|
Women in Academe:
Two Steps Forward, One Step Back
|
|
IV. Obstacles to Women's Success in Academe
Academe is not yet a level playing field. When women psychology faculty (PhDs and MAs) were asked in the NCES (1993) national survey how much they agreed with the statement, "Female faculty members are treated fairly at this institution," 40% disagreed, 11% of them strongly. Their male colleagues perceived less inequity, however. Only 15% of male faculty disagreed, 4% strongly. Although bias and stereotyping are intensified for women who are in the minority in their employment setting, the gender of one's evaluators and general societal perceptions of the gender appropriateness of the specific occupation also play a role. Thus, despite the many advances made by women in academic psychology, they are still often in the minority (e.g., among full professors), often evaluated by a group made up mostly of men (e.g., for promotion and tenure), and sometimes engaged in a field (e.g., psychophysiology) or activity (e.g., research) seen as more appropriate for males. Moreover, since leadership itself has traditionally been viewed as a masculine endeavor, women leaders in intellectual endeavors or administrative positions may confront difficulties that men do not experience.
Gender discrimination is sometimes delivered as a "knockout blow" to one's professional advancement,as Ann Hopkins discovered when she was denied a partnership by Price Waterhouse (Fiske, Bersoff, Borgida, Deaux, & Heilman, 1991). However, it probably more often occurs as a long series of "microinequities" (Rowe, 1990) or "unintended slights" (Association of American Medical Colleges [AAMC], 1996) that create a more stressful, less rewarding environment for women. For most women, it is the disadvantage that accrues over time that is most harmful to their professional achievement and their quality of life at work (Bickel, 1995; Valian, 1998). We begin this section with a discussion of issues related to stereotyping and bias, followed by a description of the possible effects of discriminatory behavior on women's access to opportunities in research, teaching, service, and leadership.
Difficulties of Difference
In an important set of meta-analyses, Eagly and her colleagues (Eagly & Johnson, 1990; Eagly, Karau, & Makhijani, 1995; Eagly, Makhijani, & Klonsky, 1992) examined the relationship between gender and leadership in a wide variety of settings, academic and otherwise. Taken as a whole, their findings clearly illuminate the gender fault lines lying just beneath an apparently benign surface. Overall, perceived effectiveness did not differ for the male and female leaders included in the studies analyzed by Eagly et al. (1995), and evaluations of female leaders were only slightly more negative than those of male leaders (Eagly et al., 1992). However, a more detailed examination of these data paints a more disturbing picture.
Women in settings where there were many more men than women were evaluated more negatively and perceived as less effective than were men (Eagly et al., 1995; Eagly et al., 1992). Women were also evaluated more negatively by male raters (Eagly et al., 1992), who, of course, predominate in groups with a preponderance of men. Interestingly, the leadership style that women displayed was related to the proportion of women in the group. Although women were generally more likely than men to engage in a democratic, participatory style of leadership, this difference was weakened in male-dominated settings (Eagly & Johnson, 1990).
Thus, when women were in the minority, they were more likely than women in the majority to be less democratic and more autocratic. This difference in style is not without consequences:An autocratic style by a female leader elicited strongly negative evaluations. Men, however, were not devalued for adopting the more female-preferred democratic leadership style (Eagly et al., 1992).
In addition, Eagly and her colleagues found that women and men in roles viewed by others as unsuitable for their gender were less task-oriented than those in more gender-congenial roles (Eagly & Johnson, 1990) and were perceived by others as less effective leaders (Eagly et al., 1995). Since perceptions of gender appropriateness are correlated with actual gender ratios, these two factors tend to be mutually reinforcing.
The meta-analyses conducted by Eagly and her colleagues included laboratory research as well as organizational studies, with participants ranging from undergraduate and graduate students to managers and supervisors in both private and public settings. This wide variation suggests that these findings would apply to women in many different occupations, including faculty as well as administrators in colleges and universities. We will now examine the considerable body of research that supports and expands upon the themes identified by these meta-analytic studies.
