Gender Bias and the Faculty Gender Gap (2005)

Gender Bias and the Faculty Gender Gap (2005)

Implications and Recommendations for the College of Arts and Sciences

It is well understood that women faculty in American higher education lag behind their male peers in a myriad of ways, ranging from their underrepresentation at the most elite universities, to underrepresentation at the highest faculty ranks, to lower salaries beyond the entry level (see Curtis 2004; Ernst 1998; Valian 2004). A 2002 study shows that the experience of faculty women at Indiana University-Bloomington basically fits these national patterns. For example, IUB female faculty are underrepresented at the highest faculty ranks and within the ranks of academic administration, are notably underrepresented in certain disciplines and fields, and have lower rates of progress toward promotion to full professor, once tenured (even though they receive tenure at the same rate as men). Further, despite the significant salary adjustments that followed from the campus salary equity review begun in 1997, as of 2002 female faculty within each rank earned less than their male counterparts, and few women were found among the highest-paid faculty members on campus (Office for Women's Affairs 2002).

What accounts for these persistent gender differences, and what, if anything, should and can the College of Arts and Sciences do about them? National discussions of the faculty gender gap generally attribute the patterns to two main sources: (1) to differences, on average, in the family responsibilities of male and female professors and (2) to gender bias, or discrimination, on the part of those who make hiring and promotion decisions and set salaries. Because of the former, it is thought, women's accomplishments lag behind men's on average. Because of the latter, women whose professional accomplishments are on a par with their male peers receive less recognition and lower rewards for them. Many discussions of gender inequality in the academy conflate these two possible sources of gender differences in rank, salary, and the like, but they contribute in entirely different ways to the gender gap and thus require different means of redress (assuming that an institution chooses to redress them). Further, in discussions of women's lower status in the academy the conflation of these two sources of gender inequality often deflects attention away from gender bias. Academics are apparently more comfortable attributing the lagging status of faculty women to the family choices they make than to the potential gender bias they encounter in the academy.

Like many other universities, Indiana University has instituted a number of "family-friendly" programs and policies to help alleviate the career disadvantages that female faculty with children are assumed to experience (as well as male faculty who have primary or equally-shared parenting responsibilities). The development, implementation, and funding of family-friendly policies is a campus responsibility, and thus we focus our attention here on what is within the purview of the College – the question of whether gender bias is a factor in the evaluation of College faculty and thus contributes to a gender gap in position and salary. In what follows, then, we direct our attention to what studies show about gender bias in the evaluations of equivalent men and women and to a discussion of steps the College could take to reduce this source of gender difference in position and salary, recognizing that other sources of gender difference may remain (e.g., differential family responsibilities between male and female faculty members).

Evidence of Gender Bias

Many studies that try to identify the sources of the gender gap in the workplace rely on analyses that link workers' salaries, positions, and other job outcomes to their background characteristics, including gender, family status, education, years of experience, and the like. In studies based solely on the statistical analysis of such background and outcome data, it is very hard to differentiate among the many factors that contribute to gender differences in outcomes, only one among which may be gender bias. Some factors are simply unmeasurable; some other factors, while measurable, conflate sources of gender difference. These studies do not study gender bias directly; rather, any unexplained difference between men and women that remains after controlling for the measurable factors known to affect job outcomes is attributed to bias (i.e., the residual is called "discrimination"), leaving the decision-making process itself inside a "black box." Thus statistical studies of the characteristics of workplace populations, including the prior studies conducted at Indiana University, do not directly measure bias nor the processes through which bias operates, and thus at best provide only partial evidence about the causal importance of gender bias in the production of gender difference in salary and position.

To address more directly the question of whether women are, on average, evaluated less positively than equivalently-qualified male peers (in other words, if gender bias results in "likes" being treated differently on the basis of gender), social scientists have turned to experimental and quasi-experimental studies in which equivalence between men and women is experimentally established. Under these conditions, any difference in the evaluations of women and men can be unequivocally attributed to bias. The findings of these studies show clearly that gender bias results in likes being treated differently, although in most cases the evaluators are unaware they are doing so. The following examples are illustrative, not exhaustive.

