Open Collections

Predicting student outcomes using office referral data from a national sample of middle school students

University of British Columbia

This study examined the adequacy of using Office Discipline Referral (ODR) data to predict student behaviour outcomes. Specifically, the study identified rates of ODRs and student trajectories in a middle school population and assessed whether end-of-year behaviour outcomes can reliably be predicted as early as the first few months of Grades 6, 7, and 8. Participants were 401,852 students from 593 public middle schools (serving Grades 6 to 8) in the United States whose ODR data had been entered in the School-wide Information System (SWIS, 2012) during the 2009-2010 school year. ODRs were categorized by final ODR cut points used in positive behaviour support systems (0-1, 2-5, and 6+ total ODRs per year). Descriptive analyses and multilevel multinomial logistic regression analyses were performed; Receiver Operating Characteristic (ROC) analyses were used to assess diagnostic accuracy. Results showed relatively stable mean increases in ODRs throughout the school year for students in each final ODR category, though median growth trajectories revealed a comparatively large increase in ODRs received in November and February. Results also showed that ODRs received in September, October, and November were statistically significant predictors of final behaviour outcomes, and the inclusion of types of referrals (especially for defiance) significantly improved prediction of the final ODR category. These results are discussed with regards to previous and future research, limitations, and the utility of ODRs for intervention decision-making in middle schools.

Text

2013-01-09

Attribution-NonCommercial-NoDerivatives 4.0 International

http://hdl.handle.net/2429/43817

School Psychology

University of British Columbia

UBCV

http://creativecommons.org/licenses/by-nc-nd/4.0/