Authors: Christopher P. Kelley, Shane D. Soboroff, Andrew D. Katayama, Mathew Pfeiffer and Michael J. Lovaglia
Christopher P. Kelley
2354 Fairchild Dr., Ste. 6L107
U.S. Air Force Academy, CO 80840-2603
Dr. Christopher P. Kelley is an Assistant Professor of Leadership in the Department of Behavioral Science and Leadership at the United States Air Force Academy. He studies complex organizations, leadership, power, and influence processes. Dr. Kelley also serves as the Managing Editor of the journal, Current Research in Social Psychology and is an active member of the American Sociological Association and the Society for Industrial and Organizational Psychology.
Institutional Reforms and the Recoupling of Academic and Athletic Performance in High-Profile College Sports
University officials and stakeholders continue to debate the role of athletics in the mission of higher education. Reforms promoted by the National Collegiate Athletics Association (NCAA) to promote academic integrity reflect this tension. This research investigates whether the most recent means for monitoring a team’s academic success, the Academic Progress Rate (APR), has led to changes in the academic and athletic outcomes of high profile football and basketball teams. Neo-Institutional theory provides a framework for understanding how regulations translate into organizational change through the coupling of organizational practices to institutional goals. Predictions that metrics used to assess academic progress among high profile student athletes will reflect increasing isomorphism among sports teams at the same school received support. Specifically, analyses of seven years of NCAA’s APR and athletic performance data found that APR scores became more similar among Division 1 programs, and increasingly correlated for high-profile sports within the same schools. Using Hallett’s ‘inhabited institutions’ framework and research on academic and athletic success factors, we also investigated whether improvements in APR could be attributed to coaches and if these changes impacted team athletic success, while accounting for resource differences between schools.Keywords: Institutional Theory, Organizations, Higher Education, Sports, Organizational Reform.
Academic reform efforts by the NCAA seek to promote two goals: student athlete academic achievement and athletic success in intercollegiate competitions. This study uses seven years of population level data comparing NCAA APR scores and following season wins for Division 1 football and basketball teams to assess the link between these two goals. The data indicate that academic reforms instituted by the NCAA are having a positive effect on athletic success for compliant teams. This suggests institutional pressures have led universities, coaches, and teams to adopt practices that, on their face, raise the priority of academic success for high profile collegiate athletic teams. The impact we found for the head coach on a team’s academic success suggests that team culture likely plays a role in the relationship between team-level academic progress and success in competition. Yet further research is needed to establish whether team-level academic success is being fostered by new practices that promote academics or, alternatively, if teams have found ways of appearing compliant while maintaining the centrality of athletics in player’s lives. Such research is important from a policy standpoint, because our initial analysis suggests that coaches’ focus on the team’s academic success increases the team’s competitive advantage on the field and on the court. This study, while limited in scope, is an important starting point for that discussion, because the data appear to support continued efforts at academic reform and suggest a link between success in the classroom and on the field. Future research is needed to establish whether reform achieves these effects in the manner intended— coercing coaches to increase the focus on the importance of player’s student role.
With extraordinary coaching salaries, national acclaim, and bonus structures that incentivize winning above all else, high-profile sports teams and universities have come under increasing pressure to address problems exemplified by highly publicized academic scandals. These scandals have led to the perception that college athletics are rife with rule violations and attempts to subvert regulations (Smith, 2015). This perception has led some critics (Sperber, 2000) to argue that the goals of athletic achievement and high academic performance are incompatible, especially in the high-profile sports of basketball and football. Partially in response to these criticisms, the NCAA instituted the APR to more precisely measure student athlete academic achievement. NCAA reforms suggest that the organization is aware that winning and providing student athletes with an education can become competing institutional values when the pressure to provide stakeholders with successful athletic teams is high.
Institutional theory (Meyer & Rowan, 1977) offers useful tools for analyzing the effects of APR reforms on the outcomes of high profile college sports programs. This perspective focuses on the relationship between regulation and organizational practices, and accounts for the actions of professionals, such as coaches and administrators, who are expected to embody institutional values. Based on institutional theory, hypotheses predict that changes to the regulatory structure of college sports will lead different organizations to adopt common practices and metrics for assessing performance, leading to isomorphism among these units (DiMaggio & Powell, 1983). Further, because changes must be reconciled with the technical requirements of routine activities, they ought to have an impact beyond those activities directly sanctioned by regulatory pressures. Coaches faced with increased pressure to adopt practices that encourage academic achievement among student athletes may seek to create a culture of accountability among students, backed by the resources of their university. Past research by Hallett (2010) on the recoupling of institutional values and organizational decision-making suggests that organizations under scrutiny by external audiences will attempt to provide concrete evidence of institutionalized myths, and that these attempts will shape organizational practice. This can have an impact both on college and high school practices, as there is some evidence that today’s student athletes are better prepared for academics out of high school and are better adjusted to college than their non-athlete peers in years concurrent with APR reforms (Melendez, 2009). While a causal relationship cannot be directly shown, understanding the APR as an institutional pressure that will diffuse through organizations due to regulatory compliance and professional pressures helps to put these changes in perspective.
One possibility is that increased discipline in the classroom leads to greater discipline on the athletic field. Research on college freshman retention shows that positive coping responses to stress and failure are important predictors of academic achievement (DeBerard, Spielmans & Julka, 2004). These traits are similar to those that sport psychologists also believe result in ‘mental toughness’ that promotes athletic success (Nicholls, Polman, Levy, & Backhouse, 2008). If, as Hallett suggests, the team culture put in place by a specific coach is responsible for promoting academic performance, then changes to APR ought to be predicted by coaching turnover. Further, if this culture is promoting academic practices that increase mental toughness, then teams that do better in the classroom ought to do better on the athletic field, with more wins following improvements in APR.
To test predictions, seven years of population-level panel data were collected, spanning the academic years 2004-5 to 2010-2011. Data included APR scores and season wins for 125 football teams in the Football Bowl Subdivision and the 214 basketball teams from corresponding schools competing within the same athletic conferences. Since access to resources and revenue could be expected to impact the success of reforms, and these resources are highly correlated with conference membership and NCAA subdivision, these memberships and designations were accounted for in our analysis. Our focus on high-profile sports as opposed to all sports at all levels is due to the impact scandals in high-profile programs played in spurring public attempts at reform.
