Advancement of Women

through the Academic Ranks:

Where are the Leaks in the Pipeline?

Interim report to the Columbia University Senate

on a study by the

Commission on the Status of Women

April 27, 2001

Presenter: Kim Kastens

Commission Chair: Jean Howard

Staff Support: Lucy Drotning

• Attrition of graduate students (PhD track)

- in the first year

- after seven years

• Hiring into the tenure-eligible ranks

• Entry into the tenured ranks

- by promotion

- by external hire

• Ten year trends & importance of microclimates

(all within the Graduate School of Arts & Sciences)

First Year Attrition of Doctoral Students

(Students who entered fall of 1999; Status as of fall 2000)

Humanities

	Women	Men
# entered	59	54
# attrition	11	4
% attrition	19%	7%

Social Sciences

	Women	Men
# entered	52	59
# attrition	7	3
% attrition	13%	5%

Natural Sciences

	Women	Men
# entered	34	79
# attrition	5	3
% attrition	15%	4%

All Divisions

	Women	Men
# entered	145	192
# attrition	23	10
% attrition	16%	5%

Ultimate Attrition of Doctoral Students

(Students who entered fall of 1993; Status as of fall 2000)

Humanities

	Women	Men
# entered	96	60
# attrition^*	44	21
% attrition	46%	35%

Social Sciences

	Women	Men
# entered	66	82
# attrition^*	29	32
% attrition	44%	39%

Natural Sciences

	Women	Men
# entered	40	65
# attrition^*	13	18
% attrition	33%	28%

All Divisions

	Women	Men
# entered	202	207
# attrition^*	86	71
% attrition	43%	34%

* students neither graduated nor registered, seven years post-matriculation

Influence of Funding Status on Attrition

(Students who entered fall of 1993; Status as of fall 2000)

Humanities

	Women	Men
# entered	96	60
# funded	37	31
% funded	39%	52%
% attrition^* among funded students	32%	19%
% attrition^* among unfunded students	54%	52%

Social Sciences

	Women	Men
# entered	66	82
# funded	29	28
% funded	44%	34%
% attrition^* among funded students	28%	25%
% attrition^* among unfunded students	57%	46%

· students neither graduated nor registered, seven years post-matriculation

Recommendations Concerning Graduate Student Attrition

• Ask students leaving the PhD program why they are doing so, in either an exit interview or a survey questionnaire. A standard set of questions should be asked, and the data should be tabulated in a way that can be compared across divisions and across cohorts.

• Examine the support structure for first-year PhD students, including orientation activities, first year course of study, and the mechanism for matching student with advisor.

• Recompile the data on graduate student attrition into a form that will allow individual entering cohorts to be tracked longitudinally.

• Examine longitudinal data for pressure points: are there points in the student trajectory where attrition of female students preferentially occurs, cohort after cohort?

• Compare longitudinal data across cohorts for evidence of change over time: is there evidence that female students in more recent cohorts are faring better than the cohorts from 5 or 10 years ago?

New Hires for Tenure-eligible Ranks

	1999-2000		1996-1998
	Current Tenure-Elig. Faculty	New Hires	Current Tenure-Elig. Faculty	New Hires
Total Arts & Sciences	33%	34%	28%	32%
Humanities	43%	45%	40%	46%
Social Sciences	36%	48%	24%	32%
Natural Sciences	23%	16%	20%	19%

	1993-1995		1990-1992
	Current Tenure-Elig. Faculty	New Hires	Current Tenure-Elig. Faculty	New Hires
Total Arts & Sciences	30%	31%	31%	29%
Humanities	42%	63%	44%	37%
Social Sciences	27%	25%	29%	31%
Natural Sciences	17%	11%	13%	16%

New Hires for Tenure-eligible Ranks

1999-2000

	% Women
	Current Ten-El. Faculty	New Hires		Applicant Pool		National Availability Pool
Total Arts & Sciences	33%	34%	>	23%	<	43%
Humanities	43%	45%	<	48%	<	51%
Social Sciences	36%	48%	>	31%	<	42%
Natural Sciences	23%	16%	>	14%	<	39%

• Women applicants for Columbia tenure-eligible positions are successful in the competition for those jobs, disproportionately successful.

• However, women are not represented within the Columbia applicant pool in proportion to their abundance in the national availability pool.

Why are women under-represented in our applicant pools?

Hypotheses to explore:

(1) New York City may be off-putting to some women (for example, women with children).

(2) Columbia’s long tradition as an educator and employer of men may be off-putting to some women.

(3) Women may tend to underestimate their own accomplishments and thus be less likely to apply for positions for which they are marginally qualified. (Or conversely, men may tend to overestimate….)

(4) Advisors / mentors at other universities may be, on average, more likely to encourage men to apply for a challenging job than women.

(5) For candidates who are discovered through networking and encouraged to apply: Male senior faculty may preferentially ask for tips from male colleagues, and/or male faculty elsewhere may preferentially recommend male candidates.

(6) Some women may choose to confine their job search to “easier” jobs which offer the possibility of a better work-life balance.

(7) On average, job-hunting women may apply for fewer jobs. This could be the case, for example, for women who constrain their job search geographically.

Entry to the Tenured Ranks

1990 – 2000

	Women/Total
	Tenured Faculty	Internal Promotions	External Hires	Targets^* of Opportunity
Total Arts & Sciences	1990: 13% 2000: 20%	33% (29/88)	22% (19/87)	27% (8/30)
Humanities	1990: 17% 2000: 25%	40% (15/38)	40% (10/25)	50% (5/10)
Social Sciences	1990: 13% 2000: 23%	40% (10/40)	19% (7/36)	33% (3/9)
Natural Sciences	1990: 8% 2000: 11%	16% (4/21)	8% (2/24)	0% (0/11)

* Applicant pool of one person, in Affirmative Action records

Ten year trend for English: Faculty numbers

(relabeled “A bad example”)

Ten year trend for English: Students & Faculty percentages

(label covered)

Ten year trend for Econ: Faculty numbers

(relabeled: “A good example”)
Change in Representation of Women on Faculty

(1990 – 2000)

Improved substantially:

Anthropology

Classics

Economics

Italian

Philosophy

Psychology

Sociology

Least Improved:

English & Comparative Literature

Germanic Languages

Middle East & Asian Languages & Culture

Recommendations Concerning Faculty Pipeline

• Investigate hypotheses concerning low representation of women in Columbia’s tenure-eligible applicant pools: comparison with other elite universities, other NYC universities.

• Recruit women pro-actively for tenure-eligible positions. Scrutinize the make up of the applicant pool, as well as considering whether women in the applicant pool were fairly considered.

• Scrutinize external hires to tenured positions, especially targets of opportunity.

• Disseminate the department-by-department data on faculty and student gender balance over time, 1990-2000, to allow individual departments to evaluate their own standing and progress.

• Investigate the gender balance of fluxes out of the faculty: by retirement/death and for other jobs, at tenured and tenure-eligible levels.

• Extend pipeline study to other Schools.