To the Editor:
The opinion piece entitled "Remediation is Not the Enemy" argues that reforming traditional remediation, specifically the corequisite model, deprives students of learning opportunities in pre-requisite developmental courses. Among the data the author cites are two studies from a randomized controlled trial conducted at the City University of New York in 2013.
Along with Mari Watanabe-Rose, we are the co-authors of those studies. The author of the opinion piece attempts to leverage our statistics to argue that a sub-group of students assigned to traditional remediation were in fact more successful than those students randomly assigned to corequisite courses. There are both factual and interpretive issues with those claims.
The crux of the problems with the author’s claims is that he is making them based on a limited subgroup from an RCT (here, students who passed traditional remediation and college level math), and this is not an appropriate way to analyze such data. These are students who, if assigned to corequisite, would likely have been even more successful. Indeed, if you pass a course that is difficult to pass, a course that is particularly alienating and discouraging for many students, and then you pass another course, that certainly says something about who you are as a student. It does not say that the remedial course made you better. This is a classic example of spurious reasoning, and of “swapping cause for effect.”
If we could randomly assign people to pass (not just take) remedial courses, that would allow us to deduce the true effect of passing a remedial course on subsequent outcomes. But such a random assignment is not possible.
What we observe causally in our RCT is that students assigned to corequisite courses were more likely to pass this first assigned math course; and that this further caused many positive short-, medium-, and long-term outcomes, including increased degree completion and wages.
What "learning opportunity" did corequisite students miss? If there was such a missed opportunity, it would have manifested at some point in the follow-up analysis. Corequisite students would have shown lower course pass rates, decreased graduation rates (associate or bachelor's), decreased persistence, and/or decreased wages after college. But in our seven-year follow-up, we show that corequisite students performed better on all of these metrics than did students assigned to traditional remediation.
A final point concerns the claim that corequisite reform is an elimination of developmental education. The co-requisite – which in our experiment was peer tutoring, but has since taken many other forms – is developmental education. What corequisite does is provide college students with just-in-time, targeted support to do well in college, rather than make them repeat work they already completed in high school as a hurdle to full participation as college students. Our study demonstrates that this is a more effective way of supporting students with long-lasting benefits.
Instructor of sociology and director of Social Science Research
--Alexandra W. Logue
Center for Advanced Study in Education
City University of New York