Ensemble Learning for Estimating Individualized Treatment Effects in Student Success Studies
- PDF / 835,732 Bytes
- 21 Pages / 439.642 x 666.49 pts Page_size
- 105 Downloads / 157 Views
Ensemble Learning for Estimating Individualized Treatment Effects in Student Success Studies Joshua Beemer1 · Kelly Spoon1 · Lingjun He2 · Juanjuan Fan3 · Richard A. Levine2,3
© International Artificial Intelligence in Education Society 2017
Abstract Student success efficacy studies are aimed at assessing instructional practices and learning environments by evaluating the success of and characterizing student subgroups that may benefit from such modalities. We propose an ensemble learning approach to perform these analytics tasks with specific focus on estimating individualized treatment effects (ITE). ITE are a measure from the personalized medicine literature that can, for each student, quantify the impact of the intervention strategy on student performance, even though the given student either did or did not experience this intervention (i.e., is either in the treatment group or in the control group). We illustrate our learning analytics methods in the study of a supplemental instruction component for a large enrollment introductory statistics course recognized as a curriculum bottleneck at San Diego State University. As part of this application, we show how the ensemble estimate of the ITE may be used to assess
Richard A. Levine
[email protected] Joshua Beemer [email protected] Kelly Spoon [email protected] Lingjun He [email protected] Juanjuan Fan [email protected] 1
Computational Science Research Center, San Diego State University, San Diego, CA, USA
2
Analytic Studies and Institutional Research, San Diego State University, San Diego, CA, USA
3
Department of Mathematics and Statistics, San Diego State University, San Diego, CA, USA
Int J Artif Intell Educ
the pedagogical reform (supplemental instruction), advise students into supplemental instruction at the beginning of the course, and quantify the impact of the supplemental instruction component on at-risk subgroups. Keywords Educational data mining · Personalized learning · Machine learning · Regularized regression · Supplemental instruction
Introduction In striving to improve graduation rates and reduce achievement gaps, Universities have experimented with a suite of instructional practices and learning environments (for example, see the 2015 issue 2 of Peer Review from the Association of American Colleges & Universities). Broadly speaking, these strategies foster student success and engagement through common and collaborative intellectual experiences, student research and internships, and study abroad experiences (Kuh 2008) as well as supplemental instruction and instructional technologies (for example see Dawson et al. 2014; Henrie et al. 2015). An analytics goal is identifying at-risk students that will benefit from one or more of these intervention strategies and, early in their college careers, advise these students accordingly. On the flip side, we also must evaluate each instructional practice and each learning environment on at-risk subgroups for purposes of strategic planning, resource allocation, and program de
Data Loading...
