Academic Standards and the NCAA
Robert K. Fullinwider

In March, a federal court in Philadelphia enjoined the National Collegiate Athletic Association (NCAA) from enforcing Proposition 16, its rule for determining whether a high school student athlete can be recruited by Division I colleges and universities. Under Proposition 16, a high school graduate with a 2.0 GPA in core academic subjects becomes a "full qualifier" by scoring 1010 on the SAT. If his GPA is as high as 2.5, he can qualify with a score of 820 on the SAT. (A student can offer comparable ACT scores as well.) As a full qualifier, he is eligible to receive an athletic scholarship and immediately play a varsity sport. A high school graduate can score as low as 720 on the SAT and still become a "partial qualifier" if the sum of his test score and GPA is at least as high as that of a full qualifier. In other words, he can partially compensate for a below-minimum test score by a very good GPA. As a partial qualifier, he is eligible to receive an athletic scholarship but not to play during his freshman year. Any high school student who fails to satisfy one of these two conditions can enter a Division I institution only as a paying, regular student.

When Proposition 16's predecessor, Proposition 48, took effect in 1986, it drew protests from black coaches, who deemed it unfair to minority athletes. When Proposition 48 was replaced by the slightly more rigorous Proposition 16 ten years later, black coaches were so incensed that they toyed with the idea of boycotting NCAA events. The gravamen of their complaint: the use of the SAT cut-off score, which blacks fail to reach in markedly higher proportions than whites. John Thompson, then-coach of Georgetown University’s basketball team, complained that poor minority kids were at a disadvantage taking the "mainstream-oriented" SAT. "Certain kids," he noted just after the federal court’s decision, "require individual assessment. Some urban schools cater to poor kids, low-income kids, black and white. To put everybody on the same playing field [i.e., to treat them the same in testing] is just crazy." John Chaney, Temple’s basketball coach and another early critic of Proposition 16, offered a similar postmortem in an editorial for the New York Times. Proposition 16, he insisted, had excluded students who could make it in college. He referred to one of his own players, Rasheed Brokenborough, who was raised in a poor part of Philadelphia and scored too low on the SAT to be even a partial qualifier.

That meant we could not provide him with any scholarship money, much as he needed it, during his freshman year, and he had to accumulate 24 credits or face expulsion. Rasheed was a serious student and finished his course work in four years plus summer school. . . . He is now doing student teaching in a Philadelphia school. He is also deeply in debt. He was punished for a crime he did not commit.

Proposition 16 eliminated too many Rasheed Brokenboroughs, in Chaney’s view, and most of them were black. In 1997, 21.4 percent of black prospective athletes failed to achieve eligibility, in contrast to 4.2 percent of whites. Thus, in the words of the Black Coaches Association, "minority and low-income student athletes, academically qualified as measured by their classroom performance," have borne the brunt of a misguided effort to set academic standards. Let a student’s grade-point average suffice for admission and participation, the black coaches insist.

Why not let grades suffice? Few colleges or universities do. They seldom forgo standardized tests altogether and rely entirely on grades for admissions decisions, and for good reason. High school grading practices and academic standards vary considerably. The same numerical GPA possessed by different students may signal quite different academic capabilities (or liabilities). Thus, an uncontextualized GPA may or may not show students to be "academically qualified as measured by their classroom performance." Colleges use the SAT (and similar exams) to provide an objective counterweight to grades. Although on average the high school GPA is a fair predictor of a student’s academic success in college, a student’s GPA combined with her SAT score provides an even better predictor.

John Thompson proposes individualized decision making, since an SAT score may not reflect a student’s full potential. Look at each applicant case by case, he suggests. But where thousands of applicants are involved, individualized assessment may be impracticable. Moreover, why, on average, would we expect individualized assessment to produce different admissions outcomes? Unless institutions of higher education give up predicting how well students will fare once admitted, GPAs and SAT scores will continue to loom large in admissions decisions, for whatever their weaknesses as predictors, they prove far superior to student essays, teacher recommendations, extracurricular activities, socioeconomic background, and the like.

