Does college football impact the size of university applicant pools and the quality of entering students?

ABSTRACT

The role played by collegiate athletics in furthering the mission of institutions of higher education has been one of the more active research streams in the economics literature. Two areas of emphasis in this particular genre concern the relationship between athletics success and the size of a university’s applicant pool, and the relationship between athletics success and the quality of a university’s incoming class. This study extends both lines of research above by employing a unique panel data set consisting of 10 institutions that either added or eliminated college football between 1997 and 2015 in order to examine the impact of the presence of college football programme on both the size of university applicant pools and the quality of the students chosen for admission. Results from a panel data estimator presented here suggest that the size of their applicant pool shrinks the year following discontinuation of a college football programme. In the case of ACT scores, the results are similar, indicating that the ACT scores of incoming freshmen decrease after discontinuation of football.

KEYWORDS:

• Student quality
• university applicant pool
• college football
• ACT scores

JEL CLASSIFICATION:

• I21
• J24

Acknowledgement

The authors thank two anonymous referees for helpful comments on a previous version. The usual caveat applies.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ Pope and Pope (2009) find that football and basketball success significantly increase the quantity of university applications from 2% to 8%.

² Interestingly, a new study by Anderson (2017) employs data on bookmaker spreads to estimate the probability of winning each game for college football teams and finds that winning increases a university’s applications.

³ Smith (2008) finds that various measures of athletics success are unrelated to either the proportion of freshmen from the top 10% of their high school class, the number of freshmen with a high school grade point average of B or better, or the number of entering freshmen who bring the National Merit Scholar designation.

⁴ Mixon (1995) finds a positive relationship between incoming SAT scores and past appearances in the NCAA Men’s Basketball Championship Tournament. Mixon, Treviño, and Minto (2004) find a positive association between the 10-year winning percentage of college football programmes and the incoming SAT scores at their institutions. Results in Pope and Pope (2009) support these two studies. Smith (2008) finds that freshmen SAT scores are marginally related to one measure of basketball success. Lastly, Trenkamp (2009) finds a positive association between football success and median SAT scores, while a similar specification substituting basketball success for football success proves unfruitful.

⁵ The ACT is administered by American College Testing, which is a non-profit organization that was founded in 1959 by University of Iowa educators E.F. Lindquist and Ted McCarrel, who launched the ACT® test. The ACT test is now accepted by all universities and colleges in the US (www.act.org).

⁶ See www.collegeboard.org for more on the SAT, and www.act.org for more on the ACT.

⁷ The ratio for APSPOOL_it controls for the expectation that more students apply to larger institutions.

⁸ The number of undergraduate students is used to measure the numerator, which the denominator captures the number of faculty with doctorate degrees.

⁹ These statistics, and the subsequent regression estimates, are, given missing data for some of the institutions during the early years of the panel, based on an overall sample size of 101.

¹⁰ We are unconcerned with this result given that we employ a short panel while this test is typically reserved for long panels. Unfortunately, there are too few observations in the sample to properly conduct a cross-sectional dependence test.

¹¹ As before, there are too few observations in the sample to properly conduct a cross-sectional dependence test on the ACT specifications.