Best Practices to Address Bias in Personnel Decisions

Promotion and Tenure (P&T) is a critical component in building the intellectual infrastructure of a university. From the university’s perspective, it determines whether a faculty member has the opportunity to remain on the faculty and further their contributions. For individual faculty, it marks a major career milestone that shapes the rest of their career trajectory.

Given the importance of P&T outcomes to all involved, it is critical that the evaluation process is fair and equitable. Yet research has shown that such evaluation processes are susceptible to bias. To mitigate such biases, personnel committee members should become informed about commonly held assumptions that can influence their own assessments of P&T candidates as well as those of external letter writers.

Assumption #1: The P&T process is completely unbiased and objective.

What we know:

• The potential for bias is present in many of the criteria used to evaluate P&T candidates.
• Specific review criteria (e.g., number of publications) are often left vague to allow flexibility in capturing the varied ways in which academics contribute to the University and their disciplines. Although well-intended, this vagueness has the downside of introducing potential bias into the review process.
• When comparing candidates for tenure, the success rate for male candidates increased by 19.4 percentage points after stopping the clock for parental leave. For women, the rate fell by 22.4 percentage points. There is evidence of a differential impact of Covid-19 on women versus men academics.

Best Practices for P&T Committee Members

• Acknowledge possible biases in P&T guidelines and process, in the perspectives of both internal and external reviewers as well as in your own perspectives, and seek to prevent them contributing to biased decision making.
• Attend training to understand the impact of bias on the process and ways to mitigate it.
• If serving as P&T committee Chair, consider making a statement about potential bias at the start of each P&T committee meeting which reminds your colleagues about the importance of being cognizant of potential sources of bias.
• Use the ADVANCE Office for Faculty Equity-recommended protocol to organize your discussion and assessment of candidates’ dossiers.
• Take some time to review the differential impact of Covid-19 discussed by Malisch, et al. (2020) focusing specifically on the questions in figure 1.

Assumption #2: The quality of a candidate’s research and scholarship is objectively measured by formal academic indicators.

What we know:

• Making determinations about the quantity and quality of publications is not straightforward; thus, committees often rely on proxy indicators like citation counts and Journal Impact Factors (JIF). However, such indices can provide misleading information about the impact of a scholar’s work and can be subject to biases.
• In terms of citation counts, there is research that indicates that women’s scholarship tends to be cited less often than men’s even when they publish in high quality/highly visible journals.vi Furthermore, there is evidence that men researchers engage in more self-citations than do women researchers, which then inflates their overall citation rates.
• Critiques of the JIF consider it to be somewhat unreliable, because it relays little about the quality of the review process or the quality/content of the journal, and tends to favor rapid rather than prolonged impact.
• White researchers are almost twice as likely to receive grants from high-profile funders, as compared to their Black peers. Furthermore, evidence suggests that men applicants receive significantly more competitive “quality of researcher” evaluations (but not “quality of proposal” evaluations) and have significantly higher funding success rates than women applicants.
• Faculty of Color do not achieve tenure in the same proportion as White men. African American and Hispanic faculty are significantly less likely to achieve tenure than their White colleagues. This differential can be attributed, at least in part, to an ambiguous tenure process as well as a dearth of effective mentors.
• Women researchers publish less frequently than their male colleagues often due to heavy teaching and committee service, which reduces their time to conduct research and publish.
• Gender disparities persist in authorship order, which is considered a critical indicator of presumed level of contribution. Although there is variation in authorship order patterns across disciplines, overall, women’s authorship remains under-represented in prestigious first author position. The reasons for this are not entirely clear, but some researchers attribute it to women failing to negotiate as effectively as men for first author positions.
• There is also broader research that shows that, in general, women’s collaborative work is often attributed to men colleagues. Similarly, men tend to receive more credit for collaborative research with women than women receive for collaborative research with men, thereby contributing to the systemic denial of women’s contributions to the scholarship. For example, while men economists get tenure 75% of the time regardless of whether the majority of their research is single-authored or collaboratively authored, women who co-author most of their papers only get tenure about 50% of the time.

