The Downstream Consequences of Race-Related Managerial Job Insecurity: Insights From College Basketball Coaching

Abstract

We examine how in men’s college basketball coaching, race-related managerial job insecurity trickles down to negatively affect the careers of the subordinates who work for them. Using panel data from a randomly selected group of assistant basketball coaches working under the most prestigious and endowed governing body of collegiate sports in the United States—the National Collegiate Athletic Association (NCAA) Division I (DI)—we find that, in men’s college basketball coaching, subordinate White coaches are less likely to be involuntarily dismissed than their non-White, predominantly Black, counterparts because non-White subordinates disproportionately work for racially minoritized, predominantly Black, head coaches who themselves face greater job insecurity. We also find involuntary dismissal correlates with whether assistant coaches leave the ranks of NCAA DI men’s college basketball coaching and explains the significant interaction between race and a teams’ performance relative to their respective conferences. These findings illustrate how race-related managerial job insecurity trickles down to negatively affect the job opportunities of their subordinates and, because of homophily, perpetuates racial disadvantage.

Keywords

job insecurity racial disadvantage college basketball coaching career dependence involuntary dismissals

Introduction

Coaching staffs for collegiate and professional sports teams operate like top management teams (TMTs) (Wolfe et al., 2005), with head coaches relying on the contributions of the assistant coaches they appoint to improve performance and satisfy expectations of various stakeholders. Recognizing parallels between coaching staffs and TMTs, researchers have used the case of college coaching for insights about careers in high-status labor markets (Day, 2015). Some of this research documents differences in the careers of White and racially minoritized coaches—differences that usually disadvantage minoritized head coaches (e.g., Cook and Glass, 2013; Cunningham, 2021; Salaga and Juravich, 2020; Turick and Bopp, 2016).

Related research seeks to explain how these racial inequalities persist despite concerted efforts to overcome them. Studies on the effects of the National Football League’s (NFL’s) Rooney Rule prove illustrative. The rule mandates teams interview at least one racially minoritized coach when hiring a head coach, allowing for the ceremonious interviewing of minority candidates with no intention to hire (Foreman and Turick, 2021; Gomer and Ossei-Owusu, 2022). Consequently, scholars question whether the rule has realized its intended goal of substantially increasing job prospects, stability, and retention for coaches of color (Foreman and Turick, 2021; Gomer and Ossei-Owusu, 2022).

At the National Collegiate Athletic Association (NCAA) level, researchers similarly question whether a lack of diversity among athletic directors allows racial bias to influence decisions regarding coaching outcomes. From this perspective, college athletics suffers from a history of institutional racism shielded by the edificial myth of “color-blindness” (Carrington, 2013; Hartman, 2000). Color-blind ideologies about inclusion, equal opportunity, and meritocracy (Rankin-Wright, Hylton, and Norman, 2016), when coupled with organizational practices that reinforce these ideologies (Ray 2019), in turn, mask real racial inequalities in college head coaching (Gomer and Ossei-Owusu, 2022).

Subsequent research has considered what race-based disadvantages for head coaches means for assistant coaches (Savage and Seebruck, 2016; Seebruck and Savage 2020). That research, drawing on organizational theories of career interdependence (Baek, Bidwell, and Keller, 2022; Ballinger, Lehman, and Schoorman, 2010; Jackall, 1988; Kanter, 1977; Li, Huisi, and Dragoni, 2020; Shapiro et al., 2016), examined the interconnectedness of the careers of head and assistant coaches by studying what happens to assistant coaches following a head coach’s departure. It finds assistant coaches are more likely to be retained if the outgoing head coach is White (Savage and Seebruck, 2016) and, if the assistant coach is also White, more likely to be promoted as the outgoing head coach’s replacement (Seebruck and Savage, 2020). These findings are consistent with claims that racially minoritized assistant coaches benefit more from the social capital that comes from having White supervisors than White assistant coaches do from having non-White supervisors (Cunningham, 2021).

Nevertheless, emphasizing only situations where the head coach experiences a job transition offers an incomplete and potentially errant understanding about the racialized effects of managerial mobility by ignoring cases of head coach stability. By only investigating what happens to assistant coaches after a head coach’s departure, this work ignores real differences in the likelihood of head coach departure while remaining silent on racial differences in job security for those assistant coaches who work for head coaches who keep their jobs. Put simply, previous research tackles only half the problem. It ignores situations where head coaches retain their jobs and, as a result, offers only a partial lens into how race and the interconnectedness of managers’ and subordinates’ careers affects the job security of White and non-White subordinates. We correct this omission in this paper by examining whether racially minoritized, predominantly Black, subordinates are more likely to experience involuntary dismissals because they work for racially minoritized, predominantly Black, head coaches, who themselves face greater job precarity.¹

Our argument is straightforward: racially minoritized assistant coaches will disproportionately experience greater job insecurity because they disproportionately work for racially minoritized head coaches who themselves face greater job uncertainty in cases of poor team performance. Importantly, these race-based differences in job security should hold regardless of whether head coaches retain their jobs. Using panel data from a randomly selected group of assistant basketball coaches working at the most prestigious and endowed governing body of collegiate sports in the United States—the NCAA Division I (DI)—we test this argument about how race and the interconnectedness of managerial and subordinate mobility affect the odds of an assistant coach being involuntarily dismissed. Our findings support our expectations: White assistant coaches are less likely to be fired because they are less likely to work for racially minoritized head coaches, who themselves face greater performance pressures. We further find involuntary dismissals are a primary predictor of whether an assistant coach exits the labor market and a key mechanism accounting for race moderating the effects of a team’s performance relative to other teams in their conference on exiting. These differences reflect the downstream negative consequences of the glass cliff or race-related managerial job precarity (Morgenroth et al., 2020)—consequences that, because of the tendency toward racially homogeneous coach staffs, perpetuate racial disparities in this profession.

While these findings can only truly generalize to NCAA DI college basketball, it is likely the theoretical processes uncovered apply somewhat broadly. Basketball in the United States has been racialized as a Black cultural space. The mainstream marketing of basketball and its products call on the racialized imagery of playground basketball in urban areas (Oates, 2017), and processes associated with collective identity result in Black youth expressing a preference for the sport (Ogden and Hilt, 2003). As a Black cultural space, there are significant internal and external pressures to promote equal opportunities for racial minorities in basketball. Consequently, NCAA DI college basketball coaching is more racially integrated than other college sports (Lapchick et al., 2013). Nevertheless, although racially minoritized individuals compose almost 50% of assistant coaches at the DI level, they constitute only about 29% of head coaches, with these numbers decreasing to 47.6% and 24.1%, respectively, after excluding coaches at historically Black colleges and universities (HBCUs) (Meyer, 2019).² Thus, explaining how the glass cliff accounts for the blockage in the assistant-to-head coach pipeline for minority coaches in this labor market may offer new theoretical insights applicable to other coaching labor markets specifically and other managerial labor markets generally. Given these possibilities, we begin by reviewing the broader literature on job insecurity and career independence before turning to address how insights from this more general literature might inform processes in the coaching labor market that we study.

