A New Way to Keep Score

75% of Division I teams perform better under NCAA's reconfigured graduation statistic than under the federal rate.
December 20, 2005

For years, coaches and sports officials have griped that the federal graduation rate inaccurately measures the academic performance of college sports programs and teams, because it does not account for the many athletes who transfer in and out. After years of work, the National Collegiate Athletic Association on Monday unveiled its alternative, the Graduation Success Rate. Perhaps not surprisingly, the results make college sports look better.

"Really spectacular," gushed Myles Brand, the NCAA's president, in announcing that the newly configured rate showed 76 percent of Division I athletes graduating, compared to the comparable federal rate of 62 percent for the classes that entered the institutions from 1995 to 1998.

The NCAA has crafted its Graduation Success Rate as part of a larger effort to change the conversation about the academic performance of athletes. In the last two years, the NCAA has imposed a new set of academic standards that seek to hold teams and institutions accountable for how well, or not, athletes progress toward a degree.

To do that, it uses two measuring sticks: an "Academic Progress Rate," which measures how many of a given team's athletes return in good academic standing each semester, and the Graduation Success Rate, which is designed to show the proportion of athletes that achieve what the NCAA perceives as the ultimate goal: earning a degree. (Most of the penalties and rewards that the NCAA plans as part of its academic reform effort are tied to the Academic Progress Rate; the Graduation Success Rate will be used mostly for historical context.)

The NCAA has collected the graduation rates of its member colleges' athletes for more than two decades, and has published them for about 15 years, since members of Congress, concerned about academic and other abuses in big-time intercollegiate athletics, enacted the Student-Athlete Right to Know Act.

The federal graduation rate, which measures the proportion of first-time, full-time freshmen who graduate within six years of entering their original four-year institution (or within three years of entering a community college), is widely derided as a flawed instrument, especially as the number and proportion of students who attend multiple colleges, and the average time students take to earn degrees, have grown. But one recent effort to find a new way to measure the rates for all students has run into political problems, so changing the federal rate is probably unlikely in the near term.

But NCAA officials have pushed ahead on crafting their own graduation rate, amid increasing complaints that the federal rate understates the academic success of athletes in two ways: by counting as a nongraduate any athlete in good academic standing who leaves a college without graduating (perhaps to transfer to another college in search of more playing time, or to go pro), and by failing to count as graduates those athletes (many from community colleges) who transfer in to an institution and then earn a degree. "The federal rate is inaccurate and unfair," Brand said in a telephone news conference Monday.

"Under the federal methodology, a student-athlete who transfers from one Division I college to another is treated as a nongraduate at the first and is ignored in the calculation at the second even if he or she graduates,“ added Todd Petr, the NCAA's managing director of research. "Similarly, a two-year transfer into a Division I institution is never included in that school’s federal graduation rate calculation. Given the mobility of today’s students, the GSR is simply a more defensible methodology.”

Like the federal rate, the NCAA's new Graduation Success Rate starts with all freshmen who enter a given college in a given year. But it then excludes from the denominator those athletes who leave the institution in good academic standing, and includes in the numerator those who transfer in to the institution and proceed to graduate. Proof that the new rate offers a more complete picture, Brand and Petr said during Monday's news conference, comes in the fact that the new system tracked 35 percent more athletes (91,051 vs. 67,277) than does the federal rate for the same years.

"Clearly, the federal methodology has been missing a significant number of scholarship student-athletes who are competing on NCAA teams,” Petr said. “By the old standards, some 24,000 students who contributed on the field of play were not included in the calculation of the rate. Additionally, more than 16,000 students who left institutions in good academic standing were universally viewed as academic failures.”

By excluding the failures and including students who transferred in, the rates rise over all and, cumulatively, in every Division I sport, as the following table shows.

