2017 ASEE Annual Conference & Exposition

One of the curiosities of remedial math education in community colleges (CC) is that faculty typically do not have access to students’ academic background or placement testing data. This is understandable given the general lack of integrated K-12 and postsecondary data and articulation between these systems. CCs therefore resort to reliance on complex and expensive assessment and placement (A&P) systems to sort students into coursework. However, while over 90% of CCs use placement tests, the data from these tests also largely do not make it into the hands of faculty. They therefore typically begin teaching their courses without many clues as to students’ abilities, strengths, and weaknesses.

Although this may be a hallmark of the open-door CC model, research suggests that more data may in fact be helpful. Higher education scholars have documented how variables such as high school GPA and prior course-taking are often stronger predictors of college success than test scores (Adelman, 2006). Challenging the over-reliance on placement testing, researchers have also found that supplementing or even replacing placement tests with high school transcript (HST) information may improve course placement (Scott-Clayton et al., 2014). Further, math diagnostics, which provide skill specific information about student math skills, can also be a potentially valuable resource for math faculty. They can improve placement accuracy and help teachers tailor instruction in math classrooms (Betts, Hahn, & Zau, 2011).

There is limited research that explicitly addresses the role and usefulness of HST and diagnostic data in the CC setting. Addressing this gap in the literature, we conducted a mixed methods study to understand whether and how a wealth of background data from HSTs and math diagnostics might improve A&P and be useful to CC math faculty. We examined longitudinal student records from high school to CC in a large California metropolitan area and tested associations between relevant academic background variables from HSTs, placement testing results, and students’ CC outcomes. We also conducted surveys and interviews of all full-time math faculty (N=22) at one CC where diagnostic data are collected during placement testing but not shared with faculty. This provided an opportunity to gather math faculty members’ insight on the usefulness of HST and diagnostic data for improving math placement and for tailoring instruction in math classrooms.

The initial survey results are illuminating. Most faculty members reported that HST information, particularly grades in math courses, were the best predictors of math success, even more so than placement tests. However, nearly half reported knowing very little about incoming student math skills. Interestingly, several indicated they did not want to receive additional information about student skills and were skeptical about the quality of additional information they would receive.

We will supplement these findings with the results of the quantitative analyses at the conference. The results of the study provide much needed information that can improve A&P practices in developmental math, as well as provide insight into how faculty can use valuable diagnostic and HS transcript information to inform their teaching.