Theoretical Background

Job Insecurity in the United States

Job insecurity concerns the relative stability of one’s employment (Greenhalgh and Rosenblatt, 1984). When work becomes unstable it becomes insecure and there is mounting evidence work in the United States is more insecure today than it was in the 1950s and 60s. Compared to then, job tenure is shorter and long-term unemployment is longer (Kalleberg, 2009). Kalleberg (2009) attributes these changes to broad structural and ideological shifts, including globalization, deregulation, technological advances, and neoliberalism. For workers, these macro-level changes have made markets more competitive and involuntary job loss more likely (Kalev, 2014; Valletta, 1999). This reality is not lost on workers, who perceive their jobs to be more insecure, especially if they are male and not White (Lowe, 2018).

Male job insecurity is particularly an issue for Black and Latino men. Members of these groups face persistent disadvantages at work (e.g., Bielby, 2012; Quillian et al., 2017; Ray 2019; Smith, 1999; Wilson, 2009). Social closure and racial bias can block minorities from good jobs (Day, 2018; Wilson and McBrier, 2005), and these effects are compounded by deficits in social capital (Pedulla and Pager, 2019), which, when paired with racial discrimination, track minorities into lower-paying, more precarious minority-dominated jobs (Huffman and Cohen, 2004).

Racialized job tracking and reliance on subjective performance evaluations, in turn, increase the chances of minorities experiencing involuntary exits (Collins, 1997; Wilson and Roscigno, 2010) and inhibits their ability to proactively manage careers (Ray 2019). The minority vulnerability thesis captures this process, describing how norms about workforce composition and cognitive biases lead employers to perceive racial minorities more negatively (Greenhaus and Parasuraman, 1993) and subsequently dismiss them more readily (Wilson, 2009; Wilson and McBrier, 2005; Wilson and Roscigno, 2010). This occurs even in situations where there is little uncertainty about performance (Norris and Moss-Pech, 2022). Simply put, greater perceived job insecurity among racial minorities is consistent with the fact that they disproportionately experience involuntary exits in racialized organizations (Ray, 2019), which over time undercuts their ability to amass the human capital needed to secure better positions and wages (Tomaskovic-Devey, Thomas, and Johnson, 2005) and stable careers (Wilson and Roscigno, 2010).

While the impact of this disadvantage for the individual is clear—that is, it is harder for racially minoritized individuals to advance up the corporate ladder and to keep supervisorial positions when they do—less obvious is what this means for subordinates of those racially minoritized individuals who do secure leadership positions. The glass cliff metaphor captures the idea that minoritized individuals are more likely to secure precarious leadership positions and are more likely to be dismissed as a result (Morgenroth et al., 2020). Some evidence exists in support of a racialized glass cliff in NCAA DI college basketball coaching, for example (Cook and Glass, 2013). Such racialized differences in the job precarity of managers may provide one mechanism by which racial inequalities at work persist—for example, if racially minoritized managers disproportionately provide job opportunities to racially minoritized subordinates and if the job uncertainty of the managers trickles down to affect their subordinates.

And indeed, research suggests the careers of managers and subordinates are frequently intertwined. Ballinger and colleagues (2010), for example, show employees who have good relationships with managers are more likely to leave organizations after the managers depart than employees who have bad relationships with their managers. Building on this observation that the quality of the superordinate-subordinate relationship affects how employees respond to managerial departure, Baek and Colleagues (2021) investigate how manager mobility affects subordinates’ performance, pay, and mobility inside a large Fortune 500 healthcare company. Using 8 years of personnel data, they find managerial mobility increases both voluntary and involuntary employee turnover, especially when the manager exited the organization.

A positive association between managerial and subordinate turnover seems particularly likely in college basketball coaching considering extant research. In their studies of NCAA Division I men’s college basketball coaching, Savage and Seebruck (2016); Seebruck and Savage (2020) find evidence consistent with the idea that managerial exits precipitated by a desire for organizational change are more likely to result in the dismissal of the other team members (Barron, Chulkov, and Waddell, 2011). Looking only at cases involving a head coach’s departure, they find, compared to when a head coach leaves for a new head coaching opportunity, the odds of assistant coaches being fired increase when the head coach is fired. Moreover, evidence also shows this career interdependence varies along racial lines, with assistant coaches most likely to be let go when the outgoing head coach is a racial minority and when the incoming head coach is of a different race than the assistant. This is more problematic given consistent evidence of racial homophily in this labor market (Cunningham and Sagas, 2005; Savage and Seebruck, 2016).

These findings, however, only address situations where the head coach experiences a job transition. Failing to account for racial differences in job insecurity regardless of head coach stability limits insight into how racial differences in managerial job insecurity trickle down to affect the careers of subordinates, and thus, offers an incomplete understanding of how the glass cliff affects the obstructed pipeline in college basketball coaching.

Race-related Managerial Job Insecurity and Subordinates’ Careers in College Basketball Coaching

Research finds White workers report faster rates of internal promotion (James, 2000) and increased job security compared to their Black counterparts (Wilson and McBrier, 2005). These findings hold even in contexts with clear performance indicators, like sports (Norris and Moss-Pech, 2022). In fact, studies show performance indicators often mask very real racial disparities in coaching. Because racially minoritized coaches disproportionately secure head coaching positions for teams in disarray (Cook and Glass, 2013), they face a glass cliff that makes it difficult for them to achieve the same level of success. This, in turn, fosters prejudices about the leadership abilities of minority coaches (Turick and Bopp, 2016). Indeed, although people view Black individuals in college basketball as physically superior, they downplay their general leadership skills and intelligence levels (Buffington and Fraley, 2008) in a manner that devalues their effort and mental acumen relative to White individuals (Bonilla-Silva and Dietrich, 2011). In NCAA Division I men’s college basketball coaching, these processes result in a glass cliff, with racially minoritized coaches more likely to be hired as head coaches at schools with losing records, more likely to have shorter tenures as head coaches, and more likely to be replaced by someone who is White (Cook and Glass, 2013).

These disadvantages highlight the greater job insecurity racially minoritized head basketball coaches face which likely has negative consequences for racially minoritized assistant coaches too. Consistent with theorizing about social closure and opportunity hoarding (Tomaskovic-Devey, 2014), research on college basketball coaching finds head coaches are more likely to hire assistant coaches of the same race (Cunningham and Sagas, 2005), resulting in non-White, usually Black, assistant coaches disproportionately working for non-White, usually Black, head coaches and White assistant coaches disproportionately working for White head coaches. Racial homophily in college basketball coaching means minority assistant coaches are likely to work for those who, because of the glass cliff, face the greatest job insecurity.