• In a classic social psychology experiment, hundreds of subjects were asked to evaluate the same article, which differed only in whether it were supposedly written by a male, a female, or an author with a sexually ambiguous name. The essays attributed to the male authors received on average a significantly higher grade, from female raters as well as male raters (Paludi and Strayer 1985).
• Experimenters showed subjects a number of photographs of individual people and asked them to judge the height of those photographed. The average height of the men was exactly the same as the average height of the women in each set, but nonetheless the subjects judged the women to be shorter than they really were and the men to be taller (Biernat, Manis and Nelson 1991). Height is a prime example of an easily-measurable characteristic, and gender bias is more likely to enter into evaluations based on subjective criteria.
• Women and minorities are more likely to be hired when the selection process is blind with respect to ascriptive characteristics (Reskin 2003). For example, when the Civil Service stopped requiring photos as part of job applications, the proportion of women and minorities hired increased dramatically (Rosenbloom 1977). Symphony orchestras that conduct blind auditions (behind a curtain) hire more female orchestra members (Goldin and Rouse 2000). Women are more likely to be hired when the application process is computerized and they can conceal their gender (Richtel 2000; Sturm 2001).
• Audit studies that experimentally create equivalent matched pairs of males and females (by arming them with equivalent fake resumes, credentials, etc.) show that male auditers are given better treatment in a variety of settings. For example, owners/managers of high-priced restaurants are more likely to hire men as waitstaff (Neumark 1996), car dealers quote significantly lower prices for new cars to white men than women or black men (Ayres and Siegelman 1995), temp agencies are more likely to refer women to clerical jobs, and golf courses are more likely to grant prime tee times to men (Prime-Time Live 1993).

In general, the more subjective and discretionary the evaluative criteria, the more gender (and racial) bias enters into the decision. Sometimes the differences are small; sometimes they are substantial. In the context of the academy, "the small but systematic undervaluation of women culminates in women's smaller salaries compared to men's, and slower rates of promotion" (Valian 2004, p. 211).

The experimental and quasi-experimental research findings discussed above show that women are generally not evaluated as positively as equally-situated men, but they say little about why. Research by social psychologists that directly studies the decision-making process identifies three major sources of bias in the evaluation of women and men (see summary in Williams 2004). First, women's performance tends to be more closely scrutinized than men's and judged by stricter standards, making it harder for women to be perceived by others as "competent." Second, men have to give more convincing demonstrations of incompetence to be judged by others as incompetent. Third, "men tend to be judged on whether they show promise, whereas women in similar circumstances are often judged strictly on what they have actually accomplished," a form of bias that is particularly likely to affect hiring and promotion decisions. All in all, different standards are used to evaluate men and women and the way in which those standards are applied results in men being judged as more competent, although those doing the evaluating generally believe they apply the same criteria to women and men; the gender bias built into these evaluative processes is an "invisible hand" that creates gender inequality in workplaces (Ridgeway 1997, p. 218).

There are further competence Catch-22s for women. One, women who are judged to be highly competent suffer from it (Williams 2004), especially if they present themselves as competent. Heilman (2001), for example, find that "assertiveness" in male colleagues is likely to be taken as a sign of brilliance, whereas assertiveness in female colleagues is generally judged negatively, and successful women in predominantly-male work settings are seen as far less likeable than equivalent males. Second, even when a woman's successes are acknowledged they tend to be "explained away" as a result of idiosyncratic factors or attributed to the efforts of others in the workplace, rather than taken as a sign of her general competence (Heilman 2001).

The largely unconscious nature of these differential evaluations makes it particularly difficult to eliminate the gender bias in evaluation processes.

Recommendations

What can an organization do to reduce, if not eliminate, gender bias in evaluations (here we use the term evaluations as a shorthand for all decisions made about faculty members' careers, including hiring, promotion, and salary setting)? Some of what is recommended below may be informal practice at present that should be formalized as policy.

