You are here

Why the GSR is a Better Methodology

The methodology for calculating the federal rate comes from the Student Right-to-Know Act passed in November 1990. It is limited in who it tracks, but the Department of Education has held onto this methodology to this day.

Under this federal formula, student-athlete cohorts must consist only of first-time, full-time freshmen entering in a given fall term while receiving athletically related financial aid. Student-athletes not receiving such aid at entry and those who transfer into the institution are not included in the cohort. In addition, transfers out of any school are considered academic failures according to this methodology. Thus, the federal rate does not account for students who transfer from their original institution and graduate elsewhere. Not only does the original school get penalized for that transfer student, the school from which the student eventually graduates does not get credit for a graduate.

Student-athletes are followed at the institution of initial entry for up to six years from enrollment, and anyone in this cohort who does not complete his or her degree by in six years from that college is considered an academic failure.

Throughout the 1990s, NCAA presidents and chancellors grew increasingly frustrated with the federal methodology, especially given modern patterns of student transfer (data show about one-third of all college graduates attend more than one institution). They asked the NCAA to develop a graduation calculation that more accurately portrayed a student-athlete’s academic success.

In 2002, the NCAA introduced a methodology for Division I called the Graduation Success Rate (GSR), which accounts for student-athletes who transfer into or out of a particular college.

The GSR has consistently resulted in student-athlete graduation rates 10-15 points higher than those obtained using the federal methodology. While a GSR methodology applied to student-body data would likely show a similar jump, the education department to date has not authorized the collection of data necessary to calculate a student-body GSR. Thus, the NCAA still uses the federal rate as the only way to compare student-athletes with non-student-athletes, but the GSR is used to more accurately reflect student-athlete graduation outcomes (comparison of student-athlete GSRs with student body federal rates is not appropriate).

Calculation of the GSR begins with the group the federal methodology identifies, but then it takes steps to track thousands of student-athletes the federal methodology omits. This includes transfers from two-year or four-year colleges who receive athletics aid on entry, freshmen on aid who first enroll at midyear, and recruited student-athletes at schools that do not offer athletics aid.

Read: How graduation rates are calculated (.pdf)

Rather than treating likely transfers out of a school as academic failures, the GSR essentially moves them from one school’s cohort to another’s.  In the absence of a data reporting system that tracks students from school to school, this is accomplished in aggregate using a couple simple assumptions. In the GSR calculation, students in poor academic standing (defined as not being academically eligible for athletics competition had they remained at the institution) at the time of departure are counted as academic failures, while those departing in good academic standing are removed from the cohort completely (that is, they are treated as neither a success nor a failure but rather as if they were never in that school’s cohort).

Although research has shown that not every student-athlete who leaves a college while academically eligible goes on to graduate, we also see that some student-athletes who leave in poor academic standing (and thus labelled non-graduates) go on to right their academic issues and graduate from another college.   Overall, these two assumptions are imperfect but allow for a reasonable estimate of the graduation rate one would calculate if each student-athlete was tracked directly from school to school. That came to light when the NCAA launched a series of studies of former student-athletes 10 years removed from college called “The Study of College Outcomes and Recent Experiences (SCORE).” The SCORE survey revealed a 10-year, student-centered (data collected by following students over time rather than by taking snapshots of students while they are attending particular colleges) graduation rate of 88 percent for 1994 entering freshman student-athletes in Division I. That’s about 25 percentage points higher than the six-year rates using the federal methodology, which doesn’t account for transfers, and several points higher than the six-year GSR that does track transfers.  GSR is close but likely underestimates actual graduation rates.

While at this point it would be overly burdensome for institutions to track student-athletes over time like the SCORE study did, it does add to the conversation that a student-centered focus is better than the institution-centered approach the federal rate incorporates.

Other organizations are realizing the value of the student-centered graduation-rate calculation, too. A recent report from the National Student Clearinghouse Research Center  (Shapiro, et al, 2013) in fact focuses on completions at students’ starting institution and transfer institutions, as well as persistence for those who had not earned a degree within six years.

The NSC report states:

“The patterns revealed in this study reflect both the complexity of students’ postsecondary pathways and the distinctive enrollment behaviors among students following nontraditional pathways. The results suggest that conventional approaches to understanding college effectiveness and student success, limited to students’ enrollment at the starting institution only, fail to fully capture national completion rates. It also demonstrates that, as students attend multiple institutions on the way to their first completion, each of these institutions is likely to have contributed, in its own way, to each student’s pursuit and achievement of their educational goals. The findings help point the way to policies that recognize and promote such student success while also crediting the institutions that contribute to it.” (Shapiro et al. 2013, p. 4)

The NCAA agrees with this assessment and is continuing to refine its methodology to reflect student-athlete academic achievement as accurately as possible without burdening its member institutions.


Shapiro, D., Dundar, A., Ziskin, M., Yuan, X., & Harrell, A. (2013, December). Completing College: A National View of Student Attainment Rates-Fall 2007 Cohort (Signature Report No. 6). Herndon, VA: National Student Clearinghouse Research Center. Retrieved from