Recipient Organization
OHIO STATE UNIVERSITY
1680 MADISON AVENUE
WOOSTER,OH 44691
Performing Department
Agriculture, Environment and Development Economics
Non Technical Summary
As of 2016, the US gender pay gap remained at 20%. Even accounting for differences in education, work experience, occupation, and collective bargaining coverage, the gap remains substantial at 8.4%. Moreover, progress on narrowing the gender pay gap has stalled and possibly even reversed. Universities are no exception when it comes to the gender pay gap. Male full professors at US institutions earned 15% more than female full professors in 2014, while UK universities had a gender gap of 10.5% across ranks in 2015-16.We quantify the gender pay gap at OSU and assess the key contributing factors. In particular, we examine differences in research productivity, tenure and promotion, teaching and service, as well as differential effects of marriage and children. We also conduct comparisons with other institutions, and assess which types of policies may be more effective for addressing pay disparities.
Animal Health Component
90%
Research Effort Categories
Basic
10%
Applied
90%
Developmental
(N/A)
Goals / Objectives
As of 2016, the US gender pay gap remained at 20%. Even accounting for differences in education, work experience, occupation, and collective bargaining coverage, the gap remains substantial at 8.4% (Blau and Kahn, 2017). Moreover, progress on narrowing the gender pay gap has stalled and possibly even reversed (AAUW, 2017). Universities are no exception when it comes to the gender pay gap. Male full professors at US institutions earned 15% more than female full professors in 2014 (Hatch, 2017), while UK universities had a gender gap of 10.5% across ranks in 2015-16 (Holmes, 2017).And yet, research on the gender pay gap, particularly in academia, remains quite slim. So-called "leaky pipelines" have been observed in many disciplines, with higher exit rates among women, beginning as early as the undergraduate level (Allen-Hermanson, 2017; Levenstein, 2015). And there is evidence of entrenched barriers unique to academia. Women are held to higher standards in the peer review process (Hengel, 2017). Co-authored publications are more heavily discounted for women (Sarsons, 2017). Gender-neutral "clock-stopping" policies (which extend the probationary period for child birth/adoption) reduce the likelihood that women receive tenure, relative to their male counterparts (Antecol et. al., 2016). But, overall, our understanding is very limited when it comes to the fundamental mechanisms underlying the gender pay gap. And, more importantly, policies that can effectively address the disparity still need to be identified.The goals of this research are to quantify the gender pay gap at U.S. universities and to identify the factors that contribute to the disparity as well as policies that can help ameliorate the gap. Specific objectives include:Estimate the gender pay gap for regular tenure-track faculty members.Identify appropriate metrics to account for differences in productivity (teaching, research, and outreach activities) across individuals.Assess gender gaps in hiring and promotion.Account for movement of individuals across institutions and out of academia.Assess how the gender gap differs between departments of agricultural/applied economics and traditional economics.Assess how the gender pay gap differs between public universities and government agencies.
Project Methods
We combine publicly available salary and human resources data from U.S. universities with proprietary data on research productivity from Academic Analytics. To estimate the pay gap, we utilize regression and decomposition analysis. Regression analysis allows us to control for various characteristics in order to isolate the portion of the gap due to institutional, rather than individual, factors. For example, women tend to be under-represented in finance, which tends to have exceptionally high salaries. Therefore, we would prefer to compare men's and women's salaries conditional on academic unit, in order to account for the fact that there are underlying differences across disciplines.However, this approach assumes that these underlying differences are, themselves, independent of gender. While this may be plausible with regard to disciplines (though we might also question why disciplines with lower female representation tend to be paid more), when we consider faculty rank, this assumption is problematic when there are gender disparities in promotion and tenure as well. Instead, we utilize an approach common in the literature on wage disparities - decomposition analysis (Oaxaca, 1973; Blinder, 1973). In effect, this approach predicts what women would earn, based on their own characteristics and qualifications, if they were paid according to the same implicit salary scale as men. Then, the difference between that predicted value and what women actually earn can be taken as an estimate of gender discrimination. By allowing the salary scale to differ for men and women for various factors (e.g., experience, rank), we can further isolate the proportion of the gender pay gap attributable to each factor.Additional data will be collected for faculty in departments of agricultural/applied economics, as well as data on government employees at the state and federal level.