By Dr. Gerry Gurney, University of Oklahoma, Dr. Woodrow Eckard, University of Colorado at Denver, and Dr. Richard Southall, University of South Carolina
Each November, as part of its continuous academic rebranding campaign, the NCAA rolls out its latest Graduation Success Rate (GSR), touting historically high “success” rates for Division-I college athletes. Since its 2003 inception, GSR scores have climbed steadily from 74 percent to 86 percent. Not surprisingly, the NCAA contends its GSR offers proof of its commitment to academic success. This year, during our national obsession that is March MadnessÔ, the NCAA will undoubtedly tout Division-I African-American basketball players’ graduation success, noting “…77 percent have earned their degrees … the highest rate ever.”[1] Not surprisingly, for the past fifteen years the NCAA has sought to rebrand academic success as synonymous with GSR statistics.
The NCAA argues the GSR is more accurate than the Federal Graduation Rate (FGR), the only alternative metric. However, the GSR is itself flawed, significantly exaggerating athlete graduation rates. This exaggeration is obscured by the NCAA’s adherence to the arcane calculation underlying the GSR (see below).
Instead of simply accepting the GSR as a more accurate measure of academic success, it is important to critically examine the history of the GSR and what it actually measures. In 2003, at the urging of university presidents, the NCAA introduced the GSR — at least in part — in response to consistently dismal FGRs in NCAA Division-I football and men’s basketball. For example, in 1984 FGRs were 38% for men’s basketball and 47% for football.[2] Consistently low FGRs had eroded the public’s faith in the NCAA’s educational mission, and conflicted with its preferred image of the scholar-athlete. Rather, athletes were increasingly viewed as commercial chattel, whose value was based on their ability to produce revenue for universities and provide entertainment for fans.
In response to calls for reform, the NCAA developed a tidy narrative that it had (1) recognized the problem, (2) introduced reforms, and (3) achieved academic “success.” In a classic bait-and-switch, the NCAA’s decades-long rebranding strategy has cast doubt upon the FGR’s validity by continually and consistently touting the GSR as a more accurate and “better methodology.” While the NCAA notes “the GSR has consistently resulted in student-athlete graduation rates 10-15 points higher than [the FGR],”[3] it also claims this inflated graduation rate indicates academic success. More likely, however, the NCAA’s GSRs are “alternative facts” (i.e., propaganda) designed to obfuscate its members’ exploitation of their most valuable assets: FBS football and men’s basketball players, and deflect attention from a shameful litany of academic-fraud scandals.
At its most basic, a graduation rate is the number of students in a particular cohort who graduate, divided by the total number in that cohort. The NCAA argues the FGR penalizes schools by (1) retaining students who transfer, while (2) not counting as graduates, transfer students who graduate elsewhere. The NCAA contends the FGR treats all transfers as academic “failures,” rendering the FGR less than a “true” graduation rate.
While this is a valid criticism, the NCAA’s GSR makes a similar error in the opposite direction, resulting in an exaggerated or inflated graduation rate. The GSR calculation removes from a school’s cohort all athletes who leave while academically eligible for sports participation. It also adds to the cohort athletes who transfer in from other schools. The NCAA contends “student-athletes who depart a school while in good academic standing, Left Eligibles (LEs) … are essentially passed from that school’s cohort to another school’s cohort”.[4] This implies that almost all LEs will become transfers-in.
But this is not the case. What the NCAA does not acknowledge, is the fact that the number of transfers-in is significantly smaller than the number of LEs (see below). Contrary to the NCAA’s claims, most Left Eligibles are not passed to another school’s cohort. These “missing” LEs are not tracked by the NCAA and so their academic outcomes are unknown and may be college dropouts. Thus, the NCAA system is not held accountable for a significant number of recruited athletes. Even for those included in the GSR cohort as transfers, the original recruiting school is absolved of responsibility for failing to retain them.
Students leaving a school can be disappointed, discouraged and/or disillusioned, undermining their incentive to continue their academic careers. This is especially likely for economically disadvantaged athletes deprived of a grant-in-aid (i.e., scholarship), who can’t afford to attend college without this financial support. While all LEs must be in “good academic standing” when they leave a school, this is a minimal threshold. Many may be marginal students unlikely to succeed elsewhere. Often athletes are simply discarded, becoming recruiting mistakes for which universities are not held accountable.[5]
The number of missing LEs is large, causing the GSR to be significantly inflated. Tellingly, the NCAA does not make public GSR data or calculations for FBS football and men’s basketball, where public concern about athlete exploitation is the greatest. However, it does provide aggregated data for all Division I sports, male and female.[6] For the cohort comprised of the 2006-2009 entering classes (the latest available GSR calculation), the total number of athletes is 95,782 and the GSR is 84%. What the NCAA conveniently ignores is that its dataset includes 23,112 LEs, but only 8,165 transfers-in. In other words, there are 14,947 more LE’s than transfers-in. Astoundingly, about 65% of all LEs are unaccounted for in the NCAA’ graduation “success” data.[7]
Missing LEs who fail to graduate cause the GSR to exaggerate the “true” graduation rate. To estimate this disparity, one must make assumptions about the academic outcomes of these students, given that they are not tracked. As noted above, sub-par outcomes can be expected. The worst-case scenario is that all missing LEs become “eligible dropouts” (the NCAA’s term for these athletes in a 1991 report),[8] yielding a 73% GSR, 11 percentage points below the NCAA’s historically successful 84% GSR. More realistically, if 33 percent of LEs remain dropouts and 50 percent of the remainder graduate, this yields a 77% GSR, seven percentage points below the NCAA’s rate.
Thus, the GSR can easily be artificially inflated by seven to eleven percentage points. Again, the above estimates are for all sports, male and female. It is of course disingenuous and deceptive for the NCAA to tout the all-sport GSR when the public controversy involves the $10 billion- industry that is FBS football and men’s basketball. Not surprisingly, the NCAA does not report the calculation details for these sports, preventing any critical analysis.
The NCAA’s “graduation success” metric significantly exaggerates academic achievement and its continued use is deceptive in the context of a public debate regarding athletes’ access to a meaningful education. We call on the NCAA to make each member university’s Graduation Success Rate data publicly available to allow critical examination. There is no reason to keep these data secret.
[1] http://www.ncaa.org/about/resources/media-center/news/african-american-men-s-basketball-players-succeeding-classroom-highest-rates-ever
[2]https://www.ncaa.org/sites/default/files/GSR%2Band%2BFed%2BTrends%2B2013_Final_0.pdf
[3] http://www.ncaa.org/about/resources/research/why-gsr-better-methodology
[4] NCAA, “How are NCAA Graduation Rates Calculated?” (November 2015), pg. 6 http://www.ncaa.org/sites/default/files/How%20is%20grad%20rate%20calculated_nov_2015.pdf
[5] NCAA Research, “Tracking transfer in Division I Men’s Basketball” (November 17, 2016). http://www.ncaa.org/about/resources/research/tracking-transfer-division-i-men-s-basketball
[6] NCAA Research, “Trends in Graduation Success Rates and Federal Graduation Rates at NCAA Division I Institutions” (November 2016), page 5. http://www.ncaa.org/sites/default/files/2016RES_GSRandFedTrends-Final_sc_20161114.pdf
[7] Authors’ calculations based on data from NCAA GSR table.
[8] NCAA. (1991, June). NCAA academic performance study: Report 91-01 — A description of college graduation rates for 1984 and 1985 freshman student-athletes. Overland Park, KS: NCAA. Retrieved from http://www.ncaa.org/about/resources/research/federal-graduation-rates-report-archive
