MIDDLEBURY, Vt. - The College Sports Project (CSP) has released a third round of analyses measuring academic outcomes for athletes and non-athletes at 84 NCAA Division III colleges and universities.  ”The data identify subgroups of intercollegiate athletes who do as well as, and sometimes better than, their non-athlete counterparts,” said John Emerson, Charles A Dana Professor of Mathematics at Middlebury College and the study’s principal investigator. “By examining various subgroups of students, college presidents can see which students and teams are doing well academically, and which may need attention and help.”

“As past President of one of the CSP institutions, I always found these reports interesting and instructive. I’m proud of Northwestern University’s role in providing a secure locale for collecting and managing this large data set and a supportive environment for the team headed by Rachelle Brooks that has performed these valuable analyses,” said Morton Schapiro, President of Northwestern University.

The data continue to indicate relatively modest differences in grade-point averages (GPAs) between female athletes and non-athletes. In contrast, male recruited athletes generally have lower GPAs than their non-athlete counterparts.

Analyses of data on 83,728 students in the 2005-06 and 2006-07 entering cohorts following their first and second years in college find that:

  • women athletes’ GPAs were higher than those of their male counterparts (for both recruited and non-recruited athletes)
  • non-recruited athletes of both genders had GPAs only slightly below those of non-athletes
  • students who were recruited by someone in the athletic department prior to their college or university admission had lower GPAs when compared with both non-recruited athletes and non-athletes; this finding is most apparent at the most highly selective colleges - where the status as a “recruited athlete” was known to the admissions office when applications were reviewed.

To facilitate comparisons of grades across a wide range of institutions, students were ranked within their own college cohorts on a scale from 1 to 100 based on the cumulative GPA. A ranking of 100 was assigned to the top student, and 1 was assigned to the student at the bottom of the ordered list of GPAs.

Percentile Class Rank GPA after Two Years: 2005-06 and 2006-07 Entering Cohorts

Student group

Athlete status

Count

Percentile rank

of the GPA

Difference in GPA ranks (athlete - non-athlete)

Male

Non-athlete

22,734

47

na

Recruited athlete

8,951

37

-10

Non-recruited athlete

4,142

43

-5

Female

Non-athlete

37,424

55

na

Recruited athlete

6,259

51

-4

Non-recruited athlete

4,218

53

-2

Note: Data from 84 NCAA Division-III Institutions in the College Sports Project; n=83,728

Analyses for subgroups of institutions based on their admissions selectivity indicate that students at a group of 24 highly selective institutions exhibit the greatest differences in grades between athletes and non-athletes, especially for male students. The differences in average GPAs for these athletes and non-athletes are shown below.  The average GPA for male recruited athletes is 15 percentile class rank units lower than the average for their non-athlete counterparts, and the corresponding difference for recruited women athletes is 9 units lower than for their non-athlete counterparts.  Highly selective colleges also exhibit the greatest difference in athlete to non-athlete comparisons for men recruited for the five sports that are the most highly recruited male sports at CSP institutions (basketball, football, ice hockey, lacrosse, and soccer); their average GPA rank is 17 percentile units lower than that of male non-athletes.

Percentile Class Rank GPA after Two Years: 24 Highly Selective Colleges

Student group

Athlete status

Count

Percentile rank

of the GPA

Difference in GPA ranks (athlete - non-athlete)

Male

Non-athlete

6,288

49

na

Recruited athlete

2,355

34

-15

Non-recruited athlete

1,452

44

-5

Female

Non-athlete

12,250

55

na

Recruited athlete

1,936

46

-9

Non-recruited athlete

1,938

52

-3

Note: Data from 24 NCAA Division-III Institutions in the College Sports Project; n=26,219

For the choice of academic major, recruited and non-recruited athletes of both genders are somewhat more likely to be social science majors and less likely to be humanities majors when compared to non-athletes. Differences in selecting a science/engineering major are modest, and they tend toward relatively fewer male athletes and relatively more female athletes with majors in these areas. Analyses planned for the fourth and fifth years, when most students have completed their studies, will examine the choice of major in greater detail.

The reports from the CSP to 84 presidents of Division III colleges and universities are the third in a series of annual analyses titled “Representativeness of College Athletes,” conducted as part of an ongoing longitudinal study of students at participating NCAA Division III institutions, and funded by a grant from The Andrew W. Mellon Foundation. The title refers to the goal, embraced by most Division-III institutions and conferences, that college athletes should be representative of their peers at their own institutions, especially with regard to academic outcomes such as chosen majors, grades, and graduation rates.

The latest reports present analyses of three groups of students who entered college in 2005-06, 2006-07, or 2007-08. All the students, athletes and non-athletes and including transfer students, are being tracked through the 2009-10 academic year, or until they graduate or leave college. Approximately 26 percent of the students at these (mostly small) colleges and universities are intercollegiate athletes, and around 18 percent of all students are recruited to play on athletic teams.

Emerson noted that various statistical analyses explore relationships among the variables available in the study. For example, preliminary analyses with regression models that examine relationships between GPA and many variables about the students and their backgrounds suggest that high school SAT scores, grades, the high school attended, gender, and race and ethnicity do not fully account for the differences in college performance between athletes and non-athletes. Instead a significant part of the difference may be ascribed to what researchers have termed “underperformance” by athletes. (The CSP defines underperformance for a student group as the difference between the average GPA observed for that group and the average GPA that is predicted based on the students’ test scores, grades, and other characteristics.)

In addition to the reports prepared for presidents of participating institutions, the research team also presented these findings at recent meetings of the American Educational Research Association (AERA) and the Association for Institutional Research (AIR).  “Disseminating information about the impact of athletics on educational outcomes to policy-makers and scholars in higher education continues to be a primary goal of the project,” says Rachelle L. Brooks, who directs CSP’s Center for Data Collection and Analysis at Northwestern University.

The College Sports Project represents colleges and universities in the NCAA’s Division III that are committed to strengthening the bonds between intercollegiate athletics and educational values. In addition to the data collection, the College Sports Project has also sponsored workshops for athletic directors, faculty, coaches and campus officials that are aimed at better integration of the academic, athletic, and student life dimensions of the college experience.

For more information, contact John Emerson at jemerson@middlebury.edu or Rachelle Brooks at rlbrooks@northwestern.edu.  Media contact at Northwestern University is Alan Cubbage, Vice President for University Relations, 847-491-4886.

— end —