Thus, we argue that racial differences in the firing of assistant coaches can be explained by the fact they disproportionately work for racially minoritized head coaches, who themselves are more likely to suffer from both perceived and real job insecurity. A series of interviews conducted by Jesse Washington (2021) with fourteen Black college basketball coaches captures this insecurity. Kevin Sutton, an assistant coach at Florida Gulf Coast, expressed this feeling of insecurity thusly, “We have to do it cleaner, and we have to do it better, and we can’t have blemishes…And clearly, we have to win. At some schools you have to win big. We have to be more professional, smarter and more consistent. We have to be twice as good.” Other coaches offered similar assessments, with one even asking, “Does a white man really go into a $500,000 coaching job with as much pressure as a Black female or Black man going into that position? I don’t think so.” These quotes provide qualitative support for the insecurity and double standards racially minoritized coaches face.

This greater job insecurity matters because, as Shoss (2017) notes, insecurity activates a job preservation motive that encourages people to behave in ways they think will help them keep their job. For racially minoritized head coaches, this should result in a more demanding managerial style and greater willingness to dismiss assistant coaches if they view the team as underperforming. Such views are likely to be made in relative terms. Although NCAA DI men’s college basketball teams can be ranked across the field based on performance, such assessments are likely made in a more localized manner, as social comparisons are less likely as differences in ability grow (Festinger, 1954). For example, teams in this field participate in athletic conferences that generally comprise similarly positioned colleges and universities. As a team underperforms relative to this neighborhood, performance pressures should increase, especially for Black head coaches, who are more likely to get fired. This, when coupled with the fact that racially minoritized assistant coaches are more likely to work for non-White NCAA DI head basketball coaches who have shorter job tenures (Cook and Glass, 2013), suggests that non-White assistants should be more likely to be fired because they work for non-White head coaches who will be less tolerant of poor performance and more susceptible to being fired themselves for poor performances. We therefore hypothesize:

Hypothesis 1

Compared to their non-White counterparts, White assistant coaches will be less likely to be fired.

Hypothesis 2

White assistant coaches being less likely to be fired will be explained by the fact that non-White assistant coaches will disproportionately work for non-White head coaches.

Hypothesis 3

Assistant coaches will be more likely to be fired when they work for a non-White head coach whose team performs worse relative to the average performance of the teams in their athletic conference.

Hypothesis 4

The interaction effect of head coach race and performance (relative to teams in one’s athletic conference) on assistant coach’s dismissal will be partially mediated by the head coach’s dismissal.

Finally, involuntary dismissals are a stain on one’s resume, fairly or unfairly signaling poor performance and thereby making the subsequent unemployment period particularly problematic. As Eriksson and Rooth (2014) and Pedulla (2016) observe, unemployment can negatively impact the likelihood of men receiving callbacks for interviews. Thus, we expect that being fired should increase the likelihood of an assistant coach exiting the NCAA Division I men’s basketball coaching ranks.

Hypothesis 5

Involuntary dismissals will increase the likelihood of a coach exiting the ranks of NCAA Division I men’s basketball.

Data and Methods

To test our hypotheses, we rely on panel data from randomly selected assistant basketball coaches working as either assistant or associate head coaches at the men’s NCAA Division I (DI) level. Using a sampling frame (N = 1049) of all assistant coaches (either at the assistant or associate head coach levels) actively employed at any of the 351 DI schools in the fall of 2015, a random sample of 380 assistant or associate head coaches, consisting of 36.2 percent of the population, was generated.³ All coaches were male as there were no female coaches at the time. Following that, graduate and undergraduate research assistants, under the direction of the first author, collected demographic and career history data for each assistant coach using publicly available secondary sources: team media guides and websites, newspapers, biographies, and other publicly available sources (e.g., LinkedIn.com and Hoopdirt.com).

We then followed the careers of these coaches annually for 5 years. Of the 380 assistant coaches in the sample, 309 (81.3%) experienced at least one job transition. Because coaching changes are publicly available—verifiable via news reports, media guides, and social media updates—we were able to track the career transitions for each assistant coach while also collecting information on team performance and the head coaches for whom they worked. This resulted in a dataset of 1900 case years (5 years nested in 380 assistant coaches), with 150 of the 380 assistant coaches (39.5%) in the sample being fired from a job and 154 of them (40.5%) leaving the NCAA DI men’s basketball coaching ranks at some point in the window. Eight of the 150 individuals were fired multiple times and 1 left the DI ranks, rejoined, and then left again. Restricting the data to just those years where the selected assistant coach maintained a position on the staff of an NCAA DI men’s head basketball coach yielded 1459 case years nested in 380 coaches. This number was further reduced to 1444 case years nested in 377 coaches after excluding those working for U.S. military academies which do not report expenditure data used to create a key control variable for our analyses.⁴ This constitutes our analytic sample.

Dependent Variables

We have two dependent variables. The first is a binary measure of whether the individual was involuntarily dismissed from their position during or at the conclusion of each season. We determined this by referencing online newspapers and websites that document coaching moves. A coach was coded as fired if there was mention of being let go or dismissed. Values of 1 indicate the individual was involuntarily dismissed from the position that year and values of 0 indicate otherwise. This dependent variable becomes an independent variable for our second set of analyses: assessing whether an assistant coach exits the NCAA Division I men’s basketball coaching ranks at the conclusion of each season. Such an exit constitutes our second dependent variable, which is also a binary measure with values of 1 indicating the assistant coach left the DI men’s basketball coaching ranks and a value of 0 indicating otherwise.

Independent Variables

We coded the race of each assistant coach and each head coach they worked for dichotomously as either White or non-White. Black coaches make up most of the non-White group, with only one Latino, four Asians, and one mixed race assistant coach in the sample. For head coaches in our sample, all but three are either White or Black. Thus, our analyses and conclusions are largely limited to White-Black comparison, as the lack of variation in our sample prevents us from speaking to homophily processes across other racial pairings, which may differ from a predominantly White-Black dichotomy (Giuliano, Levine, and Leonard, 2009). Nevertheless, we maintain the use of non-White as this more precisely captures the variation in the data and our statistical comparisons.

To assess race, we used photographs of coaches in team media guides or official team websites using a standard phenotypic coding technique, whereby coders used skin color and physical features to assess race (see Murguia and Telles, 1996). In addition, we used online searches to seek out any additional information on a coach’s race to help confirm these assessments. The coding of race occurred as follows. Initially, the first author, who is White, and an undergraduate research assistant, who is Hispanic, used photographs and online information to make racial classifications. Only in one case were we unable to make a clear assessment (i.e., intercoder reliability = 99.7%). We labeled this coach as mixed race. After the initial assessment, a different undergraduate student, who is Asian, and a graduate student, who is White, independently verified the initial assessments as they collected data for later years. Neither identified any coding as problematic. Admittedly, this technique ignores self-assessments of racial and ethnic identity as well as issues of colorism (Foy and Ray, 2019). Nevertheless, research shows that because skin color and phenotype influence perceptions of others (Murguia and Telles, 1996), it is a valid and reliable method for our purposes.