Best Practices for P&T Committee Members

• Become knowledgeable about the research on gender, race, and ethnicity-related biases that can affect determinations about the quality and impact of a researcher’s/scholar’s work and try not to let these factors influence your decisions.
• Set aside preconceived ideas about what comprises "productivity” and particularly when assessing candidates outside your discipline. Acknowledge that differences in standards across disciplines can be misinterpreted as evidence of lack of scholarly rigor or knowledge.
• Do not rely on formal indices of citation counts or the JIF as the sole sources for assessment of academic performance – particularly for people working in smaller fields and/or on niche topics where the specialized audience might affect the numerical indices even when the work is highly significant.
• Consider faculty impact achieved through less traditional means such as policy reports, legal decisions, podcasts, inclusion on syllabi at other institutions, and non-traditional curriculum development.
• Where guidelines allow, award faculty for their interdisciplinary work if appropriate to the form and purpose of their scholarship. If the work is innovative, high quality, and impactful, judge such work on its own merits. For example, such work may include the use of digital media and work that informs social justice work in the legal, healthcare, and education systems.
• Keep in mind that engagement in applied research may result in “gaps” in a candidate's formal publication record given the time needed to implementation the interventions to be researched, data gathering, and/or analyses. Publication gaps in these situations are not typically indicative of a lack of scholarly work during those periods.

Assumption #3: Outstanding candidates are best vouched for by external scholars, particularly those in prestigious institutions.

What we know:

• The P&T system can unwittingly reinforce bias by assuming that scholars at institutions of equal or greater prestige are more capable of evaluating a candidate’s work even when someone at a less prestigious institution has better knowledge of the candidate’s field. The reviewer with an impressive pedigree may not be the most familiar with the standards of the sub-discipline, the specific area of research, or the context of the candidate’s work. This issue is often intensified for the evaluation of interdisciplinary work.
• Many women faculty and faculty of color lack the same “sponsorship” and career connections required to gain the attention of high-profile reviewers in top ranked programs.
• Research on letters of recommendation may also be relevant to issues in P&T review letters. Letters of recommendation often present men and women candidates for academic positions in different ways. In one study, recommenders used significantly more standout adjectives to describe men as compared to women candidates. Recommenders emphasized women’s strong work ethic and their potential as teachers, whereas the focus in men’s recommendations included greater confidence in their research and ability.

Best Practices for P&T Committee Members

• Be aware of potential bias introduced into the P&T process by external reviews who may be selected because of their high profile rather than the content of their scholarship and similarity of methodology and framework with the candidate.
• Encourage additional input from non-academic reviewers when relevant to the case the candidate is trying to make regarding the impact of their work.

Assumption #4: Student evaluations are an objective measure of faculty teaching.

What we know:

• Students are often asked to assess aspects of faculty work that they do not have sufficient knowledge to evaluate.
• Students apply more stringent criteria for considering Black faculty as intellectually competent than they do White faculty. Students more frequently question faculty of color’s authority and knowledge in a classroom setting than they do for White faculty. Women faculty of color describe more frequently being challenged or disrespected by students in the classroom compared to their White or men colleagues which can affect evaluations.
• Regardless of instructor performance, students rate instructors perceived to be female significantly more harshly than those perceived to be male. Furthermore, women instructors must satisfy a higher threshold that their men peers to be rated equivalently. For example, a male faculty member can demonstrate competence and be unfriendly toward students but still be considered intellectually competent. A female faculty member, however, must demonstrate competence and friendliness to be judged intellectually competent.
• Men and women faculty may have different teaching styles which are assessed differently by students. For example, men college instructors reported lecturing twice as often as their women peers, while women reported using significantly more discussion. Students inaccurately interpreted these gendered differences as women’s lack of knowledge or preparedness and men’s vast knowledge and preparation. Yet, there is evidence to suggest that discussion-based learning increases gains in understanding and knowledge.
• Students are more likely to ask women professors for favors such as extra credit, extended deadlines, and alternative assignments than male professors. Furthermore, students perceive women professors as less fair than their male counterparts. In particular, women professors are expected to expend exceptional effort to help out their students in sometimes unrealistic ways, thus resulting in worse evaluations.
• Faculty who teach race-focused courses or any courses that involve controversial topics are more likely to be assigned low ratings by students.
• Faculty who speak with a foreign accent are more likely than their native English-speaking peers to be evaluated negatively by their students.

