Race, Gender, and Measures of Success in Engineering Education

B ackground Concern for workforce needs, social justice, and the diversification of the engineering profession make it critical to understand how different metrics may overestimate or underestimate the success of various race‐gender populations in engineering. P urpose (H ypothesis ) While earlier work found that women in nearly all racial groups persist to the eighth semester at rates comparable to men, results vary in studies that use other measures of success, providing an incentive to compare multiple measures of success in the same population. D esign /M ethod The eight‐semester persistence and six‐year graduation rates are compared for various race‐gender populations using a longitudinal, comprehensive dataset of more than 75,000 students matriculating in engineering at nine universities from 1988–1998. R esults Gender differences in persistence of Asian, Black, Hispanic, Native American, and White students are far outweighed by institutional differences. Racial differences are more pronounced, however, revealing some patterns that transcend institutional differences. C onclusion Our work demonstrates that trajectories of persistence are non‐linear, gendered, and racialized, and further that higher education has developed the way in which persistence is studied based on the behavior of the majority, specifically the White, male population. Even if institutions were to treat all students equally, the outcomes will not necessarily be the same because various populations respond differently to the same conditions. Using eight‐semester persistence may result in a “systematic majority measurement bias.” Therefore, multiple measures may be needed to describe outcomes in diverse populations.