Date of Award


Document Type

Doctoral Dissertation

Degree Name

Education, Ed.D.



First Advisor

Jacqueline Waggoner

Third Advisor

Bruce Weitzel

LC Subjects

High schools--U.S. states--Oregon; High schools--Evaluation; High school graduates; Academic achievement


The purpose of this causal-comparative study was to compare graduation rates, defined as completing high school within five years, of students who learned online in Oregon virtual schools to students who attended traditional schools. The study utilized longitudinal data provided by the Oregon Department of Education connected to literature-supported graduation indicators to explore and compare graduation rates of traditional and virtual students. Analysis of data in the study was guided by three research questions: (1) when examining the ratio of the number of school days spent in a virtual school setting to the student graduation rates, does the increase of days in a virtual setting impact the likelihood of high school graduation, defined as completing 24 credits, (2) do predictive graduation indicators from the literature (i.e., attendance, behavior, demographics, performance, and mobility) explain the discrepancy, if any, in virtual and non-virtual Oregon five-year high school graduation rates, and (3) what combination of indicators is most useful for predicting a virtual student’s graduation outcome? The analysis used descriptive statistics and binary logistic regression to address the research questions. Key findings of the study included: attending a virtual school was not a good predictor of whether a student would graduate; behavior was not an effective indicator in virtual schools; it was easier to predict a graduate than a non-graduate; adding measures of mobility to previously studied indicators, attendance, behavior, and course performance, improved the ability to predict graduation outcomes for both traditional and virtual students; for virtual students, results of this study suggest that mobility may be a valuable substitute for behavior in a predictive model focused on students who attend school virtually.


Copyright for this work is retained by the author.