The relationship between academics and athletic success on high-profile sports teams is a complex issue. Economic interests, academic demands, and the pursuit of prestige by colleges and universities have led to numerous studies that investigate the impact of college athletics on communities, universities, and students. Critical viewpoints on collegiate athletic programs suggest that high-profile sports teams are shaped by complex institutional and market pressures that contribute to problems within colleges and their communities (Pringle & Hickey, 2010). Among these problems are those of the athletes themselves, who may be caught in a structure intended to provide opportunity but instead may feel pressured to act against their own interests.
In 2012, a record number of college sports teams were penalized with postseason bans and loss of scholarships due to poor academic progress, mostly for Division 1 men’s basketball and football teams (Grasgreen, 2012). High-profile teams experience the greatest pressures to generate revenue, and so coaching jobs and salaries are most immediately tied to winning; These high-profile sports teams are also where problems of academic performance and related scandals are most likely to occur or be noticed (Gurney & Southall, 2013).
Universities may be slow to enact changes to athletic programs that confer rewards on their host schools. Lipschipz, Sauder and Stevens (2014) demonstrated that college athletics are institutionally linked to the prestige of universities, both affecting and benefit from the institution’s reputation. This relationship can be difficult for universities to manage because college athletics create market relationships with advertisers, merchandisers, and professional athletic associations. When teams span organizational boundaries and universities maintain a large field of stakeholders, instituitonal values can be difficult to define and entrenched practices more difficult to change (Lipschitz, Sauder, & Stevens, 2014). Benford (2007) charges that these conflicting interests lead college athletic programs to engage in an “active cycle of sports reform” without producing real change. Yet despite these very real concerns, there is little in the way of systematic longitudinal research to assess whether the goal of winning is incompatible with the goal of educating student-athletes.
There is some evidence that, across all sports, college athletes are at least as engaged and academically successful as other students (Beron & Piquero, 2016; Umbach, Palmer, Kuh, & Hannah, 2006) and that student and athlete roles need not conflict (Settles, Sellers, & Damas, 2002; Lucas & Lovaglia, 2005). However, as demonstrated in highly publicized scandals at the University of North Carolina in 2010 and Syracuse University in 2007, academic fraud remains a concern. The National Collegiate Athletic Association (NCAA) is charged with monitoring these concerns and sanctioning schools that fail to promote the values of amateur athletics and higher education.
The NCAA Academic Progress Rate Reform
The APR was developed by the NCAA to address shortcomings of graduation rate measures that failed to account for the distinctiveness of college athletes’ progress through higher education (www.NCAA.org). The NCAA argued the need for a measure that accounted for the tendency of student athletes to transfer schools before graduating or leave school to participate in professional sports. The NCAA contends that APR measures make it more likely that poor academic performance will be detected sooner and addressed. Sanctions that penalize teams through loss of scholarships for failure to meet minimum standards are meant as an incentive on coaches who are focused on winning to assure compliance. As a common metric for assessing the practices of very different organizations and teams, the APR serves as a form of commensuration, a way to compare different organizations using a common metric. Commensuration can create opportunities for organizational actors to game the system as they learn to master practices that affect the metric (Espeland & Stevens, 1998).
Smith (2015) found that, while APR sanctions were rare and did not negatively impact universities, sanctions did impact the success of individual teams. If sanctions affect team athletic success, then coaches are likely to support actions that increase APR scores for their teams. As such, the APR is as likely to pressure teams to at least appear compliant. Using an institutional approach, we assess how APR reforms impact measures important to academic institutions and athletic teams—student athlete academic progress rates and team wins.
Managing Institutional Demands
Contemporary institutional theory focuses on how organizations strive for legitimacy in order to survive. Organizations provide cognitive maps for participants, often in the form of organizational cultures (Meyer & Rowan, 1977). Institutional practices can become attractive when they are seen as an effective means for increasing legitimacy. However, these practices also become coercive when sanctions are imposed for non-compliance (Washington & Ventreska, 2004; Sauder & Espeland, 2009). This results in a formal structure that promotes perceptions of organizational actors as rational and legitimate. Institutional theory explains how organizations seek legitimacy and the appearance of rationality by adopting similar forms and practices as other organizations sharing common values.
Legitimacy for an organization can be with regards to the perception of the organization by people within the organization, such as when workers engage in rituals that reinforce their beliefs in the goals of an organization (i.e. academics attending commencement) (Meyer & Rowan, 1977). Alternatively, legitimacy can entail avoidance of sanction from external audiences or regulatory bodies (Scott, 2008). It is with this latter type of legitimacy that this study is concerned, the focus being on how institutions of higher education manage their relationship with the NCAA.
Institutional theory emphasizes the importance of organizational culture for understanding organizational decisions, practices, and stability. In situations where different institutions overlap and make competing demands on decision makers, such as college athletic departments (Buer, 2009), leaders may have difficulty explaining decisions that must take into consideration multiple stakeholders with varied agendas. These may include boards of governors, athletic boosters, alumni, or regulatory and accreditation agencies. The institutions of high-profile college athletics, academics, and the marketplace are often viewed as pursuing goals that only sometimes overlap. While each values the production of college graduates, their reasons may differ. For instance, athletic programs with poor graduation rates may receive sanctions or be viewed as illegitimate drains on institutional resources, and so graduates are needed to confer legitimacy on athletic competition by showing that educational and athletic goals are not incompatible. For academics, producing graduates is central to an institution’s educational legitimacy.
Unlike most other sports, football and basketball programs attempt to generate revenue and public support while coaches receive higher salaries that far exceed those of faculty and most administrators. For academics, professional activities are rationalized through goals of student educational and research productivity. For coaches, professional activities are rationalized through athletic performance and publically promoting the values of their host institutions. Academics and coaches work to promote institutional prestige, but these groups can differ on the rationale used in decision making (Buer 2009). The NCAA as an organization provides a framework for the institutional structure of collegiate athletics. The NCAA’s recent efforts at regulating academic standards for athletics has attempted to address the competing demands of the institutional logics of universities and sports teams (Pache & Santos, 2010).
Institutional theory explains that when multiple systems of institutional rationale are present, such as coaches making decisions about what values to emphasize for student athletes, organizational actors will act strategically to choose a system that reflects their own interpretation of available cultural frameworks and meets their most pressing concerns. In situations where the rationale for coaches’ decisions are defined by the goal of winning, organizational culture will reflect coaches’ beliefs about practices that are most likely to achieve that goal. In this paper we focus on whether there have been verifiable changes in APR and winning over time. While these changes could be the result of system gaming or compliance, institutional theory would implicate coaches as actors likely to impact either process.
Coaches and Compliance
Institutional theory offers different possible reasons why Division I universities might do better on APR measures and athletic success over time following reforms. While institutional change may result in conflict when taken-for-granted practices are questioned and changed, Hallett (2010) notes that leaders perceived as legitimate can help people accept new institutional norms.