We also examine performance using three measures derived from the work of sports statistician, Jeff Sagarin (2020), which has been used by the NCAA tournament when selecting teams for over 30 years. Our first measure relies on Jeff Sagarin’s yearly performance score for each DI basketball program. This score is calculated using a team’s win-loss record, strength of schedule, and margin of victory (West, 2008) and provides a field-level yearly ranking of teams. Next, because basketball teams are in different athletic conferences, which vary in terms of resources and expectations, our second measure offers a more localized assessment of performance by subtracting the team’s Sagarin rating for that year from a similar Sagarin measure for each athletic conference. We suspect this measure, which assesses how each team fared in its neighborhood, matters more than the first given it is a more proximal comparison to teams with similar resources, contexts, and expectations. And finally, we calculated each team’s performance relative to the team’s average performance in the previous 2 years by subtracting the Sagarin performance rating for that year from the average of the Sagarin ratings for the team in the prior 2 years.

Next, we account for the job security of the head coach by assessing whether the head coach was involuntarily dismissed, did not have his contract renewed, or was forced to resign because of poor performance. This was determined using news reports and was coded dichotomously (1 when true and 0 otherwise). Finally, we control for other relevant characteristics of the assistant coach: job transitions, job tenure, former player status, prior affiliation as a student with the organization, athletic director race, athletic direct sex, athletic director tenure, and team expenditures relative to the expenditures of conference foes. Although coaches were randomly selected into the dataset based on their positions as either assistant or associate coaches at the start of the 2015−2016 season, some coaches transitioned to non-coaching jobs on team staffs (e.g., directors of basketball operation or video coordinators). Thus, we include a dummy variable demarcating non-coaching positions on staff (1) from coaching roles (0). Research also suggests job tenure might affect dismissal. To control for the curvilinear nature of job tenure on dismissal in this profession (Wangrow, Schwartz, and Hughes-Moran, 2021), we include a variable measuring the number of consecutive years the person has been at the school as well as its squared term, with these terms being updated yearly as needed. For the dismissal analyses, we include similar measures for the head coach’s tenure. We also control for whether the assistant coach was formerly an NCAA DI basketball player with a binary measure valued 1 if this was the case as well as for whether the assistant coach was working at his alma mater, since research shows these factors affect job mobility in this labor market (Savage and Seebruck, 2016). A similar procedure using online information, names, and photographs was employed to assess the race and sex of athletic directors, with race measured White (1) or not (0) and sex measured male (1) or female (0). Athletic director tenure is a continuous measure assessed using news reports documenting the athletic director’s hiring.

Using data from the U.S. Department of Education’s Equity in Athletics Data Analysis website, we consider relative team expenditures by controlling for the difference between a team’s expenditures and the average expenditures for all of the other teams in each team’s respective athletic conference.⁵ Given the large variation in expenditures across teams and conferences, we standardized this measure. Values ranged from −3.13 to 6.79, with more negative values indicating lower team expenditures relative to the conference and more positive values indicating greater team expenditures relative to the conference.

Our data do not include any female coaches since there were no female NCAA D1 men’s basketball coaches at the time the sampling frame was generated, but they do account for time with a series of dummy variables identifying each year of the 5-year window. Table 1 presents the descriptive statistics.

Table 1.

Descriptive Statistics of Variables (N = 1444).

	Mean or %	SD	Min	Max
Dep. variables
Fired	10.09%		0	1
Left DI ranks	10.06%		0	1
Ind. variables
2015−2016 season	26.11%		0	1
2016−2017 season	22.22%		0	1
2017−2018 season	19.39%		0	1
2018−2019 season	17.11%		0	1
2019−2020 season	15.17%		0	1
AC race (White = 1)	49.79%		0	1
Staff position (yes = 1)	5.19%		0	1
DI college player	56.99%		0	1
Alma mater	12.88%		0	1
Tenure at school	4.43	3.75	1	29
Public school	64.75%		0	1
AD non-White	17.11%		0	1
AD Sex	91.07%		0	1
AD tenure	7.11	6.51	1	37
HC tenure (HCT)	6.53	5.78	1	44
Expenditures relative to conference average (standardized) (ECA)	0.04	1.10	−3.13	6.79
Performance	72.92	9.20	48.84	96.13
Performance relative to conference	−0.74	5.63	−21.7	16.32
Performance relative to past years	−0.004	4.56	−16.78	16.25
HC non-White	27.77%		0	1
HC fired	8.50%		0	1

Analytic Strategy

Given the panel nature of the data, with years nested in assistant coaches, and our research question about race differences in the likelihood of dismissal, we employ a series of random-effects probit regression models to account for variation in our two dependent variables.^6,⁷ We use the KHB framework to test for mediation where appropriate as it was specifically developed to test for mediation in nested nonlinear models such as these (Breen, Karlson, and Holm, 2013).

We also run discrete-time survival analyses using complementary log-log models. We model time as discrete rather than continuous given the limited number of years for which we have data and given the nature of the dependent variable which tracks dismissal in a coarse-grained manner (Rabe-Hesketh and Skrondal, 2012). These models shift the focus from the general odds of dismissal to differences in the discrete-time hazard of dismissal, while not making any assumptions about the shape of the baseline survivor function. Results are in the appendix and corroborate that White assistant coaches survive longer because they disproportionately work for White head coaches who themselves experience less job precarity.⁸

Results

Our expectation is that racial differences in the dismissal of assistant coaches is a byproduct of minority assistant coaches disproportionately working for minority head coaches, who themselves face greater job insecurity. Moreover, we argue that involuntary dismissal should correlate with these coaches exiting the NCAA DI college basketball coaching ranks. Assumed in this argument is that racial homophily funnels assistant coaches onto the coaching staffs of same-raced head coaches, as extant research suggests (Cunningham and Sagas, 2005). To examine this process with these data we ran a random-effects probit model regressing the race of an assistant coach on the race of the head coach. The model assumes head coaches select their staffs, which is the case in this labor market. Results show a clear positive relationship, such that White head coaches disproportionately select White assistant coaches and non-White head coaches disproportionately select non-White assistant coaches for their respective staffs (b = 0.574, p = .000). Descriptive statistics largely bear this out. Looking at the first year of the data, we see that 55% of the 273 assistant coaches working for White head coaches are White and that 39% of the 107 assistant coaches working for non-White head coaches are White. These numbers suggest non-White assistant coaches disproportionately work for non-White head coaches—a statistically significant difference (p = .005). With this established, we test our hypotheses.

Involuntary Dismissals

Models 1 through 5 in Table 2 provide the statistical bases for testing Hypotheses 1 through 4. Model 1 offers clear support for the first hypothesis: White assistant coaches are less likely to be fired compared to their non-White counterparts, controlling for season, staff position, history as a DI college basketball player, working at one’s alma mater, consecutive years working at the school, whether the school is public, relative expenditures, tenure of the head coach, and the race, sex, and tenure of the athletic director. Looking at the average marginal effects tells us the predicted probability of being involuntarily dismissed decreases by approximately four-percentage points (dy/dx =−.04, p = .01) for White assistant coaches. A four-percentage swing in the predicted probability of being dismissed is substantively meaningful given, as we show below, the strong relationship between dismissal and exiting the ranks of NCAA Division I basketball.

