We focus on the "Graduation Success Rate (GSR)," which the NCAA acknowledges is an inflated metric that produces a "graduation rate" seven points higher than the "Federal Graduation Rate." The latter is the only metric that allows a comparison between athletes and non-athlete students. We do not discuss the Federal Graduation Rate here, but we have described it and its shortcomings in a previous paper. [2]  The purpose of this paper is to reveal and propose remedies to the significant statistical and other flaws in the NCAA's "Graduation Success Rate." This metric was invented just for athletes, and The Drake Group maintains it is a statistical "shell game" that results in an inaccurate and inflated portrayal of academic success; it is not a true measure of "graduation success."  An analysis of how the NCAA is "gaming" the data, as depicted in the chart below [3] , follows:

 

           FLAW #1    For a true assessment of institutional responsibility, ALL ATHLETES should be counted as Enrolled, and their graduation data should be disaggregated based on potential for exploitation. The federal definition of athlete enrollment is all full-time freshmen scholarship athlete students entering in September in the first year of the six-year cohort. Thus, the GSR does not include recruited athletes who do not receive a scholarship in September of their freshmen year but who subsequently receive a scholarship. Further, an athlete, recruited or not, who receives a scholarship at a later date, is not added to the cohort.   It is not uncommon, especially in NCAA "equivalency" sports, for athletes to enter without athletic financial aid and to subsequently receive aid.

 

          The Drake Group believes that if the NCAA wants an accurate graduation metric that reflects a sound educational athletics program AND knows that the Federal Graduation Rate, the only metric that allows a comparison with the general student body is flawed, it would commit to using an athletes-only metric that accurately reports retention and graduation.   Thus, we recommend removing these flaws and creating a statistically and philosophically credible NCAA Graduation Success Rate metric.

 

          The institution should be held responsible for all students participating in the athletic program, demonstrating that these students can successfully compete in the classroom and on the playing fields. To do otherwise allows institutions to (a) discard athletes who do not measure up to athletic skill standards, (b) run off lesser skilled athletes because more skilled prospects have been identified, (c) make unreasonable athletics-related time demands on less academically prepared students, causing them to drop out or leave due to academic failure, and (d) deny responsibility for athletes who transfer to other institutions because they are neither an academic nor an athletic "fit" for the program or have been treated unfairly.  In other words, the coach and the athletic program should be held responsible for retention AND graduation.

 

          To determine whether the institution has wisely recruited athletes capable of competing in the classroom, the graduation data must be disaggregated to permit careful examination of vulnerable populations. The NCAA must recognize that coaches and their institutions are tempted to ignore academically underprepared students through the use of "special admits' (athletes who are admitted even though they do not meet normal academic admission standards) or automatically admit special talents (such as those who receive scholarships for special talent in the performance arts).   Overall graduation rates hide sub-group deficiencies. Therefore, transparent reporting by sport is required, identifying the following subgroups:

            FLAW #2 Ivy League and Military Academy athletes should not be considered as scholarship recipients but should be included as recruited or unrecruited "walk-ons". Athletes attending Ivy League institutions and the U.S. military academies do not receive athletics related financial aid. To include this group in the NCAA's current metric is a dishonest reflection of the admission philosophies of these institutions.   To include these athletes in any scholarship athlete cohort also reflects an NCAA purposeful effort to inflate the aggregated graduation rate, since the graduates of these institutions graduate at higher than 90 per cent graduation rates.   They should, however be included among the sub-groups of either recruited or non-recruited ("walk-on") students.

 

            FLAW #3 False Counting of Drop-outs.   The NCAA removes from the denominator 24,298 scholarship athletes who leave an institution in good academic standing. Because not all of them enroll in another institution, their removal mathematically inflates t he GSR calculation by reducing the denominator. It can easily be determined that these "Left Eligibles (LEs)" are NOT passed from that school's cohort to another school's cohort as "Transfers In" because "Transfers In" (7,945) should equal "Left Eligibles"(24,298). Thus, it is clear that 65% or 16,353 of the Left Eligibles in the Division I 2015-18 cohort chart displayed are NOT passed to another school's cohort, but rather remain unaccounted for, and are very likely college dropouts. This is a sizeable number, and it causes the GSR rates to be significantly inflated by 7 to 11 per cent.[4] This flaw alone would indicate that the NCAA GSR metric is inaccurately reporting the academic success of athletes who leave by absolving the original schools of responsibility for failing to retain them. At the same time, the GSR adjusts for transfers out, which encourages institutions to push out athletes who might not graduate or who are easily replaced. This flaw too renders all dependence on the GSR invalid.

 

          FLAW #4. The institution should be held responsible for athletes it recruits even if the athlete's sport is dropped. To contend that the institution should not be responsible for athletes recruited to play or who receive scholarships, but whose sport team is then dropped, is irresponsible. The responsibility should continue through graduation.   

 

Comparison to the Federal Graduation Rate

          If the current NCAA GSR is corrected by counting all college athletes admitted in the freshmen year and by disaggregating graduation data by the subgroups suggested in the Flaw #1 discussion, the revised GSR for all subgroups can be compared to the Federal Graduation Rate for non-athlete students. If the athlete FGR by all subgroups is not 15 points to 20 points higher than the general student body FGR, [5] a careful examination of athletic program recruiting, admissions, and initial and continuing eligibility policies should occur.

 

Postscripts:

---