Date of Award

Winter 2019

Document Type


Degree Name

Doctor of Education (EdD)


School of Education

First Advisor

Dane Joseph, PhD

Second Advisor

Scot Headley, PhD

Third Advisor

Susanna Thornhill, PhD


This study analyzed the predictive validity of key dropout indicators at the freshmen year within a rural school district. Specifically, the study examined the predictive validity of the freshmen on-track indicator and freshmen absenteeism as predictors of four year, on-time graduation attainment. While most of the Early Warning System (EWS) research has taken place in large urban and suburban schools districts, this study used secondary data spanning four years from a small, rural school district. Additionally, the study sought to explore which student academic and behavior metrics had the greatest predictive validity for the rural student sample. Binomial logistic regression was used to analyze the secondary data. The analysis found a statistically significant relationship between graduation attainment and three of the study’s variables (economically disadvantaged status, freshmen on-track status, and absenteeism). Developing effective dropout prediction models can assist educators in providing more timely and targeted interventions for potentially at-risk students. Additionally, the results of this study may provide additional insights into the predictive capabilities of these key student academic and behavioral early warning indicators with rural student samples.

Included in

Education Commons