Division I Graduation Rates for Entering Classes of 1995-98, by Sport

Men's sports Grad. Success Rate Federal Rate Women's Sports Grad. Success Rate Federal Rate
Baseball 65% 47% Basketball 81% 65%
Basketball 58 44 Bowling 72 70
Cross country/track 73 59 Cross country/track 83 67
Fencing 85 69 Crew 89 77
Football 64 54 Fencing 93 79
Golf 77 60 Field hockey 93 81
Gymnastics 85 71 Golf 87 67
Ice hockey 80 65 Gymnastics 93 86
Lacrosse 89 76 Ice hockey 84 70
Rifle 73 62 Lacrosse 94 82
Skiing 85 69 Rifle 73 60
Soccer 77 57 Skiing 93 71
Swimming 81 67 Soccer 87 70
Tennis 83 63 Softball 84 70
Volleyball 73 59 Swimming 91 76
Water polo 87 75 Tennis 88 70
Wrestling 66 50 Volleyball 86 69
      Water polo 86 81

The rates for the vast majority of individual teams improve over the federal rate, too, NCAA officials said -- three-quarters of all teams had higher Graduation Success Rates than federal graduation rates in 1995-98, according to the statistics released Monday. (The association plans to release overall Graduation Success Rates by institution, along with the latest college by college outcomes on the federal graduation rate, next month.) For a snapshot at individual squads, here is a look at how the basketball and football teams fared at the colleges ranked in the USA Today/ESPN Top 25 in men's basketball as of Monday.

  Men's basketball   Football  
Institution Grad. Success Rate Federal Rate Grad.  Success Rate Federal Rate
Duke U. 50% 40% 87% 81%
U. of Connecticut 50 33 71 51
Villanova U. 100 69 93 84
U. of Louisville 38 22 47 42
U. of Memphis 25 13 61 48
U. of Florida 100 64 80 42
U. of Texas Austin 25 10 40 31
U. of Illinois Urbana-Champaign 100 60 73 58
U. of Oklahoma 33 25 51 38
U. of Washington 90 67 75 64
Gonzaga U. 55 38 -- --
Michigan State U. 73 64 41 39
Boston College 60 31 89 83
U. of California Los Angeles 38 31 63 57
George Washington U. 55 50 -- --
Wake Forest U. 100 60 96 87
U. of Maryland College Park 30 25 63 62
Indiana U. Bloomington 91 70 77 66
North Carolina State U. 78 54 50 39
U. of North Carolina Chapel Hill 82 75 64 57
U. of Nevada Reno 14 20 56 41
U. of Kentucky 33 21 57 40
U. of Iowa 39 33 58 53
U. of Wisconsin Madison 58 60 67 50
Ohio State U. 45 25 54 49

What would cause a team's rate to drop? It depends on the circumstances, but Walter Harrison, president of the University of Hartford and chairman of the NCAA's Division I Committee on Academic Performance, which helped draft the new standards, said that if he presided over an institution where the Graduation Success Rate was lower than the federal rate, one of the conclusions it would suggest to him is "that students are transferring in and not graduating."

"Under the federal rate, transfer students are treated as if they never existed," Harrison said. "Now they exist, and you're held accountable for how well they're moving toward graduating."

Of the teams in the above table, only the men's basketball team at the University of Nevada at Reno had a lower Graduation Success Rate than its federal rate. A request for comment from the university's athletics department went unanswered.

Other men's basketball teams that showed declines on the new rate included those at California State University at Fresno (29 percent on the GSR, 33 percent federal), the University of Idaho (35 vs. 43), and East Carolina University (57 vs. 75 percent).

Teams that did not fare well on the new Graduation Success Rate minimized their value because of the outdated nature of the data. The men's athletics director at the University of Texas at Austin, DeLoss Dodds, said that the statistics in Monday's report were "from classes that entered during the years 1995-98," and that today's Longhorns "were attending middle school at that time." He added: "These rates do not reflect the work of our current student-athletes and coaching staffs."

Critics of big-time college sports said they believed it was not a coincidence that teams with declining rates were anomalies amid the overall increases within the NCAA -- and that far fewer colleges would have to explain away poor graduation rates in the future than was true in the past.

"That was the point of the NCAA calculating them their own way," said Murray Sperber, a former English professor at Indiana University who has written widely about academic and financial problems. "For the NCAA, the whole thing has always been an exercise in PR, not in education. The point of the federal rates is to indicate to an applicant and his or her parent(s) the chances of the applicant graduating from the school that the applicant originally enters -- and given six years to graduate. It is an important piece of knowledge for an applicant and his or her parents to consider when applying to a school."

Sperber added: "The NCAA has substituted a rate that is friendly to it -- especially its athletic departments and coaches -- and not as useful for the applicant as the federal rate."


Back to Top