| dc.description.abstract | An extensive body of research has shown that data-driven Early Warning Systems (EWS) are an effective tool for reducing school dropout. EWS designed to prevent dropout flags at-risk students based on a core set of indicators, including attendance, behavior, and course failures. While preventing high school dropout is a critical step to ensuring student success, in today’s 21st century workforce, a high school diploma is not enough. To strategically align interventions and strategies designed to promote college readiness, school districts must be aware of the variables that predict postsecondary success. Unfortunately, there is a dearth of succinct college readiness tools that capture multiple longitudinal indictors that can be used to identify and flag students who are not ready to meet the rigorous demands of postsecondary education. To fill this gap, the present study evaluated the utility of the core EWS dropout indicators in predicting postsecondary success. This study examined longitudinal data from 7th through 12th grade to predict postsecondary outcomes from one moderately sized Midwestern school district. A series of path analyses was used to analyze retrospective data from approximately 3,080 public school students who entered 7th grade in the 2007-08 school year and had an original on-time graduation year of 2013. Results revealed a statistically significant temporal relationship among each of the key EWS variables (i.e., attendance, behavior, GPA, and state assessment scores) across the target six years. In terms of the predictive validity of the EWS indicators, the study found that 12th grade attendance rate, 12th grade GPA, and 7th grade GPA were statistically significant predictors of postsecondary enrollment. Free and reduced priced lunch status, special education status, and mobility status were also statistically significant predictors of enrollment as well. The study also found that 11th grade GPA and 7th grade GPA were statistically significant predictors of postsecondary persistence. Similarly, the persistence model also indicated that free and reduced priced lunch status, special education status, and mobility status were statistically significant predictors. The study discusses the significance of these findings in light of prior research, the implications for practice, future directions for research, and limitations. | |
