Association for Public Policy Analysis & Management

The primary obstacle to college completion is often inadequate academic preparation for college-level coursework.  Most large-scale studies on the effects of remedial and developmental courses on student persistence and degree completion find negative or null effects for college students at the margin of passing out of remediation. A likely explanation for these discouraging findings is that college is too late to address issues of under-preparedness. As such, policymakers have developed a renewed interest in the academic behaviors of high school students, particularly in the courses they are required to take prior to graduation.  Prior research suggests that taking advanced math courses in high school, for example, leads to positive college outcomes (Berkner & Chavez, 1997; Goldrick-Rab, Carter, & Wagner, 2007; Adelman, 2006; Altonji, 1995; Attewell & Domina, 2008; Rose & Betts, 2001). However, the most common outcome measure in much of the prior research is enrollment in four-year colleges or the completion of a degree. We still know little about the role that high school courses play in the need for remediation and success in remedial courses in college.

This paper examines the relationship between the courses students take in high school and their success in remedial math and English courses once in college.  Using restricted high school transcript data from the Educational Longitudinal Study (ELS:02) and college transcript data from the Postsecondary Education Transcript Study (PETS), I examine the relationship between courses completed in high school and the success of students in remedial courses once in college.  To my knowledge, this is the first paper to examine the effects of high school course-taking on the need for remediation in college across a nationally representative sample of students.  Similar to Kim, Kim, DesJardins, and McCall (2015) and other quasi-experimental studies, I attempt to control for students’ self-selection into high school curricular pathways through both a matching model and an instrumental-variable approach in order to correct for this potential source of bias.  As such, I am able to examine the extent to which tracking in high school impacts remedial need in college, and the extent to which this influences subsequent college outcomes, such as degree completion.  I further examine the different patterns of remedial course enrollment that occur in the years following high school completion.

Preliminary results suggest that completing higher-level math and English courses during high school significantly decreases the probability of needing remediation in these subjects in college, but these results differ dramatically by student subgroup, particularly for English-language learners.  For subsets of students, high school course-taking is not necessarily predictive of success in remedial or college-level courses in college. These findings have important implications for how high school and college systems think about aligning their curricula for certain subsets of students in order to best support all students’ postsecondary success.