Table 2.

Random-Effects Probit Regression Models Predicting Dismissal.

	Model 1	Model 2	Model 3	Model 4	Model 5
2016−2017 season	−0.11 (0.13)	−0.13 (0.13)	−0.15 (0.14)	−0.17 (0.14)	0.16 (0.17)
2017−2018 season	−.19 (0.13)	−0.18 (0.14)	−0.15 (0.15)	−0.18 (0.15)	−0.19 (0.19)
2018−2019 season	−.03 (0.14)	−0.03 (0.14)	0.04 (0.15)	0.01 (0.15)	0.24 (0.19)
2019−2020 season	−0.76^*** (0.19)	−0.78^*** (0.20)	−0.77^*** (0.21)	−0.80^*** (0.21)	−0.65^* (0.27)
AC White	−0.25^* (0.10)	−0.20^* (0.10)	−0.17⁺ (0.11)	−0.20⁺ (0.11)	−0.18 (0.13)
Staff position	−0.05 (0.25)	−0.06 (0.25)	−0.02 (0.26)	0.02 (0.26)	0.12 (0.30)
DI basketball player	−0.08 (0.10)	−0.06 (0.10)	−0.06 (0.10)	−0.06 (0.11)	−0.13 (0.13)
At alma mater	−0.74^*** (0.21)	−0.76^*** (0.22)	−0.87^*** (0.24)	−0.93^*** (0.24)	−1.04^** (0.35)
Tenure at school (T)	0.07 (.05)	0.07 (0.05)	0.08 (0.05)	0.08 (0.05)	0.03 (0.06)
T^2	−0.01⁺ (.003)	−0.01⁺ (0.004)	−0.01⁺ (0.004)	−0.01⁺ (0.004)	−0.005 (0.005)
Public school	−0.01 (.10)	−0.04 (0.10)	0.06 (0.11)	0.05 (0.11)	0.07 (0.14)
AD non-White	−0.01 (.12)	−0.22 (0.13)	−0.22⁺ (0.14)	−0.31 (0.15)	−0.34⁺ (0.19)
AD sex	−0.42^** (.14)	−0.42^** (0.15)	−0.31⁺ (0.15)	−0.31^* (0.15)	−0.38⁺ (0.19)
AD tenure (ADT)	−0.001 (.01)	−0.004 (0.01)	−0.002 (0.01)	−0.001 (0.01)	0.001 (0.01)
HC tenure (HCT)	0.08^* (.04)	0.09^* (0.04)	0.13^** (0.04)	0.13^** (0.04)	0.04 (0.03)
HCT^2	−0.005^* (.002)	−0.005^* (0.002)	−0.01^** (0.002)	−0.01^* (0.002)	−0.000 (0.00)
Expenditures relative to conference average (standardized) (ECA)	0.07 (.05)	0.08⁺ (0.05)	0.20^*** (0.05)	0.18^*** (0.06)	0.04 (0.07)
HC non-White		0.51^*** (0.11)	0.40^*** (0.11)	0.27^* (0.13)	0.33^* (0.15)
Performance			−0.01 (0.01)	−0.01 (0.01)	−0.0003 (0.01)
Performance relative to conference average (PCA)			0.08^*** (0.01)	0.06^*** (0.02)	0.01 (0.02)
Performance relative to past			0.02 (0.01)	0.02 (0.01)	0.01 (0.02)
ECA × PCA				0.03^* (.01)	0.03^* (0.01)
HC non-White × PCA				0.07^** (.02)	0.05⁺ (0.03)
HC fired					2.60^*** (0.25)
Constant	−0.87^*** (0.24)	−1.03^*** (0.24)	−1.02⁺ (0.58)	−1.03⁺ (0.59)	−1.46⁺ (0.75)

Notes: ⁺p < .10, ^*p < .05, ^**p < .01, ^***p < .001 (two-tailed). N = 1444 years nested in 377 people.

Model 2 tests whether working for a non-White head coach explains this apparent difference in job insecurity for White and non-White assistant coaches. When including the race of the head coach in the model, the magnitude of the coefficient for assistant coach’s race decreases. Moreover, consideration of the average marginal effect reveals a 9.7-percent increase in the likelihood of an assistant coach being fired if they work for a minority head coach. Applying the KHB method confirms mediation and supports Hypothesis 2, with 22.95 percent of the effect of the assistant coach’s race being accounted for by the head coach’s race (C.I. [–0.092, −0.026], p < .001). The significant coefficient of the race of one’s head coach on the dismissal of the assistant coach holds after controlling for performance and relative expenditures. Per Model 3, assistant coaches were more likely to be fired when their teams performed worse than the conference average (p < .001) and when team expenditures exceeded the average expenditures in the conference (p < .001), highlighting how localized comparisons of performance and expenditures inform dismissal decisions. Yearly performance and change in performance relative to the previous 2 years were not significant at the .05 level (p = .37 and p = .18, respectively), and we note that our results hold regardless of whether these variables are included in the model, obviating concerns about collinearity. Nevertheless, with the average marginal effect revealing a 6.8-percent increase in assistant coaches being dismissed when they work for a minority head coach and working for a minority head coach mediating 18.46 percent of the effect of assistant coach’s race, even after controlling for performance, the racial disadvantage of head coaches trickles down to affect the job security of assistants.

Models 4 and 5 scrutinize this effect. Model 4 reveals a significant interaction effect between a head coach’s race and a team’s performance relative to that of their athletic conference’s average performance as well as a significant interaction between relative performance and relative expenditures. Tests for interaction effects with the other performance measures did not yield significant effects, highlighting the fact that in this labor market the effects of performance are assessed in relation to immediate peers. Before considering the race-by-performance interaction, we note the interaction of performance and expenditures indicates that dismissal becomes more likely as performance lags but as expenditures increase relative to conference averages.

We now turn to our primary focus. Figure 1 explores the race-by-performance interaction by plotting the predictive average marginal effects of involuntary dismissal for those working for White and non-White head coaches at various levels of relative performance. The figure shows that when teams outperform the conference average, the probability of an assistant coach being dismissed is small and unaffected by the race of the head coach. However, when teams underperform relative to the other teams in the conference, the probability of dismissal increases as the team underachieves, with this increase being larger for those working for minority head coaches. This supports Hypothesis 3 and is consistent with the idea that the greater job insecurity and performance pressures experienced by minority head coaches negatively affects the job security of their assistants.

Figure 1.

Predictive margins of involuntary dismissal.

Model 5 strengthens confidence in this interpretation as the likelihood of an assistant coach being fired increases when the head coach is fired, and including this variable reduces the race-by-performance interaction term’s coefficient to non-significance (p < .06), although the main effect of head coach’s race maintains its significant effect on dismissal (p < .03). This finding is in line with the qualitative evidence cited previously about how Black coaches are held to a higher standard, “hav[ing] to be twice as good” (Washington, 2021). Moreover, applying the KHB method reveals that a head coach being fired mediates approximately 40.9 percent of the effect of the interaction term (C.I. [0.014, 0.055], p = .001), highlighting how the precarity confronting minority head coaches bleeds over to affect the assistants who work for them.

