Using random forest analysis to identify student demographic and high school-level factors that predict college engineering major choice

Li Tan, Joyce B. Main, Rajeev Darolia

Research output: Contribution to journalArticlepeer-review

6 Scopus citations

Abstract

Background: Given the importance of engineers to a nation's economy and potential innovation, it is imperative to encourage more students to consider engineering as a college major. Previous studies have identified a broad range of high school experiences and demographic factors associated with engineering major choice; however, these factors have rarely been ranked or ordered by relative importance. Purpose/Hypothesis: This study leveraged comprehensive, longitudinal data to identify which high school-level factors, including high school characteristics and student high school experiences as well as student demographic characteristics and background, rank as most important in terms of predictive power of engineering major choice. Design/Method: Using data from a nationally representative survey, the High School Longitudinal Study of 2009, and the random forest method, a genre of machine learning, the most important high school-level factors in terms of predictive power of engineering major choice were ranked. Results: Random forest results indicate that student gender is the most important variable predicting engineering major choice, followed by high school math achievement and student beliefs and interests in math and science during high school. Conclusions: Gender differences in engineering major choice suggest wider ranging cultural phenomena that need further investigation and systemic interventions. Research findings also highlight two other areas for potential interventions to promote engineering major choice: high school math achievement and beliefs and interests in math and science. Focusing interventions in these areas may lead to an increase in the number of students pursuing engineering.

Original languageEnglish
Pages (from-to)572-593
Number of pages22
JournalJournal of Engineering Education
Volume110
Issue number3
DOIs
StatePublished - Jul 2021

Bibliographical note

Funding Information:
This material is based upon the work supported by the National Science Foundation (Awards 1532015, 1745287, and 1531920). Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation. The authors also give special thanks to Cory Koedel for his support on this project and the anonymous reviewers and editors for helpful feedback on the article. Journal of Engineering Education 1

Publisher Copyright:
© 2021 American Society for Engineering Education.

Keywords

  • analytics
  • diversity
  • engineering major choice
  • engineering pathways
  • high school
  • quantitative

ASJC Scopus subject areas

  • Education
  • Engineering (all)

Fingerprint

Dive into the research topics of 'Using random forest analysis to identify student demographic and high school-level factors that predict college engineering major choice'. Together they form a unique fingerprint.

Cite this