Student Impacts From Implementation of Early College Strategies in Comprehensive High Schools

• In Event: Transforming Comprehensive High Schools Into Early Colleges

Abstract

Objectives: This paper presents impact estimates for students’ college coursetaking and on dropout rates for all three of the projects with final outcome data; it also presents impacts on successful completion of college preparatory courses for the two projects with relevant data.

Theoretical framework: The Early College Model is expected to increase students’ successfully completion of a college preparatory course of study (the high school courses necessary for college) and their college-level coursetaking. It is also supposed to lead to a decrease in dropout rates and an increase in graduation rates. The anticipated long-term outcomes are increased enrollment in and completion of postsecondary education.

Methods and Data Sources: Each study used a quasi-experimental design in which treatment schools were matched with one or more similar comparison schools. Baseline equivalence was demonstrated between the two groups such that any differences in baseline outcome measures and key demographic characteristics was less than .25 standard deviations, thereby meeting What Works Clearinghouse design standards with reservations.

In each study, we obtained de-identified student-level administrative data for the entire state, which we used for both the matching and the outcomes. Results were run as multi-level models with students clustered within schools.

Results: In two of the states, transcript data were available to examine the completion of a college preparatory course of study; this was used as a measure of students’ readiness for postsecondary education. In both studies, there was no impact on 9th grade students’ successful completion of the high school courses necessary for college success.

In two of the projects, there were statistically significant impacts on students’ college-level coursetaking and earning of college credits. In the third project, there was a positive impact, but it was not statistically significant because of the high number of students already taking college-level courses (the comparison group was at 86%) and the small sample size.

In one of the states, the impact was much smaller on minority students than non-minority students, a finding that was consistent with other recent research showing that dual enrollment is not equitably distributed (Miller et al., 2017). In the other state, however, the impact was similar or higher for minority students, a finding running counter to the recent research.

There is evidence that some expansion of dual enrollment occurred because students were being shifted out of Advanced Placement courses. We also have evidence that students were much more likely to earn potential college credit through dual enrollment courses than through AP courses (more than 93% of dual enrollment courses ended up in a passing grade while less than 10% of AP courses results in a passing grade on the exam).

In one of the states, the treatment resulted in a decreased dropout rate, with a particularly large benefit for English Language Learners. In the other two states, dropout rates were non-significantly lower in treatment schools than in comparison schools.

Significance: This paper shares impacts and insights from an extensive federal investment to increase college access and success.

Author

• Eric Grebing, Montgomery Hospice