We propose that NCAA reforms promote institutional myths of accountability. APR scores are public and certain, released every year, and allow easy analysis of trends to identify and sanction schools. In this way, the myth of accountability takes on “flesh” (Hallett, 2010). Regardless of any school’s individual response to external pressures, internally there is likely to be pressure on decoupled actors, coaches, to achieve at least the appearance of compliance so as to avoid sanctions. Responses of coaches at a university, regardless of the technical demands of their sports, are likely to become more uniform as colleges respond to institutional demands. Though the quality of uniformity would ideally be through the promotion of academic responsibility, institutional theory does not require that responses to external pressures be legitimate so long as the outcomes appear legitimate. Commensuration, by distilling different institutional responses into a common metric for comparison, can gloss over differences between legitimate and illegitimate means of achieving institutional goals. However, institutional theory is clear that external pressures are likely to prompt adjustments across sports within an institution to avoid potential sanctions by the NCAA. Hallett (2010) suggests this kind of pressure could result in administrators and compliance officers taking on a more central role in the lives of student-athletes. However, it is also possible that a common response to institutional pressures could lead to malfeasance on the part of administrators and coaches. Regardless, institutional theory would lead us to expect a change in the performance of organizations over time following reform pressures, and that coaches play a significant role in these changes.
Hypotheses tested in this paper assess changes in the academic and athletic performance of teams and the impact of coaches on these changes. Institutional theory leads us to predict a positive and increasing change in measures of academic progress over time, and an increasing correlation between the scores of high-profile teams in different sports at the same school. Further, this change should not depend on the resources available to schools in different conferences or divisions. While compliance may draw on different types of resources for schools with different levels of organizational resources, changes in the practices of all institutions would need to change for apparent compliance to increase. This is reflected in our first hypothesis:
Hypothesis 1. APR scores for football and basketball programs at the same school will become positively correlated over time regardless of conference or division designation.
The Relationship between APR and Athletic Success
Two possibilities lead to our second hypothesis. First, as suggested by a recoupling argument, coaches and institutions might work to emphasize academic success among players. This could have the effect of reducing athletic success as athletes spend less time focusing on athletics, or increasing athletic success if academic discipline leads to greater discipline on the field. A positive relationship between academic and athletic success might also be found if organizations learn to game the system and maintain the eligibility of highly talented athletes. As teams that appear non-compliant are sanctioned and lose scholarships, they are likely to have less success on the field, resulting in a positive relationship between meeting APR thresholds and athletic success. If reforms do advantage teams who appear compliant, we could expect an increasing positive correlation between APR and winning over time, leading to Hypothesis 2b.
Hypothesis 2a. APR scores are increasingly and positively correlated with a team’s wins over time.
Institutional effects of the APR reform appear intended to pressure coaches and athletic programs to attend to student-athlete academic achievement. Yet it is unclear how individual schools work with compliance officers, academic resources, and athletic boosters to provide monetary or technical support for student-athlete classroom achievement. We investigate the extent to which a resource gap between major universities and colleges play a part in whether a school successfully institutes academic reforms for athletes? Money may be spent on proper academic remediation of athletes, increased pressure on academic staff to maintain athlete eligibility or clustering of athletes in less rigorous courses.
A longitudinal analysis of APR data by Smith (2015) finds that sanctioned teams won fewer games in subsequent seasons, though this may not impact the revenue or status of the institutions receiving sanctions. If money is being diverted from non-athlete education to subsidize athlete remediation, or non-athletes are assessed increasing student fees to subsidize athletics, this could account for a change in APR without necessarily affecting change in the cultures of athletic programs. These programs would simply have more resources to monitor athletes’ academic performances. Institutional theory does not claim to predict the precise manner in which organizations will negotiate their regulatory environment. An organization’s ability to loosely couple its practices to the demands of regulatory agencies allows the organization to tailor practices to its unique needs and goals.
If college athletics amounts to an “arms race” to provide athletes with more resources, such as tutors, to attract potential recruits, then the largest and historically most athletically successful universities should have an advantage recruiting, remediating, and retaining top talent. In Division 1 football, specifically the Football Bowl Subdivision, these teams are usually members of the so-called “Power Five” conferences—the South Eastern Conference, the Big Ten, the Pac 12, The Big 12, and the Atlantic Coast Conference. According to institutional theory, a common set of regulatory demands should have a similar impact across organizations governed by institutional regulations. However, some scholars have suggested that resource differences are to blame for current differences in APR and wins may still exist between schools with different resource levels. It is possible that resources might affect both APR and wins and account for the relationship between them. We control for a possible spurious correlation between APR and wins that results from differences in team resources by controlling for a team’s membership in a ‘Power Five’ conference, as these teams generally have higher revenues, booster support, and budgets.
Hypothesis 2b. The relationship between APR and wins will be similar for members of high-profile “Power Five” conferences and other Division 1 Football Bowl Subdivision programs.
Finally, if a coach is partly responsible for organizational responses to reform pressures, then coaching changes may be disruptive to the academic progress and athletic success of a program, especially if the new coach is under pressure to be successful immediately within a high-profile institution. The uncertainty created by the loss of a coach will likely negatively impact any response as the organization adjusts to new leadership.
Hypothesis 3. Head coaching changes during or after a season are negatively related to APR for Division 1 football and basketball programs in the following academic year.
By using population level data we attempt to assess the general effects of the APR on the relationship between high-profile sports teams and the educational mission of their host institutions. Analyzing seven years of data from the same institutions allow us to investigate the development of a relationship between APR on teams at the same school and the impact of reforms on long-term academic and athletic success at NCAA member institutions.
Two datasets allow us to test hypotheses 1-3. The first is a dataset we constructed over the academic years 2003-2004 to 2009-2010 that included team APR for Division 1 FBS football and Division 1 basketball programs. We added additional data from 89 Division 1 basketball teams, including APR scores, wins, conference affiliations and coaching changes to the dataset for analyses of the effects of APR on basketball programs alone. Hence, our dataset for testing hypotheses 2-3 consists of 120 football teams and 209 basketball teams. Over 7 years of panel data, this yields 1463 observations. This dataset reflects APR as originally reported by the NCAA (www.NCAA.org), and prior to the institution of a higher 930-point threshold in 2011.