Process and Effects of Bias and Discrimination
In a recent study of gender discrimination at the Massachusetts Institute of Technology (MIT),the senior women faculty asked themselves why they had been so slow to recognize inequities at that institution. They concluded, "It did not look like what we thought discrimination looked like" (MIT, 1999, p. 9). In other words, it is difficult to see discrimination in a single instance. Any specific event can be explained away--by others and, sometimes, by oneself--as the consequence of "special circumstances." Only when the women at MIT began to share their experiences with each other did the general pattern emerge.
The experience of these women--each of whom had already shattered many a glass ceiling to achieve a tenured position at MIT--vividly attests to the fact that although overt discrimination is out of style and indeed legally actionable if applied to hiring and other personnel decisions, covert and subtle forms of discrimination continue and can have a great impact on women's lives. Modern sexism and discrimination include (a) denial that discrimination against women exists, (b) resentment of complaints about discrimination, and (c) resentment of what are seen as special "favors" for women (Swim & Cohen, 1997). Addressing and ameliorating the subtle effects of today's sexism and its expression in the form of discrimination--which the MIT report characterizes as "powerful but unrecognized assumptions and attitudes that work systematically against women . . . even in the light of obvious good will" (p. 9)--requires a sophisticated understanding of the processes of sexism, stereotyping, and discrimination.
Indeed, considerable research has defined the characteristics of contemporary sexism (e.g., Biernat & Kobrynowicz, 1997; Glick & Fiske, 1996; MacDonald & Zanna, 1998; Ruggiero & Taylor, 1995; Swim, Aikin, Hall, & Hunter, 1995; Tougas, Brown, Beaton, & Joly, 1995). A brief look at what this research base tells us about the range and strength of discrimination against women in the workplace will help place in context discrimination within academe.
Note, for example, how subtle the process of gender discrimination can be. The elicitation of gender stereotypes is "strong and ubiquitous" (Blair & Banaji, 1996) and can take place as an unconscious, automatic process that seems to be independent of either overt sexism or the gender of the perceiver (Banaji & Greenwald, 1995a, 1995b; Spencer, Fein, Wolfe, Fong, & Dunn, 1998). Bias against women's contributions to a group discussion can be expressed by relatively unobtrusive nonverbal responses such as frowns and negative head shaking (Butler & Geis, 1990). In ambiguous situations where the power and status of the participants in a mixed-sex dyad are not clear, nonverbal behaviors by both males and females reinforce the stereotyped presumption of male, rather than female, expertise (Dovidio, Keating, Heltman, Ellyson, & Brown, 1988).
Even the type of rating scale used can affect the evaluations women and men receive (Biernat, Crandall, Young, Kobrynowicz, & Halpin, 1998). Stereotypic differences tend to be smaller on "subjective scales" (such as ratings) than on "objective scales" (such as rankings). Why? Because evaluators can select different comparison groups when rating (e.g., comparing a woman against other women and a man against other men), but have to use the same standard for all of those being rank-ordered in one set. Samuel Johnson had this difference in mind when he made his infamous remark, "Sir, a woman preaching is like a dog's walking on his hind legs. It is not done well; but you are surprised to find it done at all." Theory and research on shifting standards indicate that evaluations of women are fraught with difficulty: A high mark on a subjective scale may be patronizing in the Johnsonian mode and misleading about the evaluator's actual views; a low mark on an objective scale may reflect discrimination and bias rather than actual performance differences. Although there is some evidence that the elicitation of gender stereotypes can be modified by various cognitive factors (such astime available for the task) and the perceiver's intention (Blair & Banaji, 1996), these stereotypes are also quite tenacious. For instance, despite the availability of job-relevant, individuating information that logically should be more important than gender, gender stereotypical hiring evaluations (i.e., preferring men for "masculine" jobs and women for "feminine" ones) still persist (Davison & Burke, in press; Glick, Zion, & Nelson, 1988).