1. The College should make known to all faculty, but especially department chairs and members of review committees, how gender is likely to influence evaluations and rewards if potential gender bias is unexamined and unchallenged. This could take the form of a memo and/or could be a point of emphasis in the Dean's regular meetings with chairs/directors and departments. As Williams (2004) puts it, "Recognizing stereotypes is the first way to eliminate them." This may be particularly hard for academics, who generally think their meritocratic principles make them immune to bias in decision-making, including gender bias. Valian (2004, p. 211) argues that "our belief in our own good will can make it difficult for us to see what we are doing."
2. The College should require departments to review regularly their tenure, promotion, and salary-setting guidelines for unintended gender bias and should similarly examine College evaluation procedures for bias. We cannot rely on one-time fixes, such as the 1997 gender equity salary review and subsequent adjustments. The routine, taken-for-granted decision-making processes that produced the unequal outcomes addressed by those adjustments will just continue to produce future unequal outcomes if the routine decision-making processes are not themselves altered. For example: We should ask whether research on gender or publication in interdisciplinary journals devoted to research on gender is discounted or devalued. We should not rely solely on quantitative markers of productivity to assess scholarly contribution and impact. Publication counts may systematically underestimate women's scholarly contributions because women's publications receive, on average, more citations than men's, suggesting that each publication has, on average, greater impact. Further, citation counts are themselves conservative estimates of visibility for women in that women receive fewer "symbolic citations" for their publications than men do, especially if they work in male-dominated fields (Ferber 1986, 1988; Lamont et al. 2004; Ward, Gast and Grant 1992). Citation counts are also problematic in that they index works referenced in journal articles but not books.
3. The College should increase accountability for eliminating gender bias from all forms of faculty evaluations by requiring decision-makers to communicate their decisions and defend them. Holding evaluators accountable for their decisions decreases all forms of ascriptive bias (Reskin 2000, 2003). A number of possibilities should be College policy and practice, namely:
   1. Compile and publish an annual review of gender equity benchmarks, such as salary, tenure rates, time in rank, proportion of chairs and other administrators who are female, and the like.
   2. Monitor individual departments for their progress in increasing gender equity and in taking steps to improve the gender climate. Make it a routine part of the chair review and selection process.
   3. Regularly ask chairs to consider and/or to demonstrate whether women's and minorities' salaries are in line with their accomplishments. Make it a point to question chairs about salaries for women that seem too low and/or about women whose time in rank seems out of line with their overall accomplishments and contributions.
   4. If women are not in the pool for a particular position, ask why not.
4. The College should require each department to develop clear and explicit procedures for hiring and criteria for promotion and salary setting, and require each department to make those procedures and criteria known to all faculty. Ambiguity in performance criteria increases the likelihood of gender bias (and other forms of bias as well); conversely, making the criteria for the various faculty rewards more explicit generally facilitates women's success (Fox 1985). In discussions about candidates' (for hiring, for promotion, for administrative appointments, for awards) strengths and weaknesses, we should beware of vague and subjective gendered attributions that are often used to discredit women (e.g., "not a team player," "emotional," "intemperate," "pushy," "difficult"). We should also beware of suggestions that women, more than men, need to "wait their turn," aren't quite ready to be promoted, or cannot be otherwise leap-frogged over males in the department.
5. Recruitment committees, salary committees, and other review committees should be as gender-integrated as possible. Gender bias is less likely to creep into evaluations conducted by mixed-sex than by all-male or predominantly-male groups (Reskin 2000).
6. Special efforts should be made to appoint more women in key decision-making positions (e.g., to chair recruitment committees rather than simply serve on them) and to identify talented women candidates for higher-level appointments who might otherwise escape notice. Women are more likely to be selected for positions generally occupied by men (e.g., administrative positions, named chairs) when there are more women in critical administrative or other gate-keeping positions (Konrad and Pfeffer 1991; Cohen et al. 1998; Lamont et al. 2004).
7. When evaluating a faculty member, College administrators and review committees should encourage colleagues to imagine a scenario in which the person being evaluated was of another gender, and ask whether the evaluation would be the same.
8. Take steps to diminish and/or guard against gender bias in the two especially important opportunities for College faculty to receive larger-than-average salary increments: in response to an outside offer or as part of a preemptive offer, and as part of an appointment to an administrative position. There are three ways in which this source of gender bias may be diminished:
   1. The College and/or the campus should collect data for all departments over a period of several years on the faculty who receive outside offers and preemptive offers in anticipation of receipt of an outside offer, tracking whether a counteroffer was made before or after a firm offer was received, the terms of the outside offer and the counter, and the like. This will allow for a far better assessment than is possible at present of whether gender is a factor in these decisions.
   2. The College and/or the campus should collect data for all departments and for College administrators over a period of several years on the salary adjustments that faculty receive for taking administrative appointments. This will allow for a far better assessment than is possible at present of whether gender is a factor in salary-setting among those faculty who receive administrative appointments.
   3. The College should regularly reiterate to all chairs, directors, and faculty its position that gender discrimination in any form is not to be tolerated, including gender discrimination in the proffering of preemptive offers, responses to outside offers, appointment to administrative positions, and setting of administrative salaries. Further, the College should encourage any woman who feels she has experienced any of the forms of gender discrimination described above to use the same channels of redress in place for non-salary forms of discrimination (such as sexual harassment): to bring the problem to the attention of the College associate dean who handles personnel matters. He or she should in turn work with other offices on campus, as appropriate, to investigate, to determine whether redress is warranted and, if so, to recommend to the Dean the appropriate form of redress. In the formal response to the faculty member who alleges gender discrimination, the College should remind her that if she is not satisfied with the College's findings and proposed resolution she may take her case to the Office of Affirmative Action. The current salary equity review guidelines disallow an adjustment for salary differences that arise as a result of outside offers or administrative appointments, which in the process leaves no routine way within the College to adjudicate potential gender discrimination in responses to those faculty who are at risk of outside offers or receive outside offers or who are candidates for administrative appointments. Yet such cases are more likely to be resolved with a minimum of conflict if they can be resolved within the College.
9. The College and/or the campus should keep records of all departmental and College awards, and periodically review the distribution of such awards between men and women. The routine operation of unconscious gender bias may result in women receiving fewer markers of distinction (bestowed by their departments and/or the College) than they merit on the basis of their accomplishments (Office for Women's Affairs 2002).

