Zahedi, L., & Lunn, S. J., & Pouyanfar, S., & Ross, M. S., & Ohland, M. W. (2020, June), Leveraging Machine Learning Techniques to Analyze Computing Persistence in Undergraduate Programs  Paper presented at 2020 ASEE Virtual Annual Conference Content Access, Virtual On line . 10.18260/1-2--34921

\bibitem{asee_peer_34921}
  Zahedi, L., \& Lunn, S. J., \& Pouyanfar, S., \& Ross, M. S., \& Ohland, M. W. (2020, June), \emph{Leveraging Machine Learning Techniques to Analyze Computing Persistence in Undergraduate Programs}
  Paper presented at 2020 ASEE Virtual Annual Conference Content Access, Virtual On line .
  10.18260/1-2--34921

Leila Zahedi, Stephanie J. Lunn, Samira Pouyanfar, Monique S. Ross, and Matthew W. Ohland.  "Leveraging Machine Learning Techniques to Analyze Computing Persistence in Undergraduate Programs".  2020 ASEE Virtual Annual Conference Content Access, Virtual On line , 2020, June.  ASEE Conferences, 2020.  https://sftp.asee.org/34921 Internet.  26 Apr, 2024

\bibitem{asee_peer_34921}
  Leila Zahedi, Stephanie J. Lunn, Samira Pouyanfar, Monique S. Ross, and Matthew W. Ohland.
  "Leveraging Machine Learning Techniques to Analyze Computing Persistence in Undergraduate Programs".
  \emph{2020 ASEE Virtual Annual Conference Content Access, Virtual On line , 2020, June}.
  ASEE Conferences, 2020.
  https://sftp.asee.org/34921 Internet.  26 Apr, 2024

@INPROCEEDINGS{asee_peer_34921
author = "Leila Zahedi, Stephanie J. Lunn, Samira Pouyanfar, Monique S. Ross, and Matthew W. Ohland"
title = "Leveraging Machine Learning Techniques to Analyze Computing Persistence in Undergraduate Programs"
booktitle = "2020 ASEE Virtual Annual Conference Content Access"
year = "2020"
month = "June"
address = "Virtual On line "
publisher = "ASEE Conferences"
note = {https://sftp.asee.org/34921}
number = {10.18260/1-2--34921}
}

TY  - CPAPER
AB  - Student retention in computing fields remains one of the most significant concerns for science, technology, engineering, or mathematics careers. Looking further at computing fields, it is clear that enrollment in such programs has not kept pace with the industry demands. This is in part due to retention challenges in computer science. Retention refers to the number of students who start their career at a college and will graduate. Many factors may play a role in a students’ decision to drop out. Thus, finding meaningful patterns from historical data can help education researchers reveal the possible reasons for students’ withdrawal from a university and provide guidelines and mechanisms that lead to improving retention rates in such fields. Since pathway patterns for computing students are different from engineering students, it is crucial to specifically explore computing fields in order to find ways to increase the retention rate for such majors. To achieve this goal, this study is conducted on a large longitudinal database - Multiple-Institution Database for Investigating Engineering Longitudinal Development (MIDFIELD) to determine the importance of different factors (features) in the graduation of computing students. The results of this work provides a predictive model of graduation, across multiple U.S. institutions leveraging various machine learning (ML) techniques (including: random forest, support vector machine, naïve bayes, and decision tree classification) to find the most important factors, as well as the most accurate predictive model for graduation. Participants of this study included approximately 40,000 students who enrolled in computing majors and had the opportunity to graduate within six years. The identified students, at some point, were enrolled in one of the following computing disciplines; namely computer engineering, software engineering, computer science, computer programming, computing and information sciences, and information technology. Drawing on Astin’s inputs-environment-outcome (I-E-O) model, we first explored the importance of each feature on students’ graduation. According to the model, student outcomes (e.g., graduation) are dependent on two types of factors, including inputs (e.g., demographic features) and environment (e.g., experiences in college). Results suggest that educational variables, including cumulative GPA, number of terms registered, being a transfer student, and the institution, are the most important features, respectively. Results also showed that random forest produced a more accurate result on this dataset compared to other predictive models. We anticipate findings from this ongoing research will give insight to the computing education community and researchers to better understand the relative success of computing students, in turn providing strategic solutions to attain higher retention rates.
AU  - Leila Zahedi
AU  - Stephanie J. Lunn
AU  - Samira Pouyanfar
AU  - Monique S. Ross
AU  - Matthew W. Ohland
CY  - Virtual On line 
DA  - 2020/06/22
PB  - ASEE Conferences
TI  - Leveraging Machine Learning Techniques to Analyze Computing Persistence in Undergraduate Programs
UR  - https://sftp.asee.org/34921
DO  - 10.18260/1-2--34921
ER  -