Attitudes toward women's roles have become relatively more egalitarian (Spence & Hahn, 1997; Twenge, 1997), but the long-term outcome of even small degrees of bias and discrimination is not inconsequential. As Martell, Lane, and Emrich (1996) point out, most organizations have fewer positions at the top than at the entry level, and strong performance evaluations are usually necessary for promotion. According to their computer simulation of a company with these characteristics and in which men and women were equally represented in entry positions, pro-male bias accounting for only 1% of the variation in their initial evaluation would reduce the percentage of women at the top level from 50% to 35%.
Complexities of Self-Presentation
In a discussion of the relationship between self-presentation and gender, Rudman (1998) emphasizes how, to a much greater extent than men do, women "are obliged to make decisions every day regarding how to present themselves" (p. 642). Unfortunately, recent research suggests that no matter what presentation is selected, it may still be particularly difficult for women to achieve the apparently reasonable goal of being both respected and liked by both men and other women (Fiske, 1993).
Several research reports published over the last decade have described some of the barriers that women can encounter. For example, women often use "tentative speech," and this affects perceptions of them in the workplace. Carli (1990) defined tentative speech as the frequent use of disclaimers (I'm no expert, I suppose, I may be wrong), hedges (kind of, you know, maybe), and tag questions (Isn't it? Don't you think? Right?). Women who employed more direct language--which lacked such qualifiers--were perceived as more knowledgeable and competent than women who used tentative language. However, women who used direct language were less influential with men than women who used tentative language; men also found the direct female speaker to be less trustworthy and likeable. In contrast, female participants were more influenced by the direct speaking woman,and perceived her as more trustworthy and likeable. The language style of male speakers affected neither their ability to influence others nor the evaluations they received.
Investigating the role of nonverbal behaviors, Carli, LaFleur, and Loeber (1995) found that men reacted negatively to a woman with a "high task style"--defined as "a rapid rate of speech, a firm tone of voice, moderate voice volume, few hesitations, an upright posture, calm hand gestures, and a moderately high amount of eye contact" (p. 1034). Compared to a man with the same style, the high task woman was perceived by male participants as less likeable and more threatening. In addition, the high task woman had less influence on men's opinions than did the high task male and less than a woman with a "social" presentation style (which included a more pleading tone of voice and a friendlier facial expression than the high task presentation style). More generally, being liked by male participants played a greater role in how much influence female speakers exerted than it did for male speakers. Nonverbal styles and gender had relatively little effect on the responses of female participants.
Rudman's (1998) research examined both speech styles and nonverbal behaviors in combination. In these studies, both male and female participants perceived self-promoting women and men (who spoke self-confidently about their past accomplishments and made direct eye contact) as more competent than those who were self-effacing (who spoke more modestly and tentatively and avoided direct eye contact). However, self-promoting women were liked more than self-effacing women only by men who expected to work with that particular woman on an interdependent task. A similar pattern prevailed for ratings of the likelihood of being hired for employment. Across a variety of conditions, female subjects never liked or wanted to hire a self-promoting woman more than a self-effacing woman, though they did like and want to hire a self-promoting man more than a self-effacing man.
Taken together, these studies demonstrate the persistent difficulty that women have in obtaining both respect and liking from both men and women. For the most part, adopting a more "masculine" style (direct language, high task nonverbal behaviors, self-promoting) did increase the perceived competence of both men and women. But unlike competent men, women perceived as competent were often not liked--sometimes by men (Carli, 1990; Carli et al., 1995), sometimes by women (Rudman, 1998). Moreover, the usual benefits of perceived competency, such as being seen as suitable for hiring and influencing people's opinions, did not necessarily occur for competent women. Indeed, competent women were sometimes viewed as less suitable for hiring and less influential than their male peers and/or a female with a more "feminine" style who was perceived as less competent. These studies strongly suggest that, on average, life in the workplace presents complicated challenges to women that men do not face. At least for men in the dominant racial, ethnic, and socioeconomic class, there is usually some basic harmony among competence, respect, liking, influence, and career success. For women, there is still considerable dissonance.