The second dataset comes from Paskus (2014), including publically available data on APR scores, eligibility, awards, conference affiliation, and subdivision designation. These data contain FCS football teams that are missing from our original dataset. Analyses of these data are warranted because the NCAA has imposed APR reforms on all schools, regardless of subdivision. Within these data, 125 teams are designated FBS and have both a Division 1 football and basketball team, reflecting the upward movement of 5 teams since the last year of panel data in our original dataset. The dataset also contains 123 football teams and 124 basketball teams at FCS schools. For football, this yields ten years of APR data for 248 football teams (2480 observations) and 249 basketball teams (2490 observations). These data are used in an analysis of changes in APR by FCS teams when compared to FBS teams. While the relationship between APR and winning is likely different between these two divisions, an institutional argument would still be supported if APR has risen significantly across the years of our panel data. While our focus is on teams with FBS football teams, we investigate whether there are differences in APR scores between the so-called “Power 5” conferences and the rest of the FBS. In this way, we are comparing schools at similar levels of prestige and with similar schedules. The inclusion of FCS schools in our data would make it difficult to assess whether APR is associated with winning in the highest-profile teams because at least some wins by FBS schools are likely to come at the expense of FCS competition.
Analyses in this study focus only on male athletes in high-profile revenue generating sports. Compared to other programs, athletes in Division 1 football and basketball have historically performed poorly academically. If APR reforms are targeted toward impacting these sports, then focusing our analysis on football and basketball is warranted. This is not to say other sports have not been impacted by APR reforms. Rather, the 50% graduation rate threshold approximated by the APR has historically been more difficult for football and basketball teams to achieve, and so any sign that APR reforms have affected the academic progress of these student athletes should be assessed.
The academic and athletic trajectory of schools over time will likely have an impact on the effectiveness of NCAA reforms and likely account for a significant portion of yearly APR and season wins. While isomorphic pressures are likely to affect only those teams appearing non-compliant in early years of APR reforms, we include all teams in our analysis and include school level controls to account for heterogeneity in organizational responses to reforms. For example, a team that appears compliant will likely not invest huge resources to achieve compliance in response to a single poor APR score in a given year . Generalized linear regression models with lagged variables for previous season wins and APR tested hypotheses. A unique identification code for each school was included in both of the datasets and entered as the subject variable for our analysis. APR Year was the within-subject panel variable for repeated measures. Institutional theory would acknowledge the unique culture likely to emerge within any given institution over time. Controlling for the idiosyncratic variance in effects of APR reforms within schools is, therefore, necessary since the ways in which actors manage the messages implied by these reforms may differ widely across universities. A generalized linear model with repeated measures can allow us to assess the effects of APR across panel years, net of other factors such as the previous year’s wins.
Variables in Analysis
Academic progress rate. Each athletic team’s APR score is based on a 1000 point scale with a 925 point minimum standard. This standard represents an approximate 50 percent graduation rate under current federally mandated methodology. The APR calculation allocates points for eligibility and retention of current student-athletes. Every player on a team’s roster who is receiving aid for athletic participation earns a maximum of two points per term. APR scoring awards one point for academic eligibility during the academic term and one point if the athlete remains enrolled and eligible at the institution. Each team’s APR scores includes the total points awarded to the team overall divided by the total points possible, multiplied by 1000. APR scores were transformed into lagged variables in analyses investigating the impact of coaching changes on current year APR. Teams falling below a threshold meant to reflect a 50% four-year graduation rate—initially 925, 930 as of 2011—may lose up to 10% of scholarships and be banned from the postseason (Brown, 2005).
Season wins. We used season wins as the measure of athletic success, with lagged season wins used as a control variable. The ESPN website provided these numbers. It should be noted that season wins were only used to compare teams at the Division 1 FBS level, meaning that any positive relationship between APR and wins may be coming at the expense of FCS-level opponents played early in a season. In essence, this decision truncates the values of our dependent variable. This is not necessarily problematic for the current analysis. The question we seek to answer here is whether FBS schools are responding to the pressures of these reforms and if, by seeking to improve athlete academic progress rates, they are winning more often. The issue of how these wins are being generated is a separate question and an important topic for future analysis to evaluate the impact of reforms on all Division 1 teams. A future analysis could control for the number of FBS level opponents played in each year by schools in Division 1 to see if team wins are coming at the expense of smaller schools FCS schools.
Coaching change. The NCAA website provided coach turnover data for each year of APR. Coaching change was entered into the data as a dummy variable for each year, with 0 = no change and 1 = new coach. Rather than look at the year-to-year impact of a coaching change, we investigate whether any coaching change within a program over the panel years affected athletes’ APR scores.
Power 5 member. Teams that belonged to the ACC, Big Ten, Big Twelve, Pac-12, or SEC were designated “Power 5” teams. This was represented in our data set by a dummy-coded variable, where 1 = Power 5 Team and 0 = Non-Power 5 Team. This variable was used to investigate whether a team’s status as a member of elite conferences accounted for any positive relationship between APR and wins.
NCAA subdivision. Football teams in the Paskus (2014) data were designated as either FBS (large, well-funded public and private universities competing for bowl bids) or FCS (lower-tier regional or state schools, Ivy League, Historically Black Colleges and Universities). FBS schools were coded 1 while FCS schools were coded 2.
Descriptive statistics suggest general trends within our original dataset (2004 to 2010). We use this dataset for all analyses except when investigating differences between FBS and FCS team APR scores. The first year average APR for basketball was 922.04 (SD = 60.527), meaning that on average, basketball teams in 2004-5 fell below the minimum accepted APR of 925. Football teams averaged 925.20 (SD = 32.969) in APR for the same year. By year 6, 2009-10, APR scores increased on average. For the final year of our original dataset, basketball team APR averaged 948.56 (SD = 32.421), and football team APR averaged 946.56 (SD = 19.780).
Table 1 includes basketball and football APR scores for teams with the lowest APR scores in the first year of data, 2004-5. For both basketball and football, the teams with the lowest APR scores saw increased academic progress for athletes from year 1 to year 6 of data. For basketball teams, only 4 of the lowest scoring programs achieved APR scores above the minimum APR of 925, meaning that six of these programs faced sanctions following their implementation after the 2005 season. For football teams, 7 of 10 programs in the bottom ten of APR scores in the first years of APR data achieved the minimum APR of 925 by 2009-10.
