Predictors for Retention, Persistence, and Success

Many of us were unable to attend conferences this year, where we likely would have picked up on lessons to help us improve upon student and institutional success. Our aim is that this article will act as a substitute for that conference session you missed and synthesize a few lessons from our individual campuses regarding retention. Here, we describe some ways that our three institutions have addressed the commonly-discussed topic of DFW grades—that is, grades of D, F, or withdrawn—that students receive at the end of a term.

At Case Western Reserve University, much thought has gone into how we define a successful student. In addition to student GPA, we look at the number of cumulative credits and whether or not a student has received a DFW is used to help identify students of high, moderate, or low likelihood of success. At the end of their first year, students with GPAs greater than 2.5, with 28 or more credits, and no DFWs on their transcript are classified as having a high likelihood of success, and those increase to a GPA of 2.8 and 56 or more credits after year two. Students who meet two of these three criteria fall in the moderate group, and those with fewer fall into the third group.

Data from students entering in 2012 suggest those in the high likelihood of success group were over four times more likely to be retained and graduate within five years than were students in either of the other success groups. This has been extremely powerful in monitoring students and advising.

Currently, we are working with our enrollment department to incorporate applicant information such as standardized test scores and transcript data to see if we can make the groupings more sensitive predictors of retention and whether or not the pre-college characteristics are indicative of students’ success groupings after each of the first two years.

At Saint Martin’s University, the Office of Institutional Effectiveness (IE) took an in-depth look at the relationship between DFWs and one-year and two-year retention rates. Although these courses typically have relatively low DFW rates, what was found were students who received a DFW in their first term math course were nearly twice as likely to withdraw from the institution than those who received a passing grade; students who received a DFW in their first year seminar or first term English course were nearly three times as likely to withdraw.

From these findings, IE was inspired to use a student’s college-level readiness as determined by their high school academic profile and placement exam results as predictors for retention. By identifying students who may need additional support services, resources needed to successfully complete college-level work can be proactively provided, such as those who were granted provisional admittance being strategically placed with certain first-year seminar faculty and those who placed into non-college level math being placed in a math course with a peer tutor.

To track these efforts, IE developed a detailed master cohort list that included students’ GPAs, placement exam results and course enrollment. At mid-term, the list is updated to reflect any changes along with mid-term grades—with special attention to first year seminar, English, and math. Advisors of those who received a DFW at this point were notified along with the Center for Student Success and proper intervention strategies were executed. This list was revisited at the end of the fall term and repeated for the spring term.

This coming year, IE will survey incoming first-year students using a “student intake form” to give students an opportunity to self-identify where they think they may struggle. This in turn will help the university be even more proactive in providing the students the resources they need to be successful—aside from just college-level academic readiness. Although these strategies were recently implemented, there has already been a significant improvement in overall GPA for provisionally admitted students and a decrease in DFW rate for non-college level math courses. However, due to the extraordinary circumstances that many higher education institutions are facing due to COVID-19, it is difficult to fully assess the success of these efforts at this time.

At Transylvania University, we found that understanding the reasons for high DFWs is best understood when analyzing data beyond the high DFW course itself. It is an important issue for us, as during the past three years, the first-year to sophomore retention rates for students earning DFW grades is more than 20 points below those for their non-DFW peers.

Like many institutions, our courses with high DFW rates include those in math and sciences. If we focused just on those courses and offered additional support, would the effort pay off? Perhaps it would, but a deeper analysis shows that students with certain pre-entry academic characteristics are more likely to earn DFWs. We found that those students generally had lower ACT/SAT scores, were less likely to have taken at least one AP course in high school, and had lower high school GPAs. These results are not surprising, but they are instructive as we seek to build strategies to improve success rates. With such data, identifying courses with high DFW rates is only a first step. The second step involves analyzing the characteristics of students within a given class. We then have the potential to determine which students in a particular high DFW class are most likely to encounter challenges so that we can target our academic support efforts more effectively.

By pairing the knowledge of which students are at the highest risk, along with the courses that they take that may be very challenging, we are able to increase the chance of the student being successful. Different policy options have been explored, including offering supplemental instruction in courses in which higher-risk students are enrolled.