1 of 52

Plugging the leaky pipeline in Statistics

Keegan Korthauer

@keegankorthauer

UBC-SFU Joint Statistics Seminar

12 March 2022

2 of 52

The “leaky pipeline” is a metaphor often used to describe the lack of women making career advancements in STEM fields

2

source: sfu.ca/wwest/WWEST_blog/retaining-women-in-STEM

3 of 52

Focus of this talk

Description of the problem
Past: pioneering women in statistics
Present: ongoing sources of leaks
Future: outlook & strategies

3

4 of 52

The “leaky pipeline” is a metaphor often used to describe the lack of women making career advancements in STEM fields

4

source: sfu.ca/wwest/WWEST_blog/retaining-women-in-STEM

How bad is it?

5 of 52

Across STEM fields

5

Handelsman, Jo, et al. "More women in science." Science 309.5738 (2005): 1190-1191.

6 of 52

Math & Statistics

6

NSF: National Center for Science & Engineering Statistics

https://www.nsf.gov/statistics/data.cfm�Figure courtesy of Leslie McClure

Math & Statistics Degrees, US Citizens & Permanent Residents

% Women

7 of 52

Statistics

7

Plot: Amstat News 11/2020: https://magazine.amstat.org/blog/2020/11/01/degreegrowth/

2020 Figures: https://www.amstat.org/docs/default-source/amstat-documents/stattable1987-current.pdf

Statistics degrees awarded in the United States

2020 Figures

Bachelor: 42.9%

Master: 46.5%

PhD: 37.4%

8 of 52

ASA Salary Survey (2020)

8

20%

57%

35%

9 of 52

Canada

9

http://blogs.nature.com/news/2012/11/leaky-pipelines-for-canadian-women-in-research.html

10 of 52

Europe

10

Data from the Swiss Federal Statistical Office (FSO) and the European Commission

Figure: Piccoli, F., Guidobaldi, G. Swiss J Geosci 114, 1 (2021). https://doi.org/10.1186/s00015-020-00379-x

11 of 52

Industry

11

Figure: https://cen.acs.org/articles/96/i10/why-cant-the-drug-industry-solve-its-gender-diversity-problem.html

12 of 52

Publication metrics

12

West, JD, et al (2013). “The role of gender in scholarly authorship.” PLoS One.

Gender composition of authors in JSTOR (1900-2011)

Male authors

Female authors

Female author fraction

Female 1^st author fraction

Female last author fraction

13 of 52

Publication metrics

13

West, JD, et al (2013). “The role of gender in scholarly authorship.” PLoS One.

Self-citation of authors in JSTOR (1900-2011)

14 of 52

Editorial Boards: Math

14

Topaz & Sen. Gender representation on journal editorial boards in the mathematical sciences. PLoS One. 2016

15 of 52

Editorial Boards: Statistics

15

Ghosh-Dastidar et (2016). “Gender Balance in ASA Activities Update” Amstat News.

Overall: 35%

16 of 52

Editorial Boards: Stat/Biostat

16

Foulkes, A (2019). “Study reveals gender imbalance on (bio)statistics editorial boards” Amstat News.

17 of 52

SSC Gold Medal (1985-2021)

17

18 of 52

ASA Awards

18

Ghosh-Dastidar et (2016). “Gender Balance in ASA Activities Update” Amstat News.

COPPS Distinguished Achievement Lectureship

___________

19 of 52

19

How bad was it?

20 of 52

Some good news…

20

Ghosh-Dastidar et (2016). “Gender Balance in ASA Activities Update” Amstat News.

21 of 52

Looking back

21

Constance Van Eeden (1927-2021)

SSC Gold Medalist (1990)
Fellow of ISI, ASA
Pioneer in inference for restricted parameter spaces and nonparametric inference
UBC - Visiting honorary professor 1989-2012
Research career spanned >50 years

https://ssc.ca/en/publications/ssc-liaison/vol-35-6-december-2021/memoriam-constance-van-eeden-1927-2021

22 of 52

Van Eeden’s life

German occupation of Netherlands WWII
First Dutch woman to earn PhD in Statistics (1958)

Second came 29 years later

US Universities denied her tenure on the basis of sex / spousal considerations
Why the lack of Dutch women with PhDs?

discouragement from parents / teachers
lack of role models

Credits her success to parents “who just were absolutely convinced that we all had the right to the same chance in life”

22

https://ssc.ca/en/profile/conversation-constance-van-eeden

23 of 52

Another pioneer

Grace Wahba (b. 1934)

COPSS Distinguished Achievement Award and Lectureship (2014)
Fellow of AAAS, ASA, IMS
Groundbreaking methods in smoothing noisy data
Faculty at UW Madison 1967-2018

23

Nychka, D., Ma, P., & Bates, D. (2020). A conversation with Grace Wahba. Statistical Science, 35(2), 308-320.

