Virtual Conference
July 26, 2021
July 26, 2021
July 19, 2022
Educational Research and Methods
24
10.18260/1-2--36575
https://peer.asee.org/36575
675
Danika Dorris is a Ph.D student in the Edward P. Fitts Department of Industrial and Systems Engineering at North Carolina State University. She received a Bachelor's of Science in Industrial and Systems Engineering from the University of Tennessee. Her work currently focuses on modeling wellbeing of emerging adults and college student attrition
Julie Swann is the department head and A. Doug Allison Distinguished Professor of the Fitts Department of Industrial and Systems Engineering. She is an affiliate faculty in the Joint Department of Biomedical Engineering at both NC State and the University of North Carolina at Chapel Hill. Before joining NC State, Swann was the Harold R. and Mary Anne Nash Professor in the Stewart School of Industrial and Systems Engineering at the Georgia Institute of Technology. There she co-founded and co-directed the Center for Health and Humanitarian Systems (CHHS), one of the first interdisciplinary research centers on the Georgia Tech campus. Starting with her work with CHHS, Swann has conducted research, outreach and education to improve how health and humanitarian systems operate worldwide.
Julie Simmons Ivy is a Professor in the Edward P. Fitts Department of Industrial and Systems Engineering and Fitts Faculty Fellow in Health Systems Engineering. She previously spent several years on the faculty of the Stephen M. Ross School of Business at the University of Michigan. She received her B.S. and Ph.D. in Industrial and Operations Engineering at the University of Michigan. She also received her M.S. in Industrial and Systems Engineering with a focus on Operations Research at Georgia Tech. She is President of the Health Systems Engineering Alliance (HSEA) Board of Directors. She is an active member of the Institute of Operations Research and Management Science (INFORMS), Dr. Ivy served as the 2007 Chair (President) of the INFORMS Health Applications Society and is a past President for the INFORMS Minority Issues Forum. Her research interests are mathematical modeling of stochastic dynamic systems with emphasis on statistics and decision analysis as applied to health care, public health, education and humanitarian logistics.
This is a research paper focused on identifying influential factors of student dropout. Students who drop out of college can suffer negative effects on their wellbeing long after they leave college. Many of these dropout students are dropping out in the first two years which highlights the importance of identifying at-risk students early on. We analyzed data for the 1,754 students in the 2014 undergraduate engineering cohort at a large public university. Of these 1,754 students, 12.6 percent dropped out. Ninety-two factors describe each student’s application information (e.g., SAT), academic metrics, (e.g., course load, GPA), and demographic information (ethnicity, age). Defining characteristics and significant predictors of at-risk students were identified using three types of analyses: (i) statistical testing for comparisons, (ii) cluster analysis, and (iii) logistic regression predictions. We first identify significant differences between graduate and dropout populations with hypothesis testing. Then, we use clustering to identify subgroups within the cohort and label each group according to a set of defining characteristics. Lastly, significant predictors are extracted from a logistic regression model predicting eventual dropout. Statistical testing for comparisons found that there was a lower proportion of female and full-time students in the dropout population than those who graduated. Most dropout students formed a separate cluster from the rest of the cohort, and the time of dropout influenced the clusters formed within the dropout population. From the regression models, we learn that GPA and passed credits are significant predictors in the first year, and race does not become a significant predictor of dropout until the second year. The factors that influence dropout change over time which emphasize the importance of dynamic dropout prediction models. The findings from each phase of our analysis highlight the complexity of understanding the causes of dropout and the importance of personalizing interventions for specific populations within a cohort.
Dorris, D. M., & Swann, J. L., & Ivy, J. (2021, July), A Data-driven Approach for Understanding and Predicting Engineering Student Dropout Paper presented at 2021 ASEE Virtual Annual Conference Content Access, Virtual Conference. 10.18260/1-2--36575
ASEE holds the copyright on this document. It may be read by the public free of charge. Authors may archive their work on personal websites or in institutional repositories with the following citation: © 2021 American Society for Engineering Education. Other scholars may excerpt or quote from these materials with the same citation. When excerpting or quoting from Conference Proceedings, authors should, in addition to noting the ASEE copyright, list all the original authors and their institutions and name the host city of the conference. - Last updated April 1, 2015