Online Program

Return to main conference page
Wednesday, May 29
Data Visualization
Computational Statistics
Machine Learning
Opening Mixer & E-Posters
Wed, May 29, 5:30 PM - 7:00 PM
Grand Ballroom Foyer

Predicting Matriculation Rates of Dual Enrollment High School Students (306349)


*Benjamin Kenneth Brown, Oregon Institute of Technology 

Keywords: Decision tree classification, college, enrollment, accurate, prediction, matriculation rate, data analysis

The office of Educational Partnerships and Outreach (EPO) at Oregon Institute of Technology (Oregon Tech) has two primary programs that allow students to enroll in college classes while attending high school, this population significantly contributes to total enrollment especially in the last four years. EPO wanted to target marketing to these unique populations and predict the potential for future enrollment after completion of these programs. Furthermore, analysis was performed to determine what aspects of these programs may contribute to greater matriculation rates to guide the future growth of EPO and Oregon Tech.

Analysis was performed on a dataset comprising the last eight years of students in these programs and a decision tree model was created to predict students’ attendance at Oregon Tech. Through data analysis it was found that students who live within approximately 100 miles of Klamath Falls are more likely to attend Oregon Tech. The decision tree model generated predictions using 10-fold cross validation with accuracy ranges between 75.81% to 82.58%, and had a 92.58% accuracy in predicting the test set. The data analysis is useful in providing information on how EPO should develop their program in the coming years, and the decision tree provides clear guidance that we can use to predict which students will matriculate with solid accuracy. These predictions will be used by Oregon Tech Admissions to inform their decisions on which students and high schools they will focus on to encourage and recruit for admission and enrollment to Oregon Tech.