NEW YORK, NY (February 28, 2012) — A significant number of students entering community colleges around the country are at risk of being inaccurately placed in remedial classes, two new studies from the Community College Research Center (CCRC) at Teachers College, Columbia University, have found.
The studies, one of a large, urban community college system and one of a statewide system, looked at data from tens of thousands of entering students over several years and used statistical methods to simulate how students would have fared had they been placed directly into college level courses. The simulations indicate that a quarter to a third of students assigned to remedial classes based on standardized test scores could have passed college-level classes with a grade of B or better.
Most of the nation’s community colleges rely on two standardized tests—the COMPASS and the ACCUPLACER—to determine college readiness. CCRC’s research calls into question whether these tests should be used as the sole determinant of access to college-level courses. Because CCRC’s findings were similar in two very different community college systems—an urban and a statewide system—the results strongly suggest that the problem is a general one and that large numbers of community college students across the country may be able to do well in college-level courses without taking remedial courses first.
Nationwide, 60% of entering community college students who recently graduated high school are assigned to remediation. Students must pay tuition for remedial courses, but the credits they earn do not count towards graduation. The cost to schools of providing remedial instruction has been estimated at roughly $2.5 billion dollars annually.
While CCRC found that both tests also over-place some students into college-level classes, under-placement into remedial classes is of special concern because these classes are often a dead end for students. Previous research indicates that less than 25% of students assigned to remediation go on to earn a community college credential, or transfer to a four-year college. Even students who pass the required remedial classes frequently drop out before continuing on to college-level classes—often due to weak academic skills, but possibly also due to frustration or discouragement.
While no placement process can avoid making some mistakes, both studies find that students’ high school performance is a better predictor of which students could pass college classes. In one study, using high school GPA to place students was estimated to reduce the probability of both under- and over-placement by 50%. In the other study, the benefits of using GPA instead of test scores was smaller (with a predicted 10–15% reduction in severe placement errors); but using the best of either placement test scores or high school GPA was predicted to lower the remediation rate by 8 to 11 percentage points while simultaneously reducing placement errors and increasing college level success rates.
The researchers emphasize that the validity of any test depends on how it is used in practice, but their analyses indicate using placement test scores alone does not yield the levels of accuracy that colleges and students might hope for. The findings suggest that community colleges should take into account additional measures such as high school performance when making remedial placement decisions and retain flexibility to override placement recommendations based on test scores.