Why haven’t more voters defected to Hillary? Research in social psychology suggests an answer.
“Don’t you think a man who has this kind of economic genius is a lot better for the United States than a woman?” This semi-rhetorical question, posed by Rudy Giuliani earlier this month, met with widespread disapproval. But if we spot Giuliani a true belief about Donald Trump’s kind of economic genius – none – then his self-awareness would be impressive. Many people do think along these lines, if only unwittingly, according to recent research in social psychology.
We often conceptualize gender biases as something like demerit points applied to individual women, who we subsequently assess more negatively than we would do otherwise, in the domains in which we’re biased. We tend to underestimate her, making it harder for her to compete with a male counterpart, who we are disposed to assess fairly. She might have to be twice as good to beat him, say, in an extreme case.
But there’s a distinct way of conceptualizing gender biases as applying to our rankings of men and women, and disposing us to prefer a man to his female counterpart – by, e.g., supporting, promoting, liking, believing, or voting for him over her. This may result in overestimating him, and turning against her with hostility, as well as underestimating her merits. Regardless of how good she is, we may find something to doubt or dislike about her – again, in extreme cases.
Unfortunately, one such may be that in which we currently find ourselves – where a man and a woman are competing for a position of historically male-dominated authority. The following empirical evidence, yet to be cited in this connection, sheds light on the question many people have been asking lately: how is someone like Trump still receiving nearly 40 per cent of the vote, even as he continues spectacularly flaming out? Why haven’t more voters defected to Clinton? An important part of the answer is suggested by these three studies – three of many evincing a strong disposition to uphold gendered social hierarchies.
1. Researchers David Paul and Jessi Smith surveyed nearly 500 likely voters in Ohio, some two years prior to the 2008 election. Respondents were asked to assess five likely presidential candidates – three Republicans, two Democrats, and three men and two women. Voters rated the two women as the least qualified of the five, despite their both being highly qualified, objectively speaking (according to the researchers). And in each of the six head-to-head matchups that might result in the general, the male candidate beat the female one in every single contest.
Each of the men also did better against a female nominee, as opposed to another man, from the opposing party. Perhaps most tellingly of all, voters were several orders of magnitude likelier to defect from a female nominee from their own party to a male nominee from the opposing one, compared with vice versa. The researchers concluded that “the presence of a woman candidate opponent for president may aid the [male] competition.”
The three male politicians in the survey were John McCain, John Edwards, and Rudy Giuliani. One of the female politicians was Elizabeth Dole. The other woman? Hillary Clinton.
But wait, you might say, in a somewhat hopeful tone. Hillary has much more experience now than she did back then. If her qualifications were no longer in doubt, due to women in her position admittedly being held to more stringent (i.e., double) standards, might such gender biases cease to be an issue?
No. As the following research reveals, this is highly unlikely at the level of the general population. When women are not doubted as viable competitors for male-dominated roles, they are widely disliked and subject to social punishment and rejection.
2. The psychologist Madeline Heilman has a body of research geared toward answering the question: when the evidence of a women’s competence in male-dominated fields is unequivocal, are they still subject to gender discrimination? It would appear so; highly successful female managers are still promoted at rates much lower than their male counterparts. Why is this?
In one particularly striking study, Heilman and her collaborators gave participants packets of information profiling two high-status employees in a male-dominated industry: assistant vice presidents (AVPs) for sales in a company manufacturing airline parts. One of the employees was male, the other female – as signalled by their names, “James” and “Andrea.” And which information packet was assigned to which of the pair was rotated, so each of the employee profiles was associated with James for half the participants, and Andrea for the remaining half, within each of the two conditions.
In the “unclear success” condition, the evidence of James and Andrea both being outstanding AVPs was equivocal. For this group, the vast majority of participants (86%) judged James more competent than Andrea when asked to compare them. But the two of them came off as similarly likable.
In the “clear success” condition, the remaining half of the participants received an additional piece of information that made each employee’s success unambiguous: an annual performance review stating that each was a “stellar performer,” in the top 5% of all AVPs at such companies. In this case, participants judged the two about equally competent, but James as more likable than Andrea, in the vast majority of cases (83%). Andrea was also then held to be more interpersonally hostile than in any other condition – a measure which encompassed being abrasive, manipulative, conniving, and untrustworthy, notably. Heilman et. al. describe this effect as “dramatic.”
Remember, this was based on identical (alternating packets of) information. So these radically disparate judgments had no rational basis whatsoever. Insofar as the participants felt themselves to have reasons, they must have employed ad hoc standards and post hoc rationalization.
Why, though? Why did they so dislike Andrea, when her competence was undeniable?
3. The psychologist L.A. Rudman has proposed the following answer: people are (often unwittingly) motivated to maintain gender hierarchies, by applying social penalties to women who compete for, or otherwise threaten to advance to, high-status, masculine-coded positions. This “status incongruity hypothesis” is consistent with the results of the previous study, and helps to explain them. In a 2012 study citing it, among other work by Heilman, Rudman and her collaborators showed that the effect is mediated by what is known as the “social dominance penalty,” where women in such positions who are agentic (i.e., competent, confident, assertive) are perceived as extreme in masculine-coded traits like being arrogant and aggressive. They are often described as “ballbreakers” and “castrating bitches.” (Sound familiar?)
These also happen to be verboten traits for women (as Rudman et. al. confirmed experimentally). So agentic women competing with men for male-dominated roles are doubly likely to be punished and rejected in light of these mechanisms. They are perceived as having more of the qualities they are less permitted to have than their identically-described male counterparts (described, again, using the same textual stimuli, such as letters of recommendation with the names switched).
The explanation Rudman offers for the social dominance penalty was further confirmed by the following, fascinating result: it could be augmented under a “high threat” condition where, at the beginning of the experiment, participants read an article called “America in Decline,” which included the following paragraph:
These days, many people in the United States feel disappointed with the nation’s condition. Whether it stems from the economic meltdown and persistent high rates of unemployment, fatigue from fighting protracted wars in the Middle East that have cost America dearly in blood and treasure, or general anxieties regarding global and technological changes that the government seems unable to leverage to their advantage, Americans are deeply dissatisfied. Many citizens feel that the country has reached a low point in terms of social, economic, and political factors.
In this condition, agentic women who aspired to high-powered positions were significantly less liked and more often turned down for a promotion than in the “low threat” and control conditions. For agentic men, the threatening stimulus made no difference. The researchers explained:
Given that people under system threat tend to defend their worldviews, which include gender status differences… and because female agency was especially rejected by people under system threat, [these results] provide direct evidence that backlash functions to preserve the gender hierarchy.
This makes the fact that Donald Trump and even, to a certain extent, Bernie Sanders, had an unexpectedly good run against Hillary less surprising, even beyond the “boost” that Paul and Smith suggested male politicians will receive, when they compete with a woman for the presidency.
Note too that the common assumption that millennials are more or less immune to these effects, given their tendency to identify as progressive, seems dubious at best. The second and third of the above studies were done exclusively on people (still) in this age group (their being undergraduates at the time). The first included them. And their age appears to have made no difference. Nor, in any of these studies, did participants’ gender.
Donald Trump has advertised himself as the law and order candidate. And in many ways that’s accurate. But the law and order is not that of the land but rather, the gender hierarchies of which he’s long been the beneficiary, and is perpetually enforcing, by humiliating the women who challenge his authority. When Trump said that Hillary’s gender was all she had going for her, it was yet another exercise in unwitting projection on his part. If he were not a (white) man, he wouldn’t have stood the faintest chance against her. But many Americans aren’t ready for a woman like Clinton to trounce a man like Trump for the White House.