Measure Twice – Cut Once: Getting it Right with Assessment and Placement
When the New York Times runs a story on assessment and placement strategies at community colleges, developmental education educators and postsecondary administrators better sit up and take notice. The story, which reports the findings of two recent studies by the Community College Research Center, highlight how poor assessment and placement practices result in a high percentage of students who are misplaced into developmental education courses they don’t need or college courses they are not prepared for and the negative impact these practices have on student success. The bottom line of both studies are clear, institutions must develop multiple measures for placing students into or out of developmental education courses.
While many might point their fingers at Accuplacer and Compass as the villains in this sordid tail – Judith Scott Clayton makes it pretty clear in her study that the real problem is that institutions misuse the assessments and don’t exact the necessary due diligence to make sure the assessments are valid measures of college readiness at their institutions. Yes, validity testing or aligning an assessment with expected learning competencies for students is not easy work. However many states like Virginia and Florida have done the heavy lifting and now are using assessments that are more aligned with the content they expect students to know. In fact, Accuplacer, in their response to the two studies articulate that the redesigned Accuplacer is intended to be customized so that it can be more effectively aligned with the content that institutions or systems expect. Likewise, ACT maintains that their diagnostic assessment can be aligned with specific college content. The problem, according to College Board and ACT is that not many take advantage of these tools.
All of this is not to say that the Accuplacer and COMPASS are ideal instruments for measuring college readiness. They are inexpensive and short exams that meet the convenience test for many institutions. It is for this reason that both studies, particularly the second study by Clive Belfield and Peter Crosta, look at additional measures to determine whether they predict student success. In particular, the studies look at high school transcripts and high school GPAs. The studies find that while a review of high school transcripts does not help predict student success, high school gpa does and is often a better predictor of college readiness than placement exams. Despite concerns from postsecondary institutions about whether GPA is reliable given the inconsistency of high school requirements and curriculum, the studies found that GPA is a very effective proxy for measuring other factors like student effort and commitment. When combined with a placement exam, high school GPA improve the likelihood that institutions will make a correct placement.
Belfield and Crosta take their work one step further by suggesting ways that postsecondary institutions could easily incorporate GPA into their placement decisions. Their research found a rule of thumb that institutions should test. Essentially they contend that when using high school GPA, institutions could be reasonably confident that students would perform approximately one grade level below their high school GPA or better in college level gateway classes. In other words, a “B” student in high school could be expected to earn a C+ or better in gateway courses. They suggest that campuses could guarantee admission into gateway courses for any student who has a GPA that is at least one grade above what would be acceptable in a gateway course. For those who fall below this standard, you could offer a placement exam to determine whether a student should be placed in to some form of academic support.
Yes, there are many students who are years removed from high school so high school GPA may not be an ideal measure, but the findings are clear – institutions must consider multiple measures for all students. For adults, it may be nothing more than a simple survey combined with an assessment exam that provides some insight into the motivation and commitment of students to succeed in college – the very measures that the research suggested high school GPA was a reasonable proxy.
Like so much of what has been learned from research into developmental education, this research makes it pretty clear that institutions must review and change their practices. The good news is that simple changes might result in significant improvements in student success – an important accomplishment in our new era of postsecondary accountability.