Keywords: Undergraduate, Race/Ethnicity, Gender, Engineering
Abstract: We must increasingly engage and capitalize on the contributions of people from backgrounds underrepresented in engineering, especially women and people of color, if we are to educate enough engineers to meet demand and propel our nation’s competitiveness through an engineering workforce reflective of our nation’s diversity. This study focuses on broadening pathways into engineering, expanding both the diversity and size of the engineering student population. We hypothesized that engineering colleges’ over-reliance on standardized test scores in the admission process inadvertently denies admission to diverse students capable of becoming successful engineers. Using the Multiple-Institution Database for Investigating Engineering Longitudinal Development (MIDFIELD) database of 226,221 engineering students, an investigation of whether admission data supports using a singular combined threshold using both high school grade point average and standardized test scores, or whether the data suggests using another model for predicting success in engineering as measured by a six-year engineering graduation rate. Of the predictive models that spanned all 11 institutions, high school grade point average was the most influential in predicting six-year engineering graduation. The next influential variable in predicting six-year engineering graduation was the higher education institution or ethnicity—a finding that suggests that the best predictive admissions model is specific to an individual institution, not an across-institutional model. Standardized test score was the most significant predictor in only one of the 11 institutions when modeled separately and in three others after high school grade point average. In seven of the 11 institutions, test score was not found to be a significant predictor of six-year engineering graduation for underrepresented minority students. A better understanding of the admissions profile of each institution might help determine what other factors are at play. Other potential factors that come to mind are financial aid, first-generation college-attendance and socioeconomic status.
Are you a researcher? Would you like to cite this paper?
Visit the ASEE document repository at
for more tools and easy citations.