In 2003-4, 64 Division 1 (FBS) football teams received scores below the minimum acceptable APR. By 2009-10, only 17 teams had scores below 930. Similarly, the average number of wins for non-compliant teams decreased. Whereas the average number of wins for teams in the non-compliant category in 2003-4 was 6.44, the average win total for non-compliant schools in 2009-10 was 5.12. For compliant teams (those with APR above 930), the average number of wins in 2003-4 was 6.56, while the average in 2009-10 was 6.97. This shows an increasing difference in favor of compliant teams from the first year to the last year of our data. Thus, the average team in 2003-4 would have been eligible for a bowl game and the national exposure that bowl games entail, whereas the average non-compliant team in 2009-10 was not bowl eligible—a potential hindrance to team recruiting and marketing efforts. Among teams represented in this at-risk category were current national powerhouses such as Alabama, LSU, Michigan State, Ohio State (which had just won a national championship in 2002), and Oklahoma (national champions in 2000). Thus, it is apparent that winning and APR were very loosely coupled in these early years of APR reforms. By 2009-10, none of the above mentioned teams received a non-compliant APR score. Several teams were listed as non-compliant in both years, such as Akron and Washington State. Interestingly, both teams were bowl eligible in 2003-4, with 7 and 10 wins respectively. However, in 2009-10, neither team was bowl eligible, with Akron winning 3 games and Washington State winning only one. While this cannot serve as definitive evidence that APR and winning are linked, these trends offer credence to the process described in the theory section above.
For basketball, similar trends appear. In 2003-04, 107 NCAA Division 1 basketball teams received APR scores below 930, while 102 teams were compliant. This included 2004 National Champions Connecticut, which won 33 games that season with an APR of 889. By 2009-10, only 46 teams fell into the sub-930 APR group. Interestingly, the average number of wins for compliant teams increased from 16.75 in 2003-4 to 18.77 in 2009-10. The average number of wins for non-compliant teams did not change significantly (M = 16.10 in 2003-4 vs. M = 16.20 in 2009-10). Further, a team that appeared compliant and won the national championship in 2009, North Carolina later faced censure after perpetrating academic fraud by placing student athletes into non-existent programs. Importantly, North Carolina would have been subject to isomorphic pressures due to a sub-930 APR score in 2003. This shows that teams may take a number of routes to achieve the appearance of compliance, and shows why generalizations relating academic performance to athletic success come with important limitations. However, compliant teams do appear to win more often after the implementation of reforms than prior to the implementation of reforms, as an institutional perspective would suggest.
While the overall data for both basketball and football suggest general improvements in student-athlete academic progress and an impact of APR on team athletic success, basketball and football likely face different pressures given the different number of scholarships offered, the practices of stakeholders such as professional sports organizations, and the needs of coaches and athletes. However, if isomorphic pressures are affecting both football and basketball teams at the same school, the APR of basketball and football teams at the same school would likely have an increasing and positive correlation over time.
Hypothesis 1: Correlation of Basketball and Football APR
We calculated bivariate correlations between Academic Progress Rates of basketball and football programs at the same school for both Division 1 football and Basketball. Hypothesis 1 stated that the correlation between the academic performance of basketball and football players at the same institution would become positive and increase over time. Table 2 presents Pearson’s r correlation coefficients for the correlation between football and basketball APR in each year of APR collected.
In 2004, the year after the APR was instituted, there was not a significant correlation between APR scores of basketball and football teams at the same school (Pearson’s r = .167, p = .07). Two years after the APR was instituted in 2005, the correlation is positive and significant (Pearson’s r = .240, p = .009), and remained so throughout the panel years of our data. In general, the correlation between these scores increased each year. These results support hypothesis 1 and suggest that compliance pressures are affecting the classroom performance of athletes in different sports.
It is possible that the improvements in APR for football teams are only seen among Division 1 FBS schools. Within the Paskus (2015) data, paired-samples t-tests allow us to compare APR scores in the last year of data to APR scores in the first year for each team in Division 1 (FBS and FCS). Table 3 presents means and standard deviations for APR scores in 2004 and 2014 for schools in both FBS and FCS, as well as the results of paired-samples t-tests. The average APR for Football Bowl Subdivision teams in 2014 (M = 966.45, SD = 18.254) is significantly higher than the APR average for these teams in 2004. Importantly, it should also be noted that the standard deviation of APR scores is also smaller in the latest year of data. This suggests that the strategy of using a generalized linear model to account for heteroscedastic data is appropriate. While there is no FBS/FCS designation within Division 1 basketball, basketball APR scores have also increased for teams in Division 1 with FBS football teams. It is possible that FBS schools can spread greater resources between revenue generating sports to improve the academic standing of athletes. However, even with resource disparities between FBS and FCS schools, analysis of football and basketball APR scores for teams at FCS show improvements across panel years. This suggests that increases in APR scores are occurring despite differences in resources.
The average FCS football program and basketball programs at the same school have APR scores above the minimum 930 required by the NCAA . The average FBS school’s basketball team, however, fell below this threshold in the first year of the APR. FCS schools do not, on average, have APR scores indicating a graduation rate below 50%. This highlights one reason why we focus on FBS Division 1 schools in tests of hypothesis 2. It is certainly possible that wins are being generated by larger schools defeating weaker opponents with fewer resources to remediate players. Another reason for focusing on FBS schools to test the impact of APR on wins is the potential for wins during the regular season to translate into an extra opportunity to win in the post-season. Teams at the FBS level often play more regular season games, participate in conference championship games and travel to compete in high-profile bowl games if they perform well. As such, we compare only teams within FBS, and control for membership in a Power 5 conference (that often have deals with major bowl sponsors) to compare teams at similar levels of competition. Rather than make claims about the impact of APR on winning for sports teams overall, we test here whether the idea behind the APR reforms—to incentivize student-athlete academic achievement by threatening non-compliant teams with losing competitiveness on the field—has impacted the highest-profile teams.
Hypotheses 2a and 2b: Correlation between APR and Season Wins
Pearson’s r was calculated to test whether higher APR scores were correlated with season wins. Those teams whose classroom performance is higher would be less likely to lose players to academic ineligibility and be more competitive on the field or court due to increased discipline. They would also be more likely to retain and attract high-level recruits that can improve their winning percentage year to year. Therefore, we predicted an increasingly positive correlation between APR and wins over time. Correlations were done for both football and basketball for every year of APR data (See Table 4).
As expected, there is no significant correlation between football APR scores and winning percentage in 2003-2004 (Pearson’s r = .179, p > .05). However, in each subsequent year, APR is positively and significantly correlated with season wins (see Table 3). With the exception of the 2006-2007 academic year, the correlation between APR and season wins for Division 1 football teams increases from year to year, and is positive and significant in the latest year of data (Pearson’s r = .337, p < .001). These results support Hypothesis 2a.
