Development of a Multivariate Poisson Hidden Markov Model for Application in Educational Data Mining
Abstract
Managers and policymakers in higher education institutions try to improve graduation rates and decrease halt rates. To achieve this goal, it is important to understand academic and demographic factors that correlate with academic performance. Many studies in the field of education analytics have identified student grade point averages (GPA) as an important indicator and predictor of final academic outcomes (graduating or halting their studies). While semester-to-semester fluctuations in GPA are considered normal, significant changes in academic performance may warrant more thorough investigation and consideration, particularly with regard to final academic outcomes. However, it is challenging to represent complex academic trajectories over an academic career. To overcome this challenge, in this dissertation two different Hidden Markov Models (HMMs) are developed to provide a standard and intuitive classification over students' academic-performance levels. This leads to a compact representation of academic-performance trajectories. Next, the relationship between different academic-performance trajectories and their correspondence to final academic success are explored. Based on student transcript data from the University of Central Florida, the proposed HMMs are trained using sequences of students' course grades for each semester to estimate the students' academic-performance levels. Through the HMMs, the analysis follows the expected finding that higher academic performance levels correlate with lower halt rates. However, in this dissertation, we identify many scenarios in which both improving or worsening academic-performance trajectories actually correlate to higher graduation rates. This counter-intuitive finding is made possible through the two proposed HMMs.
Notes
If this is your thesis or dissertation, and want to learn how to access it or for more information about readership statistics, contact us at STARS@ucf.edu
Graduation Date
2022
Semester
Spring
Advisor
Vela, Adan
Degree
Doctor of Philosophy (Ph.D.)
College
College of Engineering and Computer Science
Department
Industrial Engineering and Management Systems
Degree Program
Industrial Engineering
Format
application/pdf
Identifier
CFE0008951; DP0026284
URL
https://purls.library.ucf.edu/go/DP0026284
Language
English
Release Date
May 2022
Length of Campus-only Access
None
Access Status
Doctoral Dissertation (Open Access)
STARS Citation
Boumi, Shahab, "Development of a Multivariate Poisson Hidden Markov Model for Application in Educational Data Mining" (2022). Electronic Theses and Dissertations, 2020-2023. 980.
https://stars.library.ucf.edu/etd2020/980