Thompson and Chaney focus on the "false negatives" generated by the SAT minimum cut-off. Because the cut-off score is not a perfect predictor of academic performance (indeed, it is very crude), it will sometimes exclude as incapable students who, as in the story of Rasheed Brokenborough, actually would succeed in college. By themselves, however, such stories, though credible and touching, are not dispositive. False negatives will be a problem for any standard of selection. There will always be the special individual who could succeed despite every sign to the contrary. Some high school graduates who score below 720 on the SAT (a truly low score, and the limit for even partial qualification) may nevertheless pass their college courses. And it is no less possible that some students with a high school GPA of 1.6 may survive the rigors of college. A GPA minimum cut-off of 2.0 creates false negatives just as does an SAT cut-off score of 720. To evaluate any selection system, we have to ask two questions: (i) is the minimum cut-off reasonable in light of the system’s goals? and (ii) does the minimum cut-off produce false negatives out of proportion to its prevention of false positives? That is to say, does the cut-off, in successfully screening out students who wouldn’t succeed in school, erroneously screen out too many who would? The story of Rasheed Brokenborough doesn’t answer these questions. Unfortunately, the legal case, Cureton v. NCAA, doesn’t answer them, either.

The core narrative in Cureton is simple enough. Tai Kwan Cureton and two other students denied full eligibility by Proposition 16 sued the NCAA under Title VI of the Civil Rights Act of 1964, a title that forbids institutions who receive federal funds from discriminating on the basis of race. The judge determined that the NCAA fell under the scope of the title.

Current Title VI case law establishes a two-part test of discrimination. Any practice by a covered institution is discriminatory if (a) it creates an adverse disproportionate impact on blacks and (b) it is not justified by "educational necessity." The second conjunct of the test itself divides into two. "Educational necessity" is established when (c) the challenged practice is shown to bear a "manifest relationship" to a legitimate educational goal and (d) no other practice producing less disproportionate impact serves the goal equally well. In Title VI litigation, the burden of proof moves backwards and forwards. First, the burden lies with the plaintiff to show that the defendant’s practice produces disproportionate impact. When that burden is met, the defendant must then show that the challenged practice bears a "manifest relationship" to a legitimate goal. If such a showing is made, the burden returns to the plaintiff to offer evidence of alternatives that can serve the goal as well as the challenged practice without generating the same adverse impact.

The judge’s decision in Cureton tracks this formula. The plaintiffs established "disparate impact," in his opinion, by showing that of those African American students who requested eligibility-certification in 1996 and 1997, 26.6 percent and 21.4 percent, respectively, were denied it. The comparable denial rate of white requests was 6.4 and 4.2 percent. Turning, then, to the defense of this disproportionate effect, the judge observed that the NCAA justified Proposition 16 as a device "designed to discourage the recruitment of athletically talented, but academically unprepared students," using graduation rates as a proxy measure of its success in achieving this goal. The NCAA further argued that the graduation rate for student athletes has, indeed, increased since Proposition 16 was put in place. Finally, the NCAA contended that using the 820 SAT score as a minimum for full eligibility bears a "manifest relationship" to raising graduation rates for student athletes. Its own research, it insisted, shows that SAT scores plus GPAs predict academic performance and graduation rates.

The judge conceded to the NCAA that the goal of raising graduation rates of student athletes is legitimate but challenged the Association’s contention that the 820 cut-off bears a "manifest relationship" to this goal. There were four parts to his challenge. First, he argued that the NCAA’s evidence indicated a temporal relationship between the implementation of Proposition 16 and improved graduation rates for athletes but fell short of showing causation. The evidence did not distinguish the many variables that influence graduation so that the contributory effects of Proposition 16 could be isolated.

Second, the judge observed that the use of the SAT to predict graduation rates was questionable. The test has not been formally validated for that purpose. It has been validated only as a predictor of first-year academic performance. That the NCAA’s own research (and the research of others) indicates a stronger or weaker correlation between SAT scores and graduation rates doesn’t change the fact that the test is being used for a purpose it wasn’t designed for. In the case law arising from Title VI (and from Title VII, the employment antidiscrimination statute which supplies much of the doctrine for Title VI cases), using a test beyond its intended purposes is legally treacherous.

Third, the judge could find in the NCAA’s evidence no argument for the 820 score as a "logical break point." Why was this the disqualifying score rather than some adjacent one? The NCAA’s reference to informal impressions that scores below 820 indicate reading problems in students wasn’t sufficient.

