Predicting Ninth Grade On-Track to Graduation Status With Mixed-Effect Random Forests

• In Event: Marriott Roundtable Session Four
  In Roundtable Session: Targeting Interventions to At-Risk Students in Schools (Table 1)

Abstract

A mixed-effects random forest (MERF) was applied to data contained in Oregon’s state longitudinal data system to predict the on-track to graduation status of 9th grade students (9G-OTG). Using a random subsample of student and school-level data obtained from six sequential cohorts (2013-14 to 2018-19; N = 33,256), we identified nonlinear relationships and several complex interactions between student and school characteristics and 9G-OTG status, correctly classifying 84% of the sample. In the paper, we demonstrate the use of modern data visualization techniques in the communication of findings and discuss how modern machine learning methods can assist school personnel in targeting interventions to students most at-risk of not graduating.

Authors

• Keith Zvoch, University of Oregon
  Presenting Author

• Christopher Loan, University of Oregon
  Non-Presenting Author

• Kathleen Scalise, University of Oregon
  Non-Presenting Author