Based on the body of evidence described below, AFFECT proposes the following set of Best Practices

 

Highlighted Latest Research

 

  • According to Budden et. al, (2007) peer review where the gender of the author is not revealed favors female academics. This can be seen in the Behavioral Ecology Journal which introduced double-blind peer review and noticed an increase in papers where the first author was female.
  • The paper by Gottlieb, Townsend and Xu (2016) explores the impact of job-protected leave and its correlation to entrepreneurship. The study finds that if employees receive an extended leave of absence, with a guarantee of returning to the job, they are more likely to become entrepreneurs. A reform in Canada, which extended maternity leave to a year, saw an increase in entrepreneurship amongst women who now own 35-37% of small and medium-sized enterprises.
  • In order to study the individual characteristics of candidates in top executive positions such as CEOs or CFOs, Kaplan and Sorensen (2016) investigated a data set with over 2600 executive assessments. The research shows that CEOs have a higher score than CFOs and that these scores can predict the career development. Interestingly enough the scores do not differ between men and women, yet women are less likely to become CEOs.
  • A new paper by Jannati, Kumar, Niessen-Ruenzi, and Wolfers (2017) finds that analysts tend have more favorable assessments of companies that are headed by people more similar to themselves.  For example, a male analyst tends to be more favorable about companies with male CEOs, and a republican analyst tends to be more favorable about companies with republican CEOs.  Because most analysts are male, American, and republican, earnings forecasts of companies headed by members of these groups are systematically higher.
  • A new paper by Schistermann, Swanson, Lu, and Mumford (2016) study citations of articles by gender of the first and last author, within the field of epidemiology where first and authors indicate more substantial contributions.  They find that articles in which the first and last authors are male incur significantly more citations.
  • Together, the results of Sarsons (2015) and of Schistermann et al (2016) suggest:
    • in finance, where ordering is alphabetical, the profession infers that the male co-author is the most important contributor
    • In epidemiology, where ordering is based on contribution, the profession infers that articles with male “key contributors” are more important

 

 

Differential outcomes for female academics

  • Has the gender gap closed?  Not in economics…
    • Although in most STEM fields women are becoming more equally represented, economics is a notable outlier.  There remain persistent sex gaps in promotion, which cannot be explained by productivity differences.  Women continue to “fall off the ladder” at each rung up, they receive lower salary adjustments, and they are less satisfied than their male colleagues.
    • Several papers address this issue, including for example:
      • Ginther and Kahn (2014) find that gender differences in tenure award have disappeared in most social sciences, BUT NOT IN economics.  In economics, they find a gender difference of 20% in promotion to tenure and 50% in promotion to full, after controlling for productivity and family choice.  Moreover, the difference is larger for women who are single and childless.
      • Ceci et al, (2014) offer a long time-series of multiple STEM fields   
      • Ginther and Kahn (2004) and Kahn (1993) focus on economics.
      • Kahn (2012) focuses on promotion in multiple fields at a single university.
      • This Bloomberg article highlights the issues.
      • The 2014 CSWEP annual report summarizes evidence from surveys over time, with a focus on the “leaky pipeline”.  See, in particular, pages 11 – 21.
      • Chen, Kim and Liu (2017) track the PhD class of 2008, and they find that women receiving PhDs in economics in 2008 were 10 percentage points less than their male peers to receive tenure as of 2016.
  • People hold implicit biases against women
    • Both males and females hold implicit biases against women.  These biases cause affect hiring, recommendation letters, grant reviews, and promotion.  Several recommendations are offered.  Dasgupta, N., and Stout, J., 2014.
  • Women get less credit when they co-author with males
    • Heather Sarsons, a PhD candidate in economics at Harvard, finds that women get less credit than men in co-authored research (Sarsons, 2015; See also, the related NYT article).  Sarsons examines the effects of an additional paper on prob(tenure):
      • Solo-authored paper increases tenure by:
        • for a man: 8%
        • for a women: 9%
      • Co-authored paper increases prob(tenure) by:
        • for a man: 8%
        • for a woman: 2%
      • Co-authoring with men increases prob(tenure) by:
        • for a man:  7%
        • for a woman: 0%
    • Schistermann, Swanson, Lu, and Mumford (2016) study citations of articles by gender of the first and last author, within the field of epidemiology where first and authors indicate more substantial contributions.  They find that articles in which the first and last authors are male incur significantly more citations.
      • Importantly, they find no difference in prior citations of the male versus female key authors (i.e., first and last authors)
      • It is noteworthy that epidemiology is a field in which the majority of faculty (54%) are women, but a minority of editorial boards are female (36%).
    • Together, the results of Sarsons and of Schistermann et al suggest:
      • in finance, where ordering is alphabetical, the profession infers that the male co-author is the most important contributor
      • In epidemiology, where ordering is based on contribution, the profession infers that articles with male “key contributors” are more important
  • Want better student evaluations?  Be male
    • Using an experiment designed around online teaching and holding everything but gender constant, MacNell et al, 2014 find that male instructors receive more positive evaluations than females.  See both the article and the Slate.com summary
  • What are the odds that a panel would be “randomly” all men?
    • Astronomically small according to this article in The Atlantic.