Finally, the judge noted that the NCAA’s own evidence showed that partial qualifiers graduate at the same rate as full qualifiers, calling into question the contention that a score of 820 represents a true minimum, below which students are at risk of academic failure.

For these reasons, the judge refused to grant the NCAA’s claim that the 820 score was "manifestly related" to the goal of raising student athlete graduation rates. Moreover, he argued that the plaintiffs would prevail in any case, since they could point to alternative eligibility criteria that would generate less adverse impact while not appreciably degrading achievement of the NCAA’s goal -- alternatives canvassed in the NCAA’s own internal research. According to projections developed by the NCAA, continued use of Proposition 16 would generate a future graduation rate of 61.8 percent for student athletes at the cost of a black ineligibility rate of 19.4 percent. Alternatively, allowing partial qualifiers to become full qualifiers by dropping the full-qualification minimum to 720 would yield a projected graduation rate of 60.7 percent coupled with a black rejection rate of 15.9 percent. A third variation, using a sliding scale of SAT scores between 600 and 1010 pegged to GPAs between 2.75 and 2.0, would produce a graduation rate of 60 percent with 15.7 percent black ineligibility. Finally, eliminating the SAT cut-off altogether would result in a graduation rate of 59.8 percent and a denial of eligibility to 15.6 percent of blacks. The NCAA "has not demonstrated there is anything special about" the 61.8 percent graduation rate, wrote the judge, and so it cannot insist that these alternative practices, which produce less adverse impact on minorities, do not substantially realize its goal.

Thus, the judge found for the plaintiffs and enjoined the NCAA from using Proposition 16, an injunction stayed while the NCAA appeals.

Legal cases tend to be formulaic; argument must be bent to fit predetermined categories. The judge’s opinion in Cureton is not only typically formulaic, it is also opaque in important ways. Since the opinion was rendered without trial, relying on submissions by the NCAA and the plaintiffs, it is difficult to decide how much the case turned on the NCAA’s defense strategy and how much on disputable legal theories about the appropriate role of tests.

One thing is clear. The judge’s opinion relies entirely on the supposed nexus, or lack of it, between the 820 SAT cut-off score and the graduation rates of student athletes. Yet the real issue in the case has nothing to do with graduation rates. The real issue has to do with gross academic deficiencies.

Proposition 16 and its predecessor arose in a particular context, a context symbolized by a picture that played in every newspaper and on every TV screen in 1982 -- a picture of 6-foot, 9-inch Kevin Ross, 23 years old, enrolled in a private elementary school. After four years of high school and four years at Creighton University on a basketball scholarship, Ross possessed the language skills of a fourth grader. His basketball eligibility ended, Creighton placed him at Marva Collins’s Westside Preparatory School in Chicago to help him learn to read. The Ross incident was not singular or isolated. Other instances of illiterate college athletes came to light in the 1980s. College and university presidents were embarrassed, and rightly so. They set out to prevent such scandals by toughening academic standards for recruitment of athletes. They created Proposition 48 and then Proposition 16.

Now, preventing the corruption of the academy caused by admitting students who cannot read has nothing directly to do with graduation rates, and the NCAA has no real interest in graduation rates as such. Even before Proposition 48, athletes in general graduated at the same rate as regular students. Athletes now graduate at a slightly higher rate, possibly as an upshot of Proposition 16, but as the judge in Cureton noted, there is nothing special about a graduation rate of 61.8 percent. Indeed there isn’t. Suppose college athletes were graduating at a 55 percent rate rather than a 60 percent rate, but at the same time were all majoring in physics, engineering, philosophy, or French literature and were all exhibiting GPAs 30 percent higher than regular students’. College presidents -- and the NCAA -- would have cause to rejoice, not to lament.