24 of 52

Wahba’s Life

Bachelor’s degree from Cornell – women had restricted privileges
Industry job – experienced pay discrimination
Single mother – earned MS (Maryland) and PhD (Stanford) part time while working
Attributes some successes to ‘lucky breaks’
Have things gotten better for women in statistics?

much improvement in equality of opportunity
substantial gender disparity in income

24

Nychka, D., Ma, P., & Bates, D. (2020). A conversation with Grace Wahba. Statistical Science, 35(2), 308-320.

25 of 52

25

Why is the pipeline still leaking?

26 of 52

Are women leaving the field because they are behind in metrics?

OR

Are women behind in metrics because they are leaving the field?

26

27 of 52

Implicit bias - resumes

27

Moss-Racusin et al. Science faculty’s subtle gender biases favor male students. PNAS 2012

28 of 52

Implicit bias – teaching evals

28

MacNell, et al. What’s in a Name: Exposing Gender Bias in Student Ratings of Teaching. Innov High Educ, 2015.

Gender bias in teaching evaluations�“What’s in a Name? Exposing Gender Bias in Student Ratings of Teachers”�https://link.springer.com/article/10.1007%2Fs10755-014-9313-4

Innovative Higher Education | December 5, 2014

In this study, Lillian MacNell, Adam Driscoll, and Andrea N. Hunt found that students rated online instructors higher when the instructors were identified as male. The study focused on two professors (one male, one female), who each taught two sections of an online course, presenting to one group of students with a male-sounding name and the other with a female-sounding name. According to the authors, “The results were astonishing. Students gave professors they thought were male much higher evaluations across the board than they did professors they thought were female, regardless of what gender the professors actually were.”

�Gender bias in teaching evaluations�“What’s in a Name? Exposing Gender Bias in Student Ratings of Teachers”�https://link.springer.com/article/10.1007%2Fs10755-014-9313-4

Innovative Higher Education | December 5, 2014

In this study, Lillian MacNell, Adam Driscoll, and Andrea N. Hunt found that students rated online instructors higher when the instructors were identified as male. The study focused on two professors (one male, one female), who each taught two sections of an online course, presenting to one group of students with a male-sounding name and the other with a female-sounding name. According to the authors, “The results were astonishing. Students gave professors they thought were male much higher evaluations across the board than they did professors they thought were female, regardless of what gender the professors actually were.”

MacNell, et al. Innov Higher Ed 2015

�Assistant instructors in an online class each operated under two different gender identities

Students (n = 43) submitted ratings of their teachers at the end of the class

��Results:

Students rated the male identity significantly higher than the female identity, regardless of the instructor’s actual gender (and the fact that the female was a (slightly) better teacher)

29 of 52

Implicit bias - mentorship

29

Milkman, K.L, et al. What happens before? A field experiment exploring how pay and representation differentially shape bias on the pathway into organizations. Journal of Applied Psychology, 2015

Bias for white men in mentorship and faculty responsiveness�“What Happens Before? A Field Experiment Exploring How Pay and Representation Differentially Shape Bias on the Pathway into Organizations”�Journal of Applied Psychology | December 13, 2014

Milkman, et al. J Applied Psych 2014

�Faculty (n = 6,500) received meeting requests from fictional prospective doctoral students who were randomly assigned identity-signaling names (Caucasian, Black, Hispanic, Indian, Chinese; male, female)

Researchers recorded whether the faculty member replied within a week

Results:

All underrepresented groups studied experienced lower response rates than Caucasian males

This study found that when letters of interest were sent to professors in different disciplines (business, education, life sciences, humanities, social sciences, fine arts, etc.), the responsiveness of faculty varied according to race and gender. In all fields but the fine arts, letters signed with names that tend to be associated with white men (Brad Anderson as opposed to Keisha Thomas, Raj Singh, Mei Chen, or Juanita Martinez) were more likely to receive a response from a faculty member. Among business professors, where the disparities were the most extreme, responses to letters signed with names associated with white men increased 25 percentage points as compared to letters signed with names that signified females or males from underrepresented minorities (87% compared to 62%).

�

30 of 52

Why don’t I see it?