Results for basketball are similar, though as seen in Table 4 the correlation between APR and wins is not significant until the fourth year of data. Football and basketball programs have different technical demands, given different numbers of scholarship players and different rules regarding player moves from college to professional teams. Unlike football players, basketball players are only required to spend a year in college before beginning a professional career. Basketball is also more likely to recruit junior college players to fill key positions open due to the graduation of key players, player academic ineligibility, or movement of players into professional leagues. In general, teams with higher APR scores appeared to win more often. Many variables may affect the athletic success of a particular team, but Academic Progress Rates are calculated for all teams, offering a common point of comparison for each team in the sample.
We conducted generalized linear regression analysis to analyze the relationship between APR and wins, treating time as a panel variable and teams as the subject variable. Analyses reported here predicted the number of wins in a season (rather than the winning percentage) that a team recorded, regressed on the academic progress of those teams. Over the panel years, the average number of wins and APR scores in the population increased, while overall variance on these variables decreased. Our models were adjusted to account for heteroscedasticity. A generalized linear model predicting season wins that accounted for within-school and across-time effects was estimated, with APR as an independent variable. APR is released in each year prior to the start of that year’s football and basketball seasons, and so represents a naturally lagged variable within our dataset. Additionally, a lagged variable for previous season wins was created and entered into the model as a control variable.
Table 5 shows results for football, and Table 6 presents results for basketball. Past winning record significantly and positively related to the following season win total, as expected. Strong support was found for hypothesis 2a. For football, APR predicts season wins, and the effect is positive and significant (B = .015, s.e. = .003, p < .001, two-tailed). Higher APR scores are positively related to more wins, controlling for past winning records. While the coefficient appears small, there is large potential variance in APR scores. The positive and significant coefficient means that as APR increases across time by 1 point, a team’s winning record goes up by .015 wins per season. An advantage of 100 points in APR would amount to a 1.5 more wins over the seven years of panel data, net of prior winning record. Since teams with initially lower APR tended to have greater increases in APR over the panel years, this means that they became more competitive. While this may not seem like much of an increase, a single win can determine bowl eligibility or the inclusion of a team in a major bowl game. Further, it suggests that universities that fail to incentivize the academic performance of student-athletes are disadvantaging their teams in competition.
Surprisingly, belonging to a Power Five conference did not provide a significant improvement in wins for football teams relative to non-Power Five teams (B = -.328, s.e. = .192, p = .087, two-tailed), though findings are in the direction of an advantage for Power Five teams. Results are similar for basketball. Over time, increases in APR are associated with more wins (B = .029, s.e. = .006, p < .001). Again, this means that a difference of 100 points in APR over the years of panel data is predicted to result in 2.9 more wins over those years. Duke’s basketball team, with a four-year average APR of 995 in 2014, could expect to win 4.6 more games than Alcorn State (APR = 839) on the basis of academic progress rate differences alone. There is evidence that basketball teams at Power Five football schools do have an advantage, winning nearly 2.5 more games from season to season than basketball teams from non-Power Five schools (B = -2.417, s.e. = .767, p = .002, two-tailed). However, these analyses support the interpretation that the academic progress of student-athletes has a positive and significant relationship on their athletic achievement. Hypothesis 2b receives support in our analysis of basketball wins, with Power Five teams winning significantly more often. However, there is still a net positive effect of APR on wins when controlling for this effect.
Hypothesis 3: Coaching Change and Subsequent Academic Progress Rates
Hypothesis 3 predicted that coaching changes would be negatively related to APR in subsequent seasons. Again we estimated a general linear model controlling for within school and across-year variance. Controlling for previous season Academic Progress Rate, a change of coach was significantly and negatively related to subsequent season APR for basketball programs (B = -5.41, s.e. = 1.39, p = .000, two-tailed). No significant relationship is found between coaching changes and APR for football programs. These findings are presented in Tables 7 and 8.
It is possible that the smaller scholarship roster of a basketball program makes a coach’s attention to creating a team culture emphasizing academic success more important in basketball programs, especially if losing a coach leads to the loss of high-quality student recruits or current players transfer to competitor schools in the off-season. These data provide further support to an institutional theory explanation for how team APR is achieved. These data suggest that basketball programs are particularly vulnerable to academic setbacks when a coach is fired or leaves.
The APR reforms have been promoted by both the NCAA and college administrators as providing a real incentive for positive change in the academic cultures of college sports. Based on institutional theory we proposed that organizations respond to institutional rule changes by generating new expectations and norms for decision makers. This can result in “recoupling” of academic practices in athletic programs to the expectations of the institution of higher education. The manner in which this recoupling occurs is not easily identified, but must be addressed, as theory development above suggests that recoupling will include cultural changes in high-profile sports teams.
By creating the APR, the NCAA assumed that common pressures on athletic departments can reduce disparities in student-athlete classroom performance between sports and among universities. The NCAA argued that reforms could reduce predatory recruiting practices and exploitation of student athletes solely for their athletic prowess. The apparent goal of the NCAA over time has been to give the institutional myth of the student-athlete more tangible “flesh.” This study assessed whether results of these reforms, proposed by the NCAA as evidence for the success, are present in FBS football schools and Division 1 Basketball schools.
Hypothesis 1 predicted that APR scores for basketball and football programs at the same schools would become positively and significantly correlated over time. This hypothesis was supported, even though basketball and football teams face technical pressures that are unique to those sports. Findings provide evidence that institutional demands are affecting both types of programs and that football and basketball teams are responding to similar institutional demands. Further, results of paired-samples t-tests show that effects of these reforms on the APR scores of teams are not limited to the highest levels of competition. Teams at both the FBS and FCS levels achieved significantly higher APR scores over time, even though FBS schools’ basketball teams were the only teams for which the average APR fell below the 925 minimum score in the first year of APR reporting. The broad impact of reforms on programs that were not identified as having academic problems supports an institutional explanation. Specifically, it makes little sense from a purely athletic or economic standpoint for smaller institutions to pursue improvement in APR given how few programs (and almost all of these are in Power 5 conferences) produce a profit. It instead may be the case that prestige is the primary concern of all but the most elite programs, and this would not be served by revelations of widespread scandal. However, we cannot rule out the possibility that a new culture for gaming the APR has developed in spite of the desire of coaches and players for integrity within sports (Cullen et al., 2012). As noted by Washington and Ventresca (2004), the way in which institutional pressures relate to the adoption of institutional logic-consistent practices may vary across institutions and may require qualitative research to uncover. Interviews with players and coaches could help uncover when ‘recoupling’ of organizational practices with legitimate institutional goals has occurred, or when institutions have taken advantage of the commensuration process provided by the APR to make adjustments that are not in the spirit of academic reform.
