Theses and Dissertations - UTB/UTPA

Date of Award


Document Type


Degree Name

Master of Science (MS)



First Advisor

Dr. Xiaohui Wang

Second Advisor

Dr. George Yanev

Third Advisor

Dr. Maria Cristina Villalobos


In this thesis, we study the time-to-graduation problem for STEM Hispanic undergraduate students. The response, time-to-graduation, was treated in two different ways: as a binary variable with graduated (by the 6th year) and not-graduated values, and as an ordinal variable with values year-4, year-5, year-6, and not-graduate. Mathematics education plays critical role in students’ timely graduation, especially for STEM students. We used students records data obtained from The University of Texas-Pan American to illustrate how mathematics background factors (including SAT math score, ACT math score, TASP math score) and mathematics performance variables (including mathematics GPA, number of dropped mathematics courses, number of repeated math courses), and some demographic factors (including gender, full-/part-time study status) are related to time-to-graduation. Logistic regression and backwards subset selection were employed to determine the significant variables and then form the model for the prediction of timely graduation. In addition, random forests method was used to predict the ordinal outcome, the year when the students graduated.


Copyright 2014 Gejun Zhu. All Rights Reserved.

Granting Institution

University of Texas-Pan American