“Relative reluctance among men, especially faculty men within STEM, to accept evidence of gender biases in STEM” - Handley, et al. (PNAS, 2015)
Extent of bias typically relatively small
Implicit – unintentional, unconscious

Take the test!

https://implicit.harvard.edu/implicit/takeatest.html

30

31 of 52

Mentorship

31

32 of 52

32

Perceived reasons why women leave STEM careers

33 of 52

Child-rearing

33

Adapted From: Cech, EA; and Blair-Loy, M (2019). “The changing career trajectories of new parents in STEM” PNAS.

34 of 52

Discrimination & Sexual Harassment

34

Funk, C and Parker, K (2018). “Women and men in STEM often at odds over workplace equity” Pew Research Center Report.

35 of 52

Sexual misconduct - ASA

35

ASA 2018 study of member experiences of sexual- and gender-based misconduct at ASA events:

https://www.amstat.org/docs/default-source/amstat-documents/langer-research-final-report.pdf?sfvrsn=5ceb51d9_3/

Figure Courtesy of Leslie McClure

“Among women, 33 percent reported at least one incident of sexually inappropriate behavior or gender-based discrimination at one or more ASA events”

36 of 52

Lack of role models

36

Back cover of Molecular Biology of the Cell textbook, 6^th edition (2015)

37 of 52

Impostor phenomenon

37

Image source: Shivani Gopal; https://sundaskhalid.medium.com/5-stratagies-for-managing-imposter-syndrome-858699c10f3e

38 of 52

Poll – have you ever felt …?

... like your success is a result of luck or error
… like people around you are smarter than you
… the need to over-prepare or work much harder than others to “make sure” that nobody finds out you are a fraud

38

39 of 52

IP impacts women differently

39

Cokley,. et al. The Roles of Gender Stigma Consciousness, Impostor Phenomenon and Academic Self-Concept in the Academic Outcomes of Women and Men. Sex Roles (2015).

sample of U.S. undergraduate women and men in the Southwest (345 women, 146 men).

In the words of the study author:

“even though there were no differences between men and women in terms of their imposter feelings, imposter feelings had no link to GPA for men. It did not impact their grades at all, but for women imposter feelings were in fact linked in clear (negative) ways to their GPA. So you can say on the one hand that there were no differences in the reporting of imposter feelings amongst both groups, but the implications of those imposter feelings were quite different for men and women.”

Cokley, K., Awad, G., Smith, L. et al. The Roles of Gender Stigma Consciousness, Impostor Phenomenon and Academic Self-Concept in the Academic Outcomes of Women and Men. Sex Roles 73, 414–426 (2015). https://doi.org/10.1007/s11199-015-0516-7

40 of 52

COVID-19 impacts

40

Viglione G. Are women publishing less during the pandemic? Here's what the data say. Nature. 2020

41 of 52

41

42 of 52

Beyond gender

42

NSF: National Center for Science & Engineering Statistics

https://www.nsf.gov/statistics/data.cfm�Figure courtesy of Leslie McClure

43 of 52

Beyond gender

43

NSF: National Center for Science & Engineering Statistics

https://www.nsf.gov/statistics/data.cfm�Figure courtesy of Leslie McClure

44 of 52

44

Where do we go from here?

45 of 52

NSERC Graduate awards

45

https://www.nserc-crsng.gc.ca/_doc/Students-Etudiants/2020StatsScholarships_e.pdf

roughly equal!

46 of 52

NSERC Discovery Grants

46

https://www.nserc-crsng.gc.ca/_doc/DGP2020_e.pdf

16%

35%

25%

20%

roughly equal!

47 of 52

ASA Award Success Rates

47

Ghosh-Dastidar et (2016). “Gender Balance in ASA Activities Update” Amstat News.

48 of 52

48

49 of 52

Mentorship

49

https://cwstat.org/membership/available-memberships/

50 of 52

Networking

50

51 of 52

Summary

BE AWARE!

implicit biases

APPLY!

Scholarships, fellowships, grants

NOMINATE!

Awards, leadership positions, committees, editorial boards

MENTORSHIP!
ZERO TOLERANCE for harassment!

51

52 of 52

Acknowledgements

Leslie McClure

Professor of biostatistics at the Drexel University School of Public Health
Chair of ASA Task Force on Sexual Harassment and Assault

Anne Carpenter

PI at Broad Institute – image analysis for cell biology & AI for drug development
AMSB Women in Cell Biology Mid-career Award - awareness of implicit biases in STEM

52