The APR reform was meant to sanction teams that failed to promote student-athlete classroom achievement. If successful, such sanctions would be associated with lower athletic performance through the loss of scholarships or key personnel to academic ineligibility. Higher APR scores should have a positive relationship with winning. Hypothesis 2a received support; APR was positively and significantly correlated with the number of wins in each year except for the first. In general, this correlation became more significant over time. These results are only from analyses of FBS football teams, though all basketball teams designated as Division 1 during the years of our panel data were included. This replicates similar findings by Smith (2015). Further, Generalized linear regression analysis found that higher APR in each year was positively associated with season wins in subsequent seasons, controlling for the past winning record. Teams that performed better over time in the classroom won more often in subsequent years. These results are important because it suggests that academic success and athletic performance need not be mutually exclusive goals for college athletes.
Coaching changes in basketball took a major toll on team APR. This suggests that coaches have a significant impact on the culture of these teams and are important institutional agents. Pressures on these coaches to win rather than promote student academic achievement, and turnover among these coaches, may work against the best interests of both the athletes and their teams. These data suggest that colleges that incentivize student-athlete academic progress and maintain coaches who promote this progress may not be at a disadvantage in competition. The question remains how these data reflect actual changes within institutions. The nature of that change deserves further scrutiny that is beyond the scope of this paper. While the relative levels of academic achievement between larger and smaller schools’ teams may remain, absolute levels of academic performance have improved for high profile football and basketball teams.
These data provide indirect support for the assertion that APR reforms promote organizational change. Recoupling is evidenced by the increasing and positive correlation between basketball and football APR within institutions. Improvements in APR are seen across conferences, and these improvements are correlated with more wins across divisions and conferences despite the advantage of Power 5 teams in athletic competition. Finally, there is some support for the notion that coaches—especially basketball coaches—act as institutional agents, and their removal is detrimental to the academic progress of their former teams. Future research could investigate the professional networks of athletic directors and coaches to identify whether shared methods for achieving compliance are spread through these associations and the characteristics of these methods.
When coaches and schools prioritize winning at the expense of educating student athletes, they are likely violating their obligations to educate these students and assure the team’s competitiveness. This is especially true in basketball, where both academic progress and leadership stability were shown to have a major impact on the academic and athletic success of teams. Despite differences in resources, schools that prioritize athlete academic progress and limit coaching turnover have advantages relative to those that do not during athletic competition. Despite evidence that colleges suffer little reputational damage when they are sanctioned for failure to comply with NCAA rules and that relative rarity of sanctions (Smith, 2015), the average team APR for basketball and football teams at every level of competition has increased significantly over time. The case of Michigan football helps to illustrate these relationships.
Michigan football is an example of a team that, despite a compliant first APR score, experienced a downturn in academic progress for its student athletes over time. Michigan won 10 games in 2003-4 with a 949 APR. In 2007, long-time coach Lloyd Carr retired, and Michigan hired Rich Rodriguez to replace him. Not only did Rodriguez fail to win, but APR at Michigan suffered to the point that by 2009-10, Michigan was in danger of facing sanctions for an APR score of 928, below the then-required 930-point threshold. After firing Rodriguez and hiring Brady Hoke, Michigan’s troubles on the field continued, as did suggestions that Michigan football did not prioritize student athlete academics. In 2014, Mark Snyder of the Detroit Free Press speculated that one reason Jim Harbaugh, a Michigan alum, replaced Hoke at Michigan was his focus on scholastically capable athletes, as evidenced by his academic records as the head coach at Stanford University and San Diego. Snyder noted that Harbaugh’s record in terms of APR scores showed that his teams consistently improved their APR scores from the beginning to the end of his coaching tenures (Snyder, 2014). Also notable is that Michigan achieved a 993 football APR score in 2015-16, enough for it to rank 3rd nationally behind Northwestern University (995 APR) and the U.S. Air Force Academy (995 APR) (Snyder, 2017). Michigan also achieved bowl eligibility in each year of Harbaugh’s tenure. The case of Michigan from 2007 to the present offers an example of the process suggested by this paper and supported by the data: coaches can have a significant impact on the academic success of student athletes, and the academic success of a team in response to NCAA reforms can affect team success on the field of competition.
However, we must caution against interpretations that go beyond noting a positive relationship between APR scores, coaching changes, and athletic success. Given the apparent compliance of teams such as North Carolina’s basketball program and later evidence of widespread fraud within the UNC athletic department, it is possible for teams to take a number of routes toward appearing compliant without maintaining academic integrity. However, the institutional perspective we take in this paper suggests that the decisions made regarding how best to achieve compliance with regulatory pressures will represent on-the-ground responses to a complex of local organizational pressures, salient audiences and stakeholders, professional networks, and institutional values. While some proportion of athletic departments are likely guilty of gaming the system and harming both their schools and athletes, evidence of increased compliance itself cannot be enough to conclude that such fraudulent practices are widespread. The case of Michigan suggests that prioritizing academics can work to the benefit for both schools and their athletes. However, the prevalence of either approach in specific athletic departments is an empirical question requiring further research.
Our data suggest that organizational change is likely occurring as result of institutional level efforts at academic reform, and these changes are affecting student academic progress, with student athletes remaining eligible for athletic competition even in the highest profile sports. Future investigation of the actual practices of athletic programs can shed light on how organizational change occurs in specific university settings. Importantly, these findings should be a signal to coaches that failure to comply with reforms may result in future disadvantages in athletic competition, and that discipline off the field can result in discipline and winning on the field. These findings also serve as evidence that continued efforts at reform are a necessary, though imperfect, tool if big time sports are to remain part of higher education.
1. Adler, P. A., & Adler, P. (1988). Intense loyalty in organizations: A case study of college athletics. Administrative Science Quarterly, 33(3): 401-417. http://dx.doi.org/10.2307/2392716
2. Beron, K. J., & Piquero, A. R. (2016). Studying the determinants of student-athlete grade point average: The roles of identity, context, and academic interests. Social Science Quarterly, 97(2), 142-160. https://doi.org/10.1111/ssqu.12235.
3. Buer, T. (2009). Organizational complexity: The athletics department and the university. New Directions for Higher Education, 148, 109-116. https://doi.org/10.1002/he.374.
4. Christie, A.M., & Barling, J. (2010). Beyond status: Relating status inequality to performance and health in teams. Journal of Applied Psychology, 95, 920-934. http://dx.doi.org/10.1037/a0019856.
5. Cunningham, G. B., & Ashley, F. B. (2001). Isomorphism in NCAA athletic departments: The use of competing theories and advancement of theory. Sport Management Review, 4, 47-63. https://doi.org/10.1016/S1441-3523(01)70069-1.