Best Practices for P&T Committee Members

• Treat student ratings and comments as just one source of ‘feedback on instruction’ rather than as conclusive evidence. Carefully consider the weight placed on student evaluations because of their potential to introduce bias associated with factors such race, gender, age and faculty speaking with an accent – and intersections among these factors. Also beware of student ratings of faculty knowledge of their discipline, which students are not qualified to assess.
• Use multiple sources of evidence about teaching such as chair observation plus peer observation to assess candidates’ teaching.
• Acknowledge the biases that can be introduced by poorly designed or administered questionnaires.

Assumption #5: Service is equitably distributed among faculty regardless of faculty status, rank, and teaching load.

What we know:

• Women faculty perform significantly more service than men do, controlling for rank, race/ethnicity, and field or department. Women, on average, allocate more hours to university service and less time to research than do men. This service is also concentrated in internal rather than professional contexts, which therefore does not garner as much external recognition or expand the faculty member’s professional networks.
• The expectation of BIPOC faculty is that they will mentor students of color and serve as the “diversity representative” on committees for their department, college, and university, which often results in a lack of time to focus on scholarship. This is referred to as ‘cultural taxation’ coined by Amado Padilla in 1994.
• BIPOC Faculty frequently asked – and often feel compelled – to participate in service activities that might benefit their racial or ethnic community. Again, this may lead to a service overload and result in less time for scholarship. 
• Committee service is largely viewed as a minimal requirement to be met during the promotion process as compared to publications or grants. This means that excess committee participation can seriously harm the career prospects of women faculty and faculty of color.
• Service work, mostly undertaken by women faculty, especially those from traditionally marginalized groups, is often invisible and therefore goes unrewarded.

Best Practices for P&T Committee Members

• Acknowledge and reward all types of service. For example, service to the university that might be very high profile should be no more important as service to students, which is less “visible” but often requires the same if not more time and effort.
• When available, consult data dashboards to get a better sense of faculty workload relative to influencing factors such as demographics, course level, course release, and overload.
• To get an informed perspective about an equitable approach to assessing service in the P&T review process see the following discussion “Creating Equitable Universities: Recognizing Faculty Workloads" on YouTube.

Further Reading

• Buser, W., Batz-Barbarich, C.L. & Hayter, J.K. Evaluation of Women in Economics: Evidence of Gender Bias Following Behavioral Role Violations. Sex Roles 86, 695–710 (2022).
• Chatman, J.A., Sharps, D., Mishra, S., Kray, L.J., & North, M.S. (2022). Agentic but not warm: Age-gender interactions and the consequences of stereotype incongruity perceptions for middle-aged professional women. Organizational Behavior and Human Decision Processes, 73, 104190, ISSN 0749-5978.
• Carnes, M., Devine, P. G., Isaac, C., Manwell, L. B., Ford, C., Byars-Winston, A., Fine, E., & Sheridan, J. (2012). Promoting institutional change through bias literacy. Journal of Diversity in Higher Education, 5(2), 63-77.
• Devine, P. G., Forscher, P. S., Austin, A. J., & Cox, W. T. L. (2012). Long-term reduction in implicit race bias: A prejudice habit-breaking intervention. Journal of Experimental Social Psychology, 48, 1267-1278.
• Fine, E. & Handelsman, J. (2012). Searching for Excellence & Diversity: A Guide for Search Committees: National Edition. Madison, WI: Women in Science & Engineering Leadership Institute, University of Wisconsin-Madison.
• Fiske, S. T. (2002). What we know now about bias and intergroup conflict, the problem of the century. Current Directions in Psychological Science, 11(4), 123-128.
• O’Meara, K.A. Change the tenure system. Inside Higher ED. January 2014. KerryAnn O’Meara
• Perna, L. W. 2005. Sex differences in faculty tenure and promotion: The contribution of family ties. Research in Higher Education, 46(3), 277-307.
• Trix, F., & Psenka, C. 2003. Exploring the color of glass: Letters of recommendation for female and male medical faculty. Discourse and society, 14, 191-220