Recruitment of Excellent and Diverse Faculty in STEM Dr. Eve Riskin Associate Dean for Diversity and Access Faculty Director of UW ADVANCE Professor of Electrical Engineering University of Washington December 7, 2015

UW ADVANCE TEAM  This is collaborative work with the UW ADVANCE team: –Dr. Joyce Yen –Alexis Nelson –President Ana Mari Cauce

Outline  My background  UW ADVANCE  A little on bias  Recruitment

My Background  NJ native (Lawrence High School)  Research in image and video compression  Twenty-five years in the UW Electrical Engineering Department  Two stories of my 1 st department chair  Worked on UW ADVANCE since 2001  Associate Dean of Engineering, 2005–

Me (2) – Research Areas  Image and video compression with HCI focus  Technology for people with disabilities  Diversity and Higher Education

Bob Gray

My Leadership Mentor, Denice D. Denton

Richard Ladner

Outline  My background  UW ADVANCE  A little on bias  Recruitment

Why Diversity Matters  If you leave out half the population, you don’t get the best solution  Two examples follow where considering gender diversity would lead to better solutions

Example 1: Playground Study  UW researchers (Kilgore, Atman, et al.) asked freshmen what five categories are most important to design a playground  Next slide has categories with statistically different response rates by gender

* p < 0.10 or ** p < 0.05, Fisher exact

Example 2: The Airbag Photo: Mike Babcock

UW ADVANCE Principles Fix system (NOT the women) Close contact with leaders & policies Systemically Focused Immediate applications Focus on faculty questions Practical Peer mentoring Speakers within STEM departments Peer-to- Near-Peer

Leadership Development  Culture is influenced by unit leader  Issues of diversity woven throughout  Sample Topics: Dual career hires, tenure clock extensions, student ratings of STEM women, building faculty offers

Impact of UW ADVANCE

% Female Faculty in Top 50 Colleges of Engineering

Professor Elizabeth Nance  Clare Boothe Luce Assistant Professor of ChemE at UW  Burroughs Wellcome Career Award  2015 Forbes 30 under 30  11 interviews  11 offers

Outline  My background  UW ADVANCE  A little on bias  Recruitment

Educate Yourself About Bias  Recognize that underrepresented candidates are subject to different expectations – “the bar goes up”  Read the Wisconsin ADVANCE brochure: “Reviewing Applicants: Research on Bias and Assumptions”

Research on Bias in Evaluation Lab Manager Appl’n (Moss-Rascusin et al., 2012) Companies Students Faculty Privileged identity (dominant group) received more positive evaluations, regardless of the identity of the evaluator vs.

624 recommendation letters for psychology faculty candidates Women: communal descriptors; Men: agentic descriptors Communal characteristics negatively correlated with hiring decisions Same CV, different name Male applicant rated better in all categories, more likely hired Pattern holds for both men and women reviewers Same CV, different name – reviewed by science faculty at R1s Males rated significantly more competent and hireable Higher starting salary and more career mentoring offered to males Pattern holds for both male and female reviewers Teaching Evaluations (MacNell et al., 2014) Online instructors, one male and one female, taught 2 courses each, one as a male and one as a female Students rated “male” teacher higher in both cases Students rated “female” teacher lower in both cases “Men get bonus points for showing up male” White vs. Black names, 2 skill levels each Whites: 50% more callbacks Highly skilled and avg. blacks virtually same number callbacks Avg. skilled whites more callbacks than highly skilled blacks

 In Nature (1997), Wenneras & Wold showed that in order for a female scientist to be awarded the same competence score as a male, she had to have 3 extra papers in Science or Nature (i.e., high impact journals) or 20 extra journal papers (e.g., good specialist journals) OR know someone on the panel.  “A female applicant needed to be 2.5 times more productive than the average male applicant to receive the same competence score.” Wenneras & Wold (1997) 7/n6631/pdf/387341a0.pdf

More Bias Research Studies  2007 meta-analysis shows men have 7% higher chance of getting grants (NSF, Wellcome Trust, NIH, Australian, Germany, …)  2014 study finds women are 55% less likely to get tenure in Computer Science when controlling for research productivity  RateMyProfessors.com study, 2015 –Men are “brilliant,” “intelligent,” and “smart” –Women are “mean,” “unfair,” and “annoying”

Outline  My background  UW ADVANCE  A little on bias  Recruitment

Before candidates come to campus (1)  Set your criteria ahead of time  Seek balance between too broad and too narrow a search – hires should move your department forward  Find a good source of female and URM candidates –Rising Stars in EECS at MIT and UC Berkeley –NextProf at U Michigan –NSF AGEP programs

Rising Stars in EECS 2013

Before candidates come to campus (2) Doctoral Degrees Awarded to Women by Engineering Discipline Biomedical Engineering 37.3% Chemical Engineering 29.5% Civil Engineering 22.8% Environmental Engineering 40.2% Civil/Environmental 28.7% Computer Science (Inside Engineering) 17.3% Computer Science (Outside Engineering) 19.1% Computer Engineering 12.8% Electrical/Computer Engineering 17.1% Electrical Engineering 15.1% Engineering Science & Engineering Physics 24.8% Industrial/Manufacturing Engineering 28.8% Mechanical Engineering 15.1%

Before candidates come to campus (3)  Have someone talk to your search committee about implicit bias  Collect data on the candidate pool (ASEE is a good source if you’re in Engineering)  Rank your finalists according to different criteria to see who rises to the top – excellence is multi- dimensional  If the proposed slate of candidates is all majority men, review files of the best woman and URM candidate(s) – some people are better in person

During the interview  Try to interview more than one woman/URM – you increase your chances of a diverse hire if you do  Consider an all-women meal during interview (if the candidate is open to it)  Have someone from the Murray Center for Women and Technology meet with female candidates  No illegal or inappropriate questions

Things to remember (1)  Always be recruiting  The candidates are interviewing you too  Create the best experience for the candidate  If you make an offer, stay in touch frequently

Things to remember (2)  Women may not negotiate as hard as men –Don’t hold this against her –Make sure she is set up to succeed –Shortchanging her can cause a retention problem  Track the women you didn’t interview or hire –Are they doing well? –Are there patterns? –Can you do something differently next time?

Two Case Studies

Case Study #1  During discussions of the faculty candidates, a faculty member in your department consistently finds a way to discredit female candidates. What do you do?

Case Study #2  During a faculty candidate’s chalk talk, a young faculty member in your department aggressively questions the candidate’s research. What do you do?

Thank you!