6. DeBerard, M. S., Spielmans, G., & Julka, D. L. (2004). Predictors of academic achievement and retention among college freshmen: A longitudinal study. College Student Journal, 38(1), 66-80.
7. DiMaggio, P. J., & Powell, W. W. (1983). The iron cage revisited: Institutional isomorphism and collective rationality in organizational fields. American Sociological Review, 48(2), 147-160. https://doi.org/10.1177/1056492608326693.
8. Espeland, W. N., & Stevens, M. L. (1998). Commensuration as a social process. Annual Review of Sociology, 24, 313-343. https://www.annualreviews.org/doi/full/10.1146/annurev.soc.24.1.313
9. Fountain, J. J., & Finley, P. S. (2011). Academic clustering: A longitudinal analysis of a division I football program. Journal of Issues in Intercollegiate Athletics, 4, 24-41.
10. Grasgreen, A. (2012). You’re not going to Disney World. Inside Higher Education. Retrieved from https://www.insidehighered.com/news/2012/06/21/ncaa-bans-teams-postseason-low-apr-scores.
11. Gurney, G. S., & Southall, R. M. (2013). NCAA reform gone wrong. Inside Higher Education. Retrieved from https://www.insidehighered.com/views/2013/02/14/ncaa-academic-reform-has-hurt-higher-eds-integrity-essay.
12. Hallett, T. (2010). The myth incarnate: Recoupling processes, turmoil, and inhabited institutions in an urban elementary school. American Sociological Review, 75, 52-74. https://doi.org/10.1177/0003122409357044.
13. Killeya-Jones, L. A. (2005). Identity structure, role discrepancy and psychological adjustment in male college student-athletes. Journal of Sport Behavior, 28, 167-185.
14. Lapchick, R. (2010). Keeping score when it counts: Graduation rates for 2010-11 bowl bound college football teams. Retrieved from http://tidesport.org/Grad%20Rates/2010-11_APR-GSR_BowlStudy.pdf.
15. Melendez, M. (2009). Psychosocial influences on college adjustment in division 1 student athletes: The role of athletic identity. Journal of College Student Retention: Research, Theory, & Practice, 11(3), 345-361. https://doi.org/10.2190/CS.11.3.c.
16. Meyer, J. W., & Rowan, B. (1977). Institutionalized organizations: Formal structure as myth and ceremony. American Journal of Sociology, 83(2), 340-363. https://doi.org/10.1086/226550.
17. National Collegiate Athletics Association. 2013. Trends in graduation success rates and federal graduation rates at NCAA division I institutions. Retrieved from https://www.ncaa.org/sites/default/files/GSR%2Band%2BFed%2BTrends%2B2013_Final_1.pdf.
18. Nicholls, A. R., Polman, R. C. J., Levy, A. R., & Backhouse, S. H. (2008). Mental toughness, optimism, pessimism, and coping among athletes. Personality and Individual Differences, 44(5), 1182-1192. https://doi.org/10.1016/j.paid.2007.11.011.
19. Pache, A., & Santos, F. (2010). When worlds collide: The internal dynamics of organizational responses to conflicting institutional demands.” Academy of Management Review, 35(3), 455-476. https://doi.org/10.5465/amr.35.3.zok455.
20. Paskus, T. (2015). NCAA division I academic progress rate, 2003-2014. ICPSR25801-v4. Retrieved from http://doi.org/10.3886/ICPSR26801.v4.
21. Pope, D. G., & Pope, J. C. (2009). The impact of college sports success on the quantity and quality of student applications. Southern Economic Journal, 75(3), 750-780. https://www.jstor.org/stable/27751414.
22. Pratt, M. G., & Foreman, P. O. (2000). Classifying managerial responses to multiple organizational identities. Academy of Management Review, 25(1), 18-42. https://doi.org/10.5465/amr.2000.2791601.
23. Pringle, R. G., & Hickey, C. (2010). Negotiating masculinities via the moral problematization of sport. Sociology of Sport Journal, 27(2), 115-138. https://doi.org/10.1123/ssj.27.2.115.
24. Rishe, P. J. (2003). A re-examination of how athlete success impacts graduation rates: Comparing student athletes to all other undergraduates. American Journal of Economics and Sociology, 62(2), 407-427. https://doi.org/10.1111/1536-7150.00219.
25. Sauder, M., & Espeland, W. N. (2009). The discipline of rankings: Tight coupling and organizational change. American Sociological Review, 74, 63-82. https://doi.org/10.1177/000312240907400104.
26. Scott, W. R. (2008). Institutions and Organizations: Ideas and Interests. Thousand Oaks, CA: Sage.
27. Settles, I. H., Sellers, R. M. & Damas, A. (2002). One role or two? The function of psychological separation in role conflict. Journal of Applied Psychology, 87(3), 574-582. http://dx.doi.org/10.1037/0021-9010.87.3.574.
28. Smith, D. R. (2015). It pays to bend the rules: Consequences of NCAA athletic sanctions. Sociological Perspectives, 58(1), 97-119. https://doi.org/10.1177/0731121414556844.
29. Snyder, M. (2014). Jim Harbaugh’s passion for academics could help michigan.” Detroit Free Press. Retrieved from https://www.freep.com/story/sports/college/university-michigan/wolverines/2014/12/19/jim-harbaugh-michigan-coach-speculation/20623661/.
30. Snyder, M. (2017). Michigan only FBS school to record 990+ APR in football and 995+ APR in men’s basketball. Detroit Free Press. Retrieved from https://www.freep.com/story/sports/college/university-michigan/wolverines/2017/05/10/michigan-football-apr-michigan-basketball/316264001/.
31. Southall, R. M., & Staurowsky, E. J. (2013). Cheering on the collegiate model: Creating, disseminating, and imbedding the NCAA’s redefinition of amateurism. Journal of Sport and Social Issues, 37(4), 403-429. https://doi.org/10.1177/0193723513498606.
32. Umbach, P. D., Palmer, M. M., Kuh, G. D. & Hannah, S. J. (2006). Intercollegiate athletes and effective educational practices: Winning combination or losing effort? Research in Higher Education, 47(6), 709-733. https://doi.org/10.1007/s11162-006-9012-9.
33. Washington, M., & Ventresca, M. J. (2004). How organizations change: The role of institutional support mechanisms in the incorporation of higher education visibility strategies, 1874-1995. Organization Science, 15(1), 82-97. https://doi.org/10.1287/orsc.1030.0057.