Behavioral Freedom: Perceptions by Others and by Oneself
As mentioned earlier, the meta-analysis by Eagly and her colleagues (1992) on gender and the evaluation of leaders indicated that while women were devalued for adopting stereotypically masculine leadership styles (e.g., autocratic), men were not devalued for adopting stereotypically feminine leadership styles (e.g., democratic). Based on these findings, Eagly et al. concluded that "It appears that all other factors being equal, men may have greater freedom than women to lead in a range of styles without encountering negative reactions" (p. 16). The research that we have just reviewed on the effects of self-presentation is also consistent with this proposition. In women's lives, personal and professional, this more narrow range of "acceptable" behaviors creates both the sense, and sometimes the reality, that they are walking a tightrope, with harmful consequences awaiting any misstep.
Perceptions of the range of behavioral freedoms available to women can have other important effects. Beyer and Bowden (1997) documented the disadvantages for women engaged in tasks that are not stereotypically perceived as appropriate for females. In this study, both males and females who engaged in gender-incongruent tasks had lower expectations, performance, and self-evaluations than did those who engaged in gender-congruent tasks. However, men were more positive and successful than women on the gender-neutral task. There was also a specific disadvantage for women on the masculine task. Accuracy of performance estimates did not differ between men and women on either the feminine or neutral tasks, but female participants significantly underestimated their success on the masculine task.
For women, then, the stereotypically feminine task was the only one on which they had positive expectations, performed well, and could provide an accurate estimate of their actual success. Men, on the other hand, had positive expectations and good performance on the neutral as well as the masculine task and never underestimated their accuracy on any of the three tasks. Reflecting on these results, Beyer and Bowden (1997) commented that the psychological processes displayed here could have widespread and long-lasting effects, affecting "the kinds of tasks, courses, careers, and so forth that one chooses, and one's persistence and performance in those areas" (p. 169).
Steele's (1997) work on stereotype threat suggests that the impact of stereotypes on performance has a long reach: Self-relevant negative stereotypes can still affect even those who have not internalized the inferiority feelings and low expectations of success generated by the stereotype. Those who are vulnerable to this kind of situational stereotype threat are those who have achieved, despite the stereotype, in the very domains to which the stereotype applies (e.g., African Americans and individuals from a low socioeconomic background who achieve in school; women who achieve in math and science). Since their definition of self rests, in part, upon their achievements in these domains, having others believe and/or their own behavior confirm the stereotype would threaten an important part of their identity. Situations emphasizing the existence of such stereotypes produce impaired performance on stereotype-relevant tasks and prompt avoidance of identification with the group stigmatized by the stereotype (Brown & Josephs, 1999; Croizet & Claire, 1998; Steele, 1997; Steele & Aronson, 1995).
Although no research has yet been published on the relationship among stereotype threat, gender, and faculty behavior, the potential connections seem clear. The social psychological position of women psychology faculty often bears a close resemblance to that of other individuals whose achievements occur in domains for which they are stereotypically viewed as unqualified. They are minorities, easily identified as different from the majority, subject to bias and discrimination, confronted with complicated issues of effectiveness in self-presentation, and required to walk a narrow behavioral line in order to succeed. If this analysis is correct, conditions that highlight stereotyped beliefs would make it harder for these women to perform as well as they can under more benign conditions. The concept of stereotype threat also helps explain the unrelenting pressure to prove themselves experienced by high achievers in an unlikely domain: "the work of dispelling stereotype threat through [the quality of one's] performance probably increases with the difficulty of work in the domain, and whatever exemption is gained has to be rewon at the next new proving ground" (Steele, 1997, p. 618).
We turn now to an examination of the obstacles, and the opportunities, that female academic psychologists face across various academic roles.
Previous Page
Table of Contents
Next Page
|