Figure 2 revisits Figure 1 after accounting for a head coach’s firing. When the average conference performance is more than eight points higher than the team’s performance there is no longer a significant difference in the probability of involuntary dismissal for those working for White or minority head coaches, ostensibly because these head coaches are themselves dismissed due to their team’s poor performance. The racial difference in assistant coaches being dismissed, however, persists for middling teams, with assistant coaches experiencing a greater probability of being dismissed when they work for minority head coaches whose teams’ range in relative performance from 0 to 8, where a score of 0 indicates the team equaled the conference average and a score of 8 indicates the team underperformed relative to the conference average. When teams clearly outperform the other teams in the conference, assistant coaches are unlikely to be dismissed and this does not vary by the race of the head coach.

Figure 2.

Predictive margins of involuntary dismissal.

These results support Hypothesis 4 and the downstream consequences of race-related managerial job insecurity. Assistant coaches are more likely to be dismissed when they work for minority head coaches whose teams underperform because middling and underperforming minority head coaches face greater performance pressures and less job security. Compounding this racial disadvantage is the fact that minority head coaches disproportionately hire minority assistants, resulting in minority assistant coaches disproportionately being fired.

Moreover, switching the focus to the hazard of dismissal reveals similar results to those of the random-effects probit models. Results are in the Appendix, and we note that these results hold when we restrict analyses to just instances of first dismissal. Again, the message is clear: minority assistant coaches face a greater hazard of dismissal because they work for minority head coaches who themselves face greater job pressures when they underperform.

Exiting NCAA Division I Men’s College Basketball

Involuntary dismissal matters not only because it stains one’s employment record, but because it should affect whether assistant coaches exit the ranks of NCAA DI men’s college basketball coaching. The models in Table 3 examine this.

Table 3.

Random-Effects Probit Regression Models Predicting Exiting the NCAA DI Men’s College Basketball Coaching Ranks.

	Model 1	Model 2	Model 4
2016−2017 season	−0.13 (0.14)	−0.14 (0.14)	−0.07 (0.16)
2017−2018 season	−0.03 (0.14)	−0.05 (0.14)	0.09 (0.16)
2018−2019 season	0.08 (0.15)	0.08 (0.15)	0.10 (0.17)
2019−2020 season	−0.27 (0.17)	−0.27 (0.17)	0.09 (0.19)
AC White	0.10 (0.10)	0.16 (0.11)	0.25^* (0.12)
Staff position	0.28 (0.22)	0.31 (0.22)	0.44⁺ (0.24)
DI basketball player	−0.27^** (0.10)	−0.28^** (0.10)	−0.34^** (0.12)
At alma mater	−0.14 (0.17)	−0.15 (0.17)	0.24 (0.19)
Tenure at school (T)	0.07 (0.03)	0.07^* (0.03)	0.04 (0.04)
T^2	−0.003⁺ (.002)	−0.003⁺ (0.002)	−0.002 (0.002)
Public school	−0.07 (0.10)	−0.08 (0.10)	−0.12 (0.11)
HC non-White	0.15 (0.10)	0.17 (0.10)	−0.05 (0.13)
Expenditures relative to conference average (standardized)	0.08 (0.05)	0.08 (0.05)	−0.01 (0.06)
Performance	−0.03^*** (0.01)	−0.03^*** (0.01)	−0.03^*** (0.01)
Performance relative to conference average (PCA)	0.04^** (0.01)	0.07^*** (0.02)	0.02 (0.02)
Performance relative to past	0.02 (0.01)	0.02 (0.01)	0.003 (0.01)
AC White × PCA		−0.05^* (0.02)	−0.03 (0.02)
AC fired			1.98^*** (0.14)
Constant	0.51 (0.52)	0.40 (0.52)	0.64 (0.63)

Notes: ⁺p < .10, ^*p < .05, ^**p < .01, and ^***p < .001 (two-tailed). N = 1444 years nested in 377 people.

Although Model 1 shows no race differences in exits for the assistant coaches we study, Model 2 reveals that race interacts with team performance relative to the conference average to affect exits. The significant interaction effect reveals that the effects of relative performance on exiting the field are sensitive to race, with the effects of performance mattering most for non-White assistant coaches. Including involuntary dismissals in Model 3 erases this effect, and mediation analyses reveal that is because of differences in who is fired. Moreover, Model 3 reveals statistical suppression with the direct effect of assistant coach’s race becoming significant at the .05 level. The results indicate White and non-White assistant coaches exit under different circumstances. Whereas White coaches are more likely to leave these ranks voluntarily in the pursuit of other opportunities, non-White coaches frequently leave the ranks involuntarily after being fired. We suspect this difference is in part due to the greater freedom of Whites to navigate jobs. This point becomes starker because, per Hypothesis 5, we find involuntary dismissal increases the likelihood of an assistant coach exiting.

Discussion and Conclusion

We find that in NCAA DI men’s college basketball coaching, subordinate Whites are less likely to be involuntarily dismissed than their non-White counterparts because said counterparts disproportionately work for minority head coaches who themselves experience greater job insecurity. We also find involuntary dismissal correlates with whether assistant coaches leave the ranks of NCAA DI men’s college basketball coaching and some evidence of White and non-White assistant coaches leaving under different circumstances. Where it appears minority assistant coaches disproportionately exit because of dismissal, White assistant coaches frequently leave voluntarily to pursue new career opportunities. We interpret this as indicative of White assistant coaches having more occupational freedom than non-Whites as well as the negative consequences of the greater job insecurity for primarily Black assistant coaches in this profession.

These findings complement prior work on manager-employee career interdependence (e.g., Baek et al., 2022; Ballinger et al., 2010; Li et al., 2020; Shapiro et al., 2016) by explicating in college coaching how the interdependent careers of managers and subordinates perpetuate racial disadvantage. Where past research documents how homophily processes encourage head coaches to fill their coaching staffs with same-raced others (Cunningham and Sagas, 2005), we document here how this exposes minority subordinates to the greater job insecurities of their minority supervisors, thereby directing attention to the career interdependence of head and assistant coaches. Our results show one reason minority assistant coaches experience greater job instability is in large part because they disproportionately work for minority head coaches who themselves experience greater job insecurity when they underperform relative to their peers.

