Peter M. Crosta

Improving the Targeting of Treatment: Evidence From College Remediation

Remediation is one of the largest single interventions intended to improve outcomes for underprepared college students, yet little is known about the remedial screening process. Using administrative data and a rich predictive model, we find that severe mis-assignments are common using current test-score-cutoff-based policies, with “underplacement” in remediation much more common than “overplacement” college courses. Incorporating high school transcripts into the process could significantly reduce placement errors, but adding test scores to already available high school data often provides little marginal benefit. Moreover, the choice of screening policy has significant implications for the racial and gender composition of college-level courses. Finally, the use of more accurate screening tools would enable institutions to remediate substantially fewer students without compromising college success.

Predicting Success in College: The Importance of Placement Tests and High School Transcripts

This paper uses student-level data from a statewide community college system to examine the validity of placement tests and high school information in predicting course grades and college performance. We consider the ACCUPLACER and COMPASS placement tests, using two quantitative and two literacy tests from each battery. We find that placement tests do not yield strong predictions of how students will perform in college. Placement test scores are positively—but weakly—associated with college grade point average (GPA). When we control for high school GPA, the correlation disappears. Placement test scores are positively associated with college credit accumulation even after controlling for high school GPA. After three to five semesters, a student with a placement test score in the highest quartile has on average nine credits more than a student with a placement test score in the lowest quartile. In contrast, high school GPAs are useful for predicting many aspects of students’ college performance. High school GPA has a strong association with college GPA; students’ college GPAs are approximately 0.6 units below their high school GPAs. High school GPA also has a strong association with college credit accumulation. A student whose high school GPA is one grade higher will have accumulate approximately four extra credits per semester. Other information from high school transcripts is modestly useful; this includes number of math and English courses taken in high school, honors courses, number of F grades, and number of credits. This high school information is not independently useful beyond high school GPA, and collectively it explains less variation in college performance. We also calculate accuracy rates and four validity metrics for placement tests. We find high “severe” error rates using the placement test cutoffs. The severe error rate for English is 27 to 33 percent; i.e., three out of every ten students is severely misassigned. For math, the severe error rates are lower but still nontrivial. Using high school GPA instead of placement tests reduces the severe error rates by half across both English and math.

Does Age of Entrance Affect Community College Completion Probabilities? Evidence From a Discrete-Time Hazard Model

Research has consistently shown that older students—those who enter college for the first time at age 25 or older—are less likely to complete a degree or certificate. The authors estimate a single-risk discrete-time hazard model using transcript data on a cohort of first-time community college students in Florida to compare the educational outcomes of older and traditional-age students. Contrary to conventional wisdom, the authors’ results suggest that after controlling for cognitive mathematics ability, older students enrolled in Florida community colleges have a higher conditional probability of completing a degree or certificate in the observed event period.

Intensity and Attachment: How the Chaotic Enrollment Patterns of Community College Students Relate to Educational Outcomes

This study examines the relationship between community college enrollment patterns and student outcomes—credential completion and transfer to a 4-year institution—introducing a new way of visualizing the various attendance patterns of community college students. Patterns of enrollment intensity (full- or part-time status) and continuity (enrolling in consecutive terms or skipping one or more terms) are graphed and then clustered according to their salient features. Using data on cohorts of first-time community college students at five colleges in a single state, the study finds astounding variation in student enrollment patterns. Clustering these patterns reveals two relationships: the first is a positive association between enrollment continuity and earning a community college credential, and the second is a positive association between enrollment intensity and likelihood of transfer.

Can Community Colleges Afford to Improve Completion? Measuring the Cost and Efficiency Consequences of Reform:

Community colleges are under pressure to improve completion rates and efficiency despite limited economic evidence on how to do so and the consequences of different reform strategies. Here, we set out an economic model of student course pathways linked to college expenditures and revenues. Using detailed data from a single college, we calculate baseline efficiency and differences in efficiency for students who follow different pathways. We simulate changes in output, expenditures, revenues, net revenues, and efficiency assuming that the college meets performance targets. We find substantial differences in efficiency across pathways and significant differences in efficiency across strategies to help students complete college. The model has wide practical application for community colleges.

What Can Student Right-to-Know Graduation Rates Tell Us About Community College Performance?

This paper examines the validity of the Student Right-to-Know (SRK) graduation rates as measures of community college performance. The SRK rates are the only performance measures available for every undergraduate institution in the U.S. Many community college educators argue that the SRK rates give an inaccurate picture of community college outcomes. Using data from national longitudinal surveys of college students, we examined criticisms commonly leveled against the SRK measures and found that the SRK rates do indeed yield a biased and potentially misleading picture of individual community college student outcomes. We then analyzed the usefulness of the SRK rate as a measure of relative institutional performance. Specifically, we considered whether using different measures of performance would result in substantially different rankings of Florida’s 28 community colleges. Contrary to initial expectations, we found that the relative performance of the colleges did not change substantially as different students or outcomes were used. Even after we adjusted for student characteristics that might affect outcomes, the college rankings were still fairly stable.

Can Community Colleges Afford to Improve Completion? Measuring the Costs and Efficiency Effects of College Reforms

Community colleges are under pressure to increase completion rates and efficiency despite limited evidence of the economic consequences of different reform strategies. We introduce an economic model of student course pathways linked to college expenditures and revenues. Using detailed data from a single college, we calculate baseline efficiency and differences in efficiency for students who follow different pathways. We simulate changes in output, expenditures, revenues, net revenues, and efficiency assuming that the college meets particular performance targets. Findings indicate substantial differences in efficiency across pathways and significant differences in efficiency across strategies to help students complete college. They also suggest that increasing the completion rate is difficult and typically requires additional resources beyond the costs of implementing particular strategies. The model has wide practical application for community colleges.

The Value of Student Right-to-Know Data in Assessing Community College Performance

Traditionally, community colleges were judged on their number of enrollments and their ability to provide postsecondary education to a wide variety of students. Recently, however, state and federal policymakers have become increasingly concerned with student outcomes, and some states have even begun to consider linking the funding of community colleges to their performance on student outcome measures. In 1990, Congress passed the Student Right-toKnow (SRK) and Campus Security Act. It requires that all colleges report graduation rates to the National Center for Education Statistics (NCES) in order for their students to receive federal financial aid. These Student Right-to-Know graduation rates are part of the Integrated Postsecondary Education Data System (IPEDS). The SRK rates are the only performance measures available for virtually every undergraduate institution in the nation, including community colleges, but critics assert that the rates understate the success of community colleges in several important ways. This Brief summarizes a study by the Community College Research Center (CCRC) that investigated the nature and validity of the SRK rates for community colleges by analyzing data on students attending Florida’s 28 community colleges. It sought to determine whether the rates provide useful information that can guide educators and policymakers working to improve the performance of community colleges, or whether the rates contain biases serious enough to negate their usefulness. This Brief also suggests how the process of determining the rates might be improved.

Balancing Work, Family and School: Enrollment Pathways and Outcomes of Older Community College Students Compared to Traditional Age Students

This paper presents findings from a new study of the experiences and outcomes of older community college students - those who enter college for the first time at age 25 or later. We estimate a discrete-time hazard model using transcript data on a cohort of first-time community college students in Florida to compare the effect of enrollment pathways on educational outcomes of older students with those of traditional age students. Results suggest that reaching milestones such as fall-to-fall retention, obtaining 20 credits or completing 50% of the program is a more important positive factor affecting graduation probabilities for younger students than it is for older students. We also find that although remediation decreases the odds of graduating in any given term, older students who enroll in remediation are less negatively impacted than younger ones who do.