Name
Title | ||
|---|---|---|
SAT Does Not Spell Success: How Non-Cognitive Factors Can Explain Variance in the GPA of Undergraduate Engineering and Computer Science Students |
Abstract | ||
|---|---|---|
This work-in-progress research paper uses multiple regression of both cognitive and non-cognitive factors to model current GPA of engineering and computer science students. High school GPA and ACT/SAT scores are among the most common scores used as admission criteria, which result in a relatively homogeneous engineering population. Prior research, however, shows that these scores do not account for the variance in GPA once these students start undergraduate studies. In this work, we explore how students’ cognitive (e.g., study skills, test performance, regulatory behaviors, etc.) and non-cognitive factors (e.g., identity, motivation, personality, etc.) predict student success in their engineering pathways. The data for this initial study comes from a pilot survey, deployed in the summer of 2017, of 490 engineering and computing students from two large, public institutions, one on the West Coast, the other in the Midwest. We used multiple linear regression to control for demographic variables while examining the predictive value of particular cognitive and non-cognitive factors for student academic achievement (i.e., GPA) in university. Our analysis shows, not surprisingly, that standardized test scores only explain a small portion of the variance of undergraduate GPA. Including non-cognitive and affective factors into a regression model produced a marked increase in the explained variance. Our work is novel in examining a constellation of possible factors that predict undergraduate student GPA, by combining both cognitive and non-cognitive factors as predictors. This analysis begins to unpack particular factors that have potential to predict GPA of engineering and computer science students. |
Year | DOI | Venue |
|---|---|---|
2018 | 10.1109/FIE.2018.8658989 | 2018 IEEE Frontiers in Education Conference (FIE) |
Keywords | Field | DocType |
Predictive models,Analytical models,Computer science,Engineering education,Computational modeling,Linear regression | Study skills,Population,Grading (education),Regression analysis,Sociology,Standardized test,Engineering education,Knowledge management,Mathematics education,Academic achievement,Explained variation | Conference |
ISSN | ISBN | Citations |
0190-5848 | 978-1-5386-1174-6 | 0 |
PageRank | References | Authors |
0.34 | 0 | 5 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Matthew Scheidt | 1 | 0 | 0.34 |
| Ryan Senkpeil | 2 | 0 | 0.34 |
| john chen | 3 | 197 | 26.31 |
| Godwin, A. | 4 | 6 | 10.75 |
| Edward J. Berger | 5 | 1 | 4.40 |