By linking its justification of Proposition 16 directly to graduation rates, the NCAA severely weakened its ability to defend the 820 SAT cut-off. First, as the judge was quick to point out, the SAT was not designed to predict graduation rates (even though it might actually do so), and the NCAA had been warned on this point by the president of the Educational Testing Service, creator of the test. Second, many factors enter into graduation besides academic ability, making it difficult to show a tight linkage between a particular cut-off score and a particular graduation rate. Third, since student athletes were already graduating at the same rate as regular students, how could the NCAA make a case for the urgency of increasing the rate? Although the judge in Cureton granted the educational legitimacy of the NCAA’s putative goal, the lack of a compelling reason why the graduation rate should be at one level rather than another weighed heavily in his opinion. It magnified the gravity of the disparate impact caused by the 820 cut-off (a large cost to achieve a seemingly trivial gain), and it ratified the plaintiffs’ alternatives (a 60 percent rate is not seriously worse than 61.8 percent).

Thus, because of its focus on graduation rates, a focus prompted by the NCAA’s own defense strategy, the judge’s opinion in Cureton occluded the real argument -- reasonable or not -- for Proposition 16. The NCAA has not been insensitive to the effect of Proposition 16 on minorities and to the attendant problem of excluding potential successes; indeed, its own internal research was the basis for the plaintiffs’ case. But it sought to balance those evils against another evil, the admission of students entirely unprepared for college-level work -- an evil that had been condoned under the system in operation before 1986.

It is only when the NCAA’s educational goal is understood directly in terms of avoiding gross academic deficiencies rather than increasing average graduation rates that the need to strike a balance between two different evils stands in relief. When, instead, the NCAA goal is construed in terms of graduation rates, the character of the trade-off takes on a different appearance. It presents itself as a matter of balancing a substantial evil (adverse impact on minorities) against slight increases in a good (a one or two percentage point gain in overall graduation rates).

Look how the judge compared Proposition 16 and the three alternatives. For each, he listed as a basis for comparison its projected graduation rate, projected black ineligibility rate, and projected black false negatives. Where is the fourth projection, of false positives -- those students predicted to succeed but who won’t? Consider the alternative that drops standardized test scores altogether and relies only on high school GPAs, an alternative the judge thought plenty good enough. This alternative was the scheme in place before Proposition 48, the scheme that permitted scores if not hundreds of Kevin Rosses on college campuses. Yet the judge projected no false positives from this scheme that might weigh in the balance against the exclusion of potential successes under Proposition 16.

This blindness to the real issue is an artifact of focusing on graduation rates. The desire to avoid the real damage to academic integrity that comes from admitting students who can’t read gets transformed into an unexplained preference for a barely visible increment in graduation rates. The real cost of the fourth alternative for determining eligibility, the alternative that dispenses with standardized test scores altogether, is Kevin Ross, who graduated from high school with a "little bit over a C" average.

Suppose, then, that a minimum high school GPA of 2.0 doesn’t screen out actually illiterate students. Suppose, further, that the functional illiteracy rate among high school football and basketball players is 25 to 30 percent, as one report suggests, a rate double that of regular students. Suppose, finally, that the intense competition among Division I basketball and football programs drives all institutions to the absolute minimum academic standard of selection. How, then, is the NCAA to prevent illiterate players from populating campuses of higher education? As imperfect as standardized tests are, they can detect gross deficiencies. Kevin Ross’s very low score on the ACT in 1978, had Proposition 16 been in place, would have barred his recruitment by Creighton University.

Why must the scandal of illiterate athletes be addressed by an NCAA rule? Why can’t individual campuses protect their own integrity? John Thompson sees nothing especially troubling about the loss of Proposition 16. "Institutions can still establish academic standards higher than the NCAA’s" if they want to, he argues, "but they’re all concerned that some schools will try to seek a competitive advantage. Well, people who believe that are questioning the academic integrity of all the presidents, the deans and the professors. They question the integrity of the student, but they don’t trust their own academic integrity." Thompson’s reaction betrays a naive conception of the situation confronting the academy. When it comes to big-time athletics, colleges and universities don’t trust their own integrity for good reason. Were they not faced with a classic collective action problem in competing for, and with, athletes, there would be no NCAA standards. There would be no recruiting rules, no limit on athletic scholarships, no bar to paying athletes whatever the market will bear, no restriction on training periods, no ban on drugs, no restraint on athletes dealing with sports agents or gamblers. Each college or university would set its own high standards and do the honorable thing. But, of course, the logic of competition undermines this happy scenario. Individual institutions cannot set standards higher than the prevailing norms without suffering competitive disadvantage. The only way a norm -- if it is thought to be too low -- can be raised is by collective action, by all the institutions together binding themselves to a new rule. There won’t be a new norm without an NCAA rule to bring it about.