Prior research on the glass cliff finds that, among DI men’s college basketball head coaches, minorities have shorter tenures than White head coaches (Cook and Glass, 2013). The consequence of this for minority assistant coaches is unequivocal: they are more likely to be fired because they are more likely to work for those who themselves are fired. Even when these head coaches are not fired, we speculate that the greater job insecurity they experience when their teams underperform elicits a job perseverance motive that makes them more likely to fire subordinates they see as underperforming (Shoss, 2017). Our data are consistent with this idea. Thus, the disadvantages that inhibit the careers of minority head coaches disproportionately stain the employment records of minority assistant coaches. Tomaskovic-Devey and colleagues (2005) document how race differences in employment affect the accumulation of human capital over one’s career, which in turn accounts for significant race differences in earnings. Although we do not have salary data, our findings are consistent with this idea and offer one explanation for it: when the careers of minority assistant coaches are tethered to the less secure careers of minority head coaches, assistant coaches are more likely to experience dismissal, which thwarts their ability to amass the human and social capital necessary for career advancement.

Thus, tracking minority head coaches into positions that make it difficult for them to succeed does more than simply create prejudices about the leadership abilities of minority coaches (Turick and Bopp, 2016). It also, whether because of homophily or job tracking processes, negatively affects the ability of minority assistant coaches to create careers that signal head coaching ability. This likely perpetuates stereotypical beliefs about racial minorities in sports as physically superior but less capable of leading (Buffington and Fraley, 2008; Cunningham, 2021; Foreman and Turick, 2021; Gomer and Ossei-Owusu, 2022; Hartmann, 2000). Perhaps this is one reason why, despite approximately 50 percent of assistant coaches being racial minorities in 2019, only 29 percent of the head coaches were (Meyer, 2019).

The exploration of how aforementioned stereotypes, or racial schemas (Ray 2019), negatively impact Black coaches falls under a culturalist orientation view of the relationship between sport and racism—or the “symbolic role of sport in the formation of racial meanings” (Hartmann, 2000). While some racial schema such as the notion of Black Americans’ physical superiority might seem positive, they are, ultimately, stereotypes rooted in a culture of racism and disadvantage. This culturalist perspective falls outside the scope of this paper’s analysis, though, which takes a largely institutionalist approach in analyzing how racial inequality in coaching is perpetuated by the organization-level structures and practices in place.

That said, the two perspectives are not inherently at odds and may be synthesized (Hartmann, 2000). This synthesis may arise from a recognition that sport is a “contested racial terrain” (Harmann, 2000), or a venue where racial meanings and dynamics are grappled over, and a site for the reproduction and resistance of both hurtful and helpful notions about race, as agents at all levels vie for power. Future work may benefit from focusing on bridging the gap between the two perspectives while situating sport as a contested racial terrain. This paper contributes quantitative evidence to the existence of this contested terrain, undergirding the struggle of combating inequality at the levels of coaching, management, and administration.

Furthermore, by highlighting the downstream consequences of race-related managerial job insecurity in a context resembling some of the features of TMTs, we extend prior research on race differences in job security and mobility (Bozeman and Fay, 2013; Cunningham and Sagas, 2005; James, 2000; Maume, 1999; Smith, 1999). NCAA DI college basketball coaching is more racially integrated, at least along Black-White lines, than many labor markets (Lapchick, 2020; Lapchick et al., 2013), making it a hard case for the study of Black-White racial inequality (Eckstein, 2000). Sussing out mechanisms that produce racial disparities in this labor market, therefore, hints at theoretical processes likely to apply more generally (Lieberson, 1997).

And yet, college basketball coaching is undeniably a unique profession, consisting of small staffs who serve at the pleasure of the head coach whose job is tied to seemingly more objective measures of success. It is therefore imperative for future research to examine these processes in fields characterized by less career dependence and, given the Black-White predominance of the field, greater racial diversity.

Nevertheless, the high degree of patrimonialism, when paired with performance, makes this labor market more akin to top management teams (TMTs), where CEOs create management teams of executives who work to realize the CEOs’ organizational goals (e.g., Finkelstein and Hambrick, 1990). And indeed, research on CEO and head coach turnover finds advisers are more likely to be displaced when turnover comes with a call for change (Barron et al., 2011; Savage and Seebruck, 2016). Perhaps it is this similarity that explains why journalists and researchers liken DI college coaches to CEOs of corporations (Katz, 2008; Wangrow et al., 2021).

Thus, our findings offer valuable insights for combating racial disparities in this labor market and perhaps more broadly. First, our findings illustrate how critical it is to break practices that lead to strong racial homophily preferences in work settings (Bacharach et al., 2005; Ibarra, 1995; Stainback, 2008). These processes, when paired with the greater job insecurity of minority managers, disproportionately destabilize the careers of minority subordinates, hindering their ability to amass the human capital necessary for advancement. In demonstrating this we identify one process accounting for race-based differences in the accumulation of human capital over one’s career (Tomaskovic-Devey et al., 2005): differences in the job stability of minority subordinates stemming from the pairing of homophily processes with race-related managerial job insecurity. This points to a second lesson: disproportionately funneling minorities into fraught leadership positions, a process known as the glass cliff (Cook and Glass, 2013), perpetuates racial inequality and racialized organizations by having the greater job insecurity of minority managers trickle down to affect their subordinates—something that is not a race-neutral process. Thus, an equitable distribution of leadership positions should consider not only the proportion of job openings filled by individuals from different racial backgrounds but also the relative quality of those openings. And finally, it is important to ensure that once appointed to leadership positions, minority managers are not held to a racial double standard (Foschi, Ndobo, and Faure, 2019) whereby they are expected to perform at higher levels than their White counterparts. In our data we observe assistant coaches are more likely to be dismissed when they work for head coaches whose teams underperform and this likelihood is greater when the head coach is a minority, in part because these individuals are themselves more likely to be dismissed. When organizations fail to address these three inhibitors, racial disadvantage at the managerial level seeps down to disproportionately inhibit the careers of their subordinates, which, in this case, leads to greater unemployment and forced exits.

Our research, paired with the existing scholarly attention to the intersection of race and sport, may have implications for policy goals seeking to address racial inequalities in coaching. The NCAA may focus on hiring more head coaches of color while ensuring that assistant coaches of color are not disproportionately working under head coaches of color (Foreman and Turick, 2021). Efforts could be made by the NCAA to implement policies establishing minimum standards for teams’ coaching staff demographics over a given period (Foreman and Turick, 2021). Additionally, the NCAA may choose to emulate professional basketball initiatives, such as the NBA’s own Diversity and Inclusion practices, with programs such as an NBA HBCU Fellowship and a Coaches Equality Initiative (CEI), to identify and promote the talent of all coaches. Similar programs from other sports—such as the NFL-NCAA Coaches Academy, which seek to provide an opportunity for former Black quarterbacks to provide quarterback coach training (Foreman and Turick, 2021)—could also be adapted. Future research could provide insights into how, if any, such policies could affect the racial landscape of NCAA coaching.

Footnotes

Acknowledgments

We thank Jarime Chaco, Christine Pham, and Hannah Lewis for their research assistance.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by a University of Houston Grant to Enhance Research on Racism.