Benefits to the College

The process of over-time cumulative disadvantage to faculty women that results when gender bias is unchecked is also a process of cumulative discouragement. This cumulative discouragement can lead to disaffection and alienation, which in turn can lead to either withdraw or departure from the institution. For example, faculty women who feel less involved in their departments and institutions are more likely to leave (Trower and Chait 2002). Cumulative discouragement also affects productivity: Irrespective of gender, scholars who work in organizations where they feel valued and appropriately rewarded for their scholarly contributions are more productive (Pelz and Andrews 1976). Anecdotally, senior faculty women can often tell several stories apiece of senior women who left a department or institution they found unhospitable to women to take an appointment in a department they perceived as having a more favorable gender climate. All of these results of cumulative discouragement represent losses to the institution.

Addressing the problem of gender bias in all forms of faculty evaluation clearly stands to benefit those individual women faculty whose achievements may have been undervalued and underrewarded. It also stands to benefit departments within the College and the College as a whole in significant ways.

References

Ayres, Ian and Peter Siegelman. 1995. "Race and Gender Discrimination in Bargaining for a New Car." American Economic Review 85:304-21.

Biernat, Monica, Melvin Manis, and Thomas E. Nelson. 1991. "Stereotypes and Standards of Judgment." Journal of Personality and Social Psychology 60:495-502.

Clemens, Elisabeth S., Walter W. Powell, Kris McIlwaine, and Dina Okamot. 1995. "Careers in Print: Books, Journals, and Scholarly Reputations." American Journal of Sociology 101:433-494.