According to the judge in Cureton, the NCAA is free to set any standard it wants -- including an 820 SAT minimum score -- so long as it actually shows the standard to be "manifestly related" to, and necessary for, a legitimate educational goal. Thus, the NCAA might revisit Proposition 16 and supply stronger evidence that the 820 cut-off enhances athletes’ graduation rates. This appears to be its present course of action. In its brief filed this summer with the federal court of appeals, the NCAA claims its research now shows a "direct link" between the SAT scores and graduation rates of athletes. At best, however, this evidence will allow the NCAA to prevail on the claim that the 820 SAT cut-off is "manifestly related" to its educational goal, but it won’t establish necessity. As long as the NCAA poses its goal in terms of graduation rates, there are other, weaker eligibility rules, producing less adverse impact on blacks, that yield outcomes "substantially similar" to those yielded by Proposition 16.

Better that the NCAA drop the focus on graduation rates. Graduation rates entered the Cureton argument in the first place only as a proxy for the real goal -- preventing the recruitment of "academically unprepared students" -- though the NCAA and the judge both then immediately substituted the proxy for the goal and proceeded as though the matter at stake were the connection between the 820 cut-off and graduation rates. To defend Proposition 16, the NCAA should specify its goal directly: blocking the recruitment of illiterate students. Then it can argue that the plaintiff's alternatives don't substantially realize the NCAA's goal if they allow -- as they might -- Kevin Rosses to be recruited. Alternatively, in keeping with its goal, the NCAA could drop the SAT altogether and administer its own reading comprehension test to prospective athletic recruits.

Coach Denny Crum of the University of Louisville basketball team suggests that the NCAA can accomplish its goal through already existing "continuing eligibility" standards. But it is hard to see how. These standards require athletes, once admitted to college, to "maintain satisfactory progress toward" a degree. However, they leave the interpretation of "satisfactory progress" up to member institutions and the athletic conferences to which they belong. Presumably, Creighton University had its own "satisfactory progress" standards in place in the early 1980s, but they did not keep an athlete who couldn't read from spending four years in residence. Kevin Ross accumulated 96 hours of college credit in such courses as first aid, theory of track and field, marksmanship, and introduction to ceramics. To minimize such time-serving, the NCAA now further requires athletes to declare a major by their fifth semester, to have completed 50 percent of their courses in that major by the beginning of their fourth year, and to possess a GPA at the beginning of their fourth year that is 95 percent of the minimum required by their institution to graduate. Moreover, it requires students to satisfy 75 percent of their course requirements in regular semesters, not summer school or correspondence courses. These requirements block some of the more flagrant deviations from normal academic standards witnessed in the 1980s, but they do not do the work of Proposition 16.

Another option gaining favor among coaches, college presidents, and NCAA officials is the elimination of freshman eligibility. This reform is long overdue in any case, and as a substitute for Proposition 16 it might accomplish some of the latter's aims. Freshmen were not eligible to participate on varsity teams for most of this century. When this norm was abandoned in the 1970s, the change had nothing to do with academic concerns. It derived largely from pressures arising out of developments in college football. With the advent of the two-platoon system in the 1960s, teams needed to carry greatly enlarged squads. Coaches -- and college presidents -- looked at the quarter of their scholarship athletes who couldn't suit up and saw a way to leverage huge squads without greatly increasing the number -- and cost -- of scholarships: make those ineligible athletes -- the freshmen -- eligible to play. The basic consideration today against barring freshman play remains cost: to maintain the current level of competition, teams will need more athletic scholarships. No one advances an academic reason for maintaining freshman eligibility; no one advanced an academic reason in 1970 for adopting it.

Ending freshman eligibility would obviate the objection to Proposition 16 mounted by John Chaney and John Thompson. They complain that Proposition 16 denies to some disadvantaged kids the chance to prove they can succeed at college work. Let colleges give these kids the chance, then. Let the low-scoring kids show they can succeed at college work by actually succeeding -- but at real college work, not at the "bonehead" curriculum served up to Kevin Ross. Then let them play ball.

