may have simply responded to the researchers' expectations. Or, as a general statement, women may be less career-oriented than men and, so, more willing to take risks that could damage them professionally. Even if the data were valid, there are several plausible explanations of what it suggests.
The "study" entertains only one explanation -- an explanation it had predicted.
"First, our findings suggest that, under some circumstances, males are concerned with the dominance of a female when choosing a mate. Second, these results highlight the importance of relational dominance in mate selection."
In short, the current brouhaha surrounding Dowd's column is based on an insignificantly small and unrepresentative sample, processed by biased researchers who attached their own pre-determined explanation to results that were never valid.
Indeed, I have bracketed the word "study" with quote marks throughout this column because the word "survey" seems more appropriate.
Students were asked to fill out a questionnaire. The study wraps itself in hyper-technical language. For example, "Before examining the effects of relative dominance on romantic attraction, a 2 (Participant Gender) 2 (Target Gender) 3 (Dominance) ANOVA was conducted on the manipulation check. As expected, there was a main effect of dominance level [F(2,315) = 59.52, p b .001] and no interaction of dominance level with target or participant gender."
But the bottom line is that students were asked to fill out a questionnaire.
Such is the substance (or lack thereof) behind Dowd's explosive contention. As I said, her article is best viewed as a promo for her book.
Copyright © 2005 Wendy McElroy.