Notes

Appendix

Discrete-Time Hazards From a Complementary Log-Log Analysis of Involuntary Dismissal

	Model 1	Model 2	Model 3	Model 4	Model 5
1 year to event	0.14^*** (0.02)	0.20^*** (0.08)	0.14^*** (0.06)	0.10^* (0.10)	0.05^* (0.05)
2 years to event	0.11^*** (.02)	0.16^*** (0.07)	0.12^*** (0.05)	0.08^* (0.08)	0.07^* (0.09)
3 years to event	0.10^*** (0.02)	0.14^*** (0.06)	0.10^*** (0.04)	0.07^* (0.07)	0.04^** (0.05)
4 years to event	0.15^*** (0.03)	0.20^*** (0.09)	0.14^*** (0.06)	0.11^* (0.12)	0.08^* (0.09)
5 years to event	.04^*** (.02)	.06^*** (.03)	0.04^*** (0.02)	0.03^** (0.03)	0.03^** (0.03)
AC White		0.64^* (0.11)	0.70^* (0.12)	0.70^* (0.12)	0.70⁺ (0.15)
Staff position		0.82 (0.40)	0.79 (0.38)	0.78 (0.42)	0.61 (0.46)
DI basketball player		0.86 (0.15)	0.87 (0.15)	0.86 (0.15)	0.82 (0.17)
At alma mater		0.23^** (0.11)	0.24^** (0.11)	0.21^** (0.10)	0.26^** (0.11)
Tenure at school (T)		1.17⁺ (0.10)	1.16⁺ (0.10)	1.15 (0.11)	1.04 (0.08)
T^2		0.99⁺ (0.01)	0.99⁺ (0.01)	0.99 (0.01)	0.99 (0.005)
Public school		0.94 (0.17)	0.90 (0.18)	1.02 (0.20)	0.90 (0.17)
AD non-White		0.99 (0.21)	0.70 (0.16)	0.71 (0.18)	0.55^* (0.16)
AD sex		0.48^*** (0.11)	0.49^** (0.11)	0.57^* (0.13)	0.52^* (0.16)
AD tenure		1.00 (0.01)	0.99 (0.01)	1.00 (0.01)	1.00 (0.01)
HC tenure (HCT)		1.17^* (0.09)	1.20^* (0.10)	1.36^** (0.14)	1.09 (0.06)
HCT^2		0.99^* (0.004)	0.99^* (0.005)	0.99^* (0.01)	1.00 (0.002)
Expenditures relative to conference average (standardized) (ECA)		1.13⁺ (0.08)	1.15⁺ (0.09)	1.38^*** (0.13)	1.07 (0.12)
HC non-White			2.29^*** (0.43)	1.61^* (0.36)	2.03^** (0.47)
Performance				0.99 (0.01)	1.00 (0.01)
Performance relative to conference average (PCA)				1.16^*** (0.04)	1.05 (0.04)
Performance relative to past				1.01 (0.02)	1.01 (0.03)
ECA × PCA				1.05^** (0.02)	1.05^* (0.02)
HC non-White × PCA				1.07^* (0.04)	1.02 (0.04)
HC fired					32.7^*** (7.27)

Notes: ⁺p < .10, ^*p < .05, ^**p < .01, and ^***p < .001 (two-tailed). N = 1444 years nested in 377 people. Standard errors correct for multiple events nested in assistant coaches.

References

Bacharach

Samuel B.

Bamberger

Peter A.

Vashdi

Dana

. 2005. “Diversity and Homophily at Work: Supportive Relations Among White and African-American Peers.” Academy of Management Journal 48(4):619–644.

Baek

Minseo

Bidwell

Matthew

Keller

J. R.

2022. “My Manager Moved! Manager Mobility and Subordinates’ Career Outcomes.” Organization Science 33(5):1701–2083.

Ballinger

Gary A.

Lehman

David W.

Schoorman

F. David

. 2010. “Leader-member Exchange and Turnover before and After Succession Events.” Organizational Behavior and Human Decision Processes 113(1):25–36.

Barron

John M.

Chulkov

Dmitriy V.

Waddell

Glen R.

2011. “Top Management Team Turnover, CEO Succession Type, and Strategic Change.” Journal of Business Research 64(8):904–910.

Bielby

William T.

2012. “Minority Vulnerability in Privileged Occupations: Why Do African American Financial Advisers Earn Less than Whites in a Large Financial Services Firm?” The Annals of the American Academy of Political and Social Sciences 639(1):13–32.

Bonilla-Silva

Eduardo

Dietrich

David

. 2011. “The Sweet Enchantment of Color-Blind Racism in Obamerica.” The Annals of the American Academy of Political and Social Science 634(1):190–206.

Bozeman

Barry

Fay

Daniel

. 2013. “Minority Football Coaches’ Diminished Careers: Why is the ‘Pipeline’ Clogged?” Social Science Quarterly 94(1):29–58.

Breen

Richard

Karlson

Kristian Bernt

Holm

Anders

. 2013. “Total, Direct, and Indirect Effects in Logit and Probit Models.” Sociological Methods & Research 42(2):164–91.

Buffington

Daniel

Fraley

Todd

. 2008. “Skill in Black and White: Negotiating Media Images of Race in Sporting Context.” Journal of Communication Inquiry 32(3):292–310.

10.

Carrington

Ben

. 2013. “The Critical Sociology of Race and Sport: The First Fifty Years.” Annual Review of Sociology 39(1):379–98.

11.

Chin

Christina

. 2015. “Aren’t You a Little Short to Play Ball? Japanese American Youth and Racial Microaggressions in Basketball Leagues.” Amerasia Journal 41(2):47–65.

12.

Collins

Sharon M.

1997. Black Corporate Executives: The Making and Breaking of a Black Middle Class. Philadelphia, PA: Temple University Press.

13.

Cook

Alison

Glass

Christy

. 2013. “Glass Cliffs and Organizational Saviors: Barriers to Minority Leadership in Work Organizations?” Social Problems 60(2):168–187.

14.

Cunningham

George B.

2021. “The Under-Representation of Racial Minorities in Coaching and Leadership Positions in the United States.” Pp. 3–21 in ‘Race’, Ethnicity, and Racism in Sports Coaching, edited by Bradbury

Steven

Lusted

Jim

van Sterkenburg

Jacco

. New York: Routledge.

15.

Cunningham

George B.

Sagas

Michael

. 2005. “Access Discrimination in Intercollegiate Athletics.” Journal of Sport and Social Issues 29(2):148–163.

16.

Day

Jacob C.

2015. “Transitions to the Top: Race, Segregation, and Promotions to Executive Positions in the College Football Coaching Profession.” Work and Occupations 42(4):408–446.

17.

Day

Jacob C.

2018. “Climbing the Ladder or Getting Stuck: An Optimal Matching Analysis of Racial Differences in College Football Coaches’ Job-level Career Patterns.” Research in Social Stratification and Mobility 53(1):1–15.

18.

Eckstein

Harry

. 2000. “Case Study and Theory in Political Science.” Pp. 119–164 in Handbook of Political Science. Vol. 7, edited by Greenstein

Fred I.

Polsby

Nelson W.

19.

Eriksson

Stefan

