Date of Award

5-10-2018

Degree Type

Thesis

Degree Name

Master of Public Health (MPH)

Department

Public Health

First Advisor

Brian Barger, PhD

Second Advisor

Emily Graybill, PhD

Third Advisor

Andrew Roach, PhD

Abstract

INTRODUCTION: There has been a demand for processes and appropriate tools to identify and route students with challenging behavior to the proper school-based supports. However, there is also a need for intuitive outcome measures that are feasible to be used in school settings.

AIM:In this study, we aim to assess the predictive validity of the Strengths and Difficulties Questionnaire (SDQ) using office discipline referrals (ODRs), a valid metric of student behavior.

METHODS:A series of non-parametric count models were used for analysis including Poisson (P), Negative Binomial (NB), Zero-Inflated Poisson (ZIP), and Zero-Inflated Negative Binomial (ZINB) models. These approaches were used to examine whether the internalizing and externalizing scales from the SDQ, administered at the beginning of the year, predicts total ODRs by the end of the year, even when controlling for the first 3 months of ODRs.

RESULTS: The ZINB model was chosen as the final model, as it had the best model fit (AIC). Our findings indicate that the SDQ’s internalizing and externalizing subscales are significant predictors of ODRs. Specifically, the internalizing scale was a significant negative predictor of total ODRs, while the externalizing scale was a significant positive predictor of total ODRs. Additionally, the externalizing scale is a significant negative predictor of excess zeros in ODRs.

DISCUSSION:Findings from our study suggest that the SDQ is a psychometrically valid tool with predictive utility in relationship to an outcome of interest in schools across an array of statistical approaches. Future studies should validate the SDQ with other samples.

Available for download on Tuesday, April 30, 2019

Share

COinS