 

Evidence on biases against women, across fields

  • Goldin and Rouse (2000) find that “blind” auditions by orchestras, where the identity and gender of the musician are hidden, contribute to a 25% increase in the percent of orchestra musicians who are female.
  • Gender Action Portal (GAP), which is part of the Women and Public Policy Program and the Harvard Kennedy School, has aggregated an impressive set of studies on biases, outside of academia.  http://gap.hks.harvard.edu/
  • This page summarizes some recent social science studies on the implicit biases against women, which use strictly controlled methodologies.  The body of research offers important policy implications for the processes leading to hiring promotion and tenure.

 

Strategies to address the gender bias

  • “Our lives begin to end the day we become silent about things that matter.” Martin Luther King, Jr.
    • There are multiple examples of men being accused of wrongdoing, but schools attempting to protect them by not disclosing details.  Unfortunately, this can result in such practices continuing.  If there is an inappropriate behavior, it is important to be specific regarding what happened and to go through appropriate legal channels.  This will contribute toward fewer such problems in the future.
  • Benefits to increasing awareness of gender biases
    • Participants of a 2.5 hour workshop on gender bias were significantly increased their likelihood of promoting gender equity, compared to people that did not participate in such a workshop. Carnes et al, 2015
    • The American Society for Microbiology took concrete steps to achieve gender equity at a major meeting.  Both highlighting the under-representation of women at prior conferences, and encouraging the Program Committee to strive for more equal representation of women were found to be effective.  Casedevall, 2015
  • Increasing gender diversity on faculty
    • Smith et al conduct a randomized and controlled three-step intervention in faculty hiring, at one US university.  The approach focused on:  educating the search committee on the potential influence of gender-related biases, and providing the candidate with somebody outside the search committee to discuss work/life balance issues.  The intervention significantly increased the percent of women at every stage of the process:   initial interview to acceptance of an offer.
  • Avoiding gender bias in reference writing
    • Reference letters for men are more likely to emphasize accomplishments, while letters for women are more likely to mention personal life.  This one-page guide provides some checkpoints, for ensuring that the recommendation letter you are writing does not unintentionally portray a female student differently than a male student.
  • Family Friendly policies – benefits to women vs men
    •  As highlighted in a recent NYT article, research by Antecol, Bedard, and Stearns (2016) suggested that gender-neutral tenure-extension policies benefits men much more than women.  Specifically, the study concludes that such policies increased the probability that a male economist would receive tenure at his first job by 19 percentage points, compared to an analogous 22% decrease for women.
  • Ensure adequate representation of women at conferences, on panels
    • A recent New York Times article discusses the male-skewed lineups at conferences, and the efforts of a Princeton neuroscientist to combat this.  Specifically, a group of scientists started a website focusing on the gender ratios at conferences.

 

The importance of mentoring

 

The impact of female faculty

  • Female professors positively affect female students’ performance in math and science classes.
  • Female faculty positively affect proportion of women entering PhD programs

 

Different performance expectations for women

  • Women are stereotyped as not possessing certain talents. 

    Such stereotypes contribute to under-representations in academia.  In contrast, differences in aptitude do not explain observed patterns.

  • Women are less likely to negotiate over resources, and are more likely to agree to serve on committees.

 

Advice

  • How to navigate the tenure process?
    • How should a junior faculty member best allocate time?  Should somebody go on the market during the tenure-track period?  A series of articles in the CSWEP newsletter addresses these questions and more.
  • Tips on writing referee reports, presenting papers, and econometric techniques.

 

Personal experiences

  • Barbara Spencer recounts her experiences as a women economist, in Australia, Canada, and the United States.
  • A TEDX talk that highlights how subtle implicit bias can be – only 9 minutes long, and quite illuminating

 

Other resources and organizations