Cozened by the plaintiffs' fallacious argument, the judge in Cureton rejected the legitimacy of Proposition 16's 820 cut-off score partly on the grounds that partial qualifiers, who score only 720 on the SAT, graduate at the same rate as full qualifiers. This was proof, he thought, that the 820 score bore no "manifest relationship" to the NCAA's goal of increasing graduation rates. Of course, it is no such proof at all. The academic parity of full and partial qualifiers can just as easily be seen as proof of the wisdom of the NCAA policy of keeping partial qualifiers out of competition their freshman year. It can just as easily stand as evidence that forcing academically weak students to concentrate on the classroom their first year enables them, in the end, to do as well as those with greater academic strengths who are thrown immediately into the cauldron of big-time competition.

John Chaney thinks Proposition 16 did a disservice to Rasheed Brokenborough. If Proposition 16 hadn't existed, Brokenborough would have suited up immediately for the Temple Owls, without sitting out his freshman year. But how does Chaney know that it wasn't precisely that freshman year free of athletic distraction that allowed Brokenborough to lay the foundation for his ultimate academic success? Proposition 16, in fact, may have saved Rasheed Brokenborough's academic life, even if it required him to pay for his freshman year out of his own pocket. Ironically, Rasheed Brokenborough may be the poster child for freshman ineligibility.

--Robert K. Fullinwider

Sources: Greg Garber, "Turning the Tables on the NCAA," GO.com.sports (March 10, 1999), <www.go.com>; John Chaney, "Losing Proposition," New York Times (March 10, 1999); Welch Suggs, "Fight Over NCAA Standards Reflects Long-Standing Dilemma," Chronicle of Higher Education, vol. 45 (April 9, 1999), <http://chronicle.com/weekly/v45/i31/31a04801.htm>; "Letter from Black Coaches Association and Fair Test to NCAA" (June 2, 1998); Robert Klitgaard, Choosing Elites (Basic Books, 1985), pp. 91, 96-7, 108; Jean H. Fetter, Questions and Admissions: Reflections on 100,000 Admissions Decisions at Stanford (Stanford University Press, 1995), p. 24; James Crouse and Dale Trusheim, The Case Against the SAT (University of Chicago Press, 1988), p. 216 and elsewhere (conceding that the use of the SAT along with high school class rank permits slightly more accurate predictions of academic success than use of rank alone); Cureton v. NCAA, Civil Action 97-131 (United States District Court, Eastern District of Pennsylvania, 1999); William G. Bowen and Derek Bok, The Shape of the River: Long-Term Consequences of Considering Race in College and University Admissions (Princeton University Press, 1998), pp. 59, 65; Edward Menaker, "Casualty of a Failed System," New York Times (October 3, 1982); Ross v. Creighton University, 740 F. Supp. 1319 (Northern District of Illinois, 1990); Danny Robbins, "A Question of Responsibility," Los Angeles Times (April 12, 1992); Bill Brubaker, "Minimum Standard, Maximum Dispute," Washington Post (July 25, 1999); "Race Discrimination Lawsuit Filed Against NCAA Test-Score Rules," Fair Test Examiner (Winter 1996-97); Diana Nyad, "How Illiteracy Makes Athletes Run," New York Times (May 28, 1989); Welch Suggs, "NCAA Says It Can Show ‘Direct Link' in Athletes' Test Scores and Graduation Rates," Chronicle of Higher Education (July 23, 1999); Tony Barnhart, "Crum's Reply on Eligibility," Atlanta Journal and Constitution (March 26, 1999); Gordon S. White, "Should Major Colleges Restore Rule for Freshmen?" New York Times (November 4, 1980); Craig Neff and Robert Sullivan, "Call for a Fresh Start," Sports Illustrated (December 22, 1986); Curry Kirkpatrick, "The Class of ‘92," Sports Illustrated (November 16, 1988); John A. DiBiaggio, "Don't Let Freshmen Play," Washington Post (June 22, 1999). 

About IPPP | Search Our Web Site | Publications | The Report
Links of Interest | Order | Comments | NCCR
For reprint permissions, contact Verna Gehring, Institute Editor: vgehring@umd.edu