Cohen, Lisa E., Joseph P. Broschak, and Heather A. Haveman. 1998. "And Then There Were More? The Effect of Organizational Sex Composition on the Hiring and Promoting of Managers." American Sociological Review 63:711-727.

Curtis, John W. 2004. "Faculty Salary and Faculty Distribution Fact Sheet 2003-04." American Association of University Professors.

Ernst, B. 1998. "Disparities in the Salaries and Appointments of Academic Women and Men." American Association of University Professors.

Ferber, Marianne. 1986. "Citations: Are They an Objective Measure of Work of Men and Women?" Signs 11: 381-389.

Ferber, Marianne. 1988. "Citations and Networking." Gender & Society Gender & Society 2:82-89.

Fox, Mary Frank. 1985. "Publication, Performance, and Reward in Science and Scholarship." Pp. 255-282 in Higher Education: Handbook of Theory and Research, edited by John C. Smart. NY: Agathon.

Goldin, Claudia and Cecelia Rouse. 2000. "Orchestrating Impartiality." American Economic Review 90:715-41.

Heilman, Madeline E. 2001. "Description and Prescription: How Gender Stereotypes Prevent Women's Ascent Up the Organizational Ladder." Journal of Social Issues 57:657-674.

Konrad, Alison and Jeffrey Pfeffer. 1991. "Understanding the Hiring of Women and Minorities in Educational Institutions." Sociology of Education 64:141-157.

Lamont, Michele, Alexandra Kalev, Shawna Borden, and Ethan Fosse. 2004. "Recruiting, Promoting, and Retaining Women Academics: Lessons from the Literature." Report prepared for the Standing Committee for the Status of Women, Faculty of Arts and Sciences, Harvard University. Distributed at the December 14, 2004 meeting.

Neumark, David. 1996. "Sex Discrimination in Restaurant Hiring: An Audit Study." Quarterly Journal of Economics 3:915-41.

Office for Women's Affairs. 2002. "Report on the Status of Women, Indiana University-Bloomington."

Paludi, Michele A. and Lisa A. Strayer. 1985. "What's in an Author's Name? Differential Evaluations of Performance as a Function of Author's Name." Sex Roles 12:353-361.

Pelz, Donald and Frank Andrews. 1976. Scientists in Organizations. Ann Arbor: Institute for Social Research.

Prime-Time Live. 1993. "The Fairer Sex?" ABC News, aired October 7.

Ridgeway, Celilia L. 1997. "Interaction and the Conservation of Gender Inequality: Considering Employment." American Sociological Review 62:218-235.

Reskin, Barbara. 2000. "The Promixate Causes of Employment Discrimination." Contemporary Sociology 29:319-328.

__________. 2003. "Including Mechanisms in Our Models of Ascriptive Inequaltiy." American Sociological Review 68:1-21.

Richtel, Matt. 2000. "Online Revolution's Latest Twist: Computers Screening Job Seekers." New York Times, February 6, sec. 1, page 1.

Rosenblum, David H. 1977. Federal Equal Employment Opportunity: Politics and Public Personnel Administration. NY: Praeger.

Sturm, Susan. 2001. "Second Generation Employment Discrimination." Columbia Law Review 101:458-568.

Trower, Cathy and Richard Chait. 2002. "Faculty Diversity: Too Little for Too Long." Harvard Magazine 104:33.

Valian, Virginia. 2004. "Beyond Gender Schemas: Improving the Advancement of Women in Academic." NWSA Journal 16:215-220.

Ward, Kathryn B., Julie Gast, and Linda Grant. 1992. "Visibility and Dissemination of Women's and Men's Sociological Scholarship." Social Problems 39: 291-298.

Williams, Joan. 2004. "Hitting the Maternal Wall." Academe, 90 (7 pages).

NOTE: Heilman (2001), Reskin (2000), Valian (2004), and Williams (2004) are particularly useful for overviews and recommendations for action.