Back to search

West Virginia University

Using Machine Learning and Traditional Statistics to Explore Retention and Knowledge Structure in STEM with an emphasis on Physics

Abstract

dc:description.abstract

Retention of Science, Technology, Engineering, and Mathematics (STEM) students is a serious problem as STEM graduation rates continue to lag the growing demand for the skills taught by these degree programs. Critical to fixing this “leaky pipeline” is investment in improving retention in the first two years of college study and increasing and maintaining the interest of K-12 students in STEM. This thesis will address this in three parts. The first is through evaluation of conceptual tests used to evaluate course improvements to determine the structure student knowledge measured by them. The second part uses machine learning to construct early warning models of student failure in introductory physics courses to aid instructors in better targeting of interventions. The final part assesses the effectiveness of the Pulsar Search Collaboratory, an authentic science experience for middle and high school students, at encouraging K-12 students to pursue STEM degrees.

Degree

thesis:*
Name thesis:degree_name
PhD
Level thesis:degree_level
Dissertation
Discipline thesis:degree_discipline
Physics and Astronomy

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Zabriskie, Cabot Alexander

Subjects

dc:subject × 7

Identifiers

dc:identifier.*

Chain of custody

source
Harvested from
West Virginia University
Base URL
researchrepository.wvu.edu/do/oai/
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
citation

Zabriskie, Cabot Alexander. Using Machine Learning and Traditional Statistics to Explore Retention and Knowledge Structure in STEM with an emphasis on Physics. Dissertation thesis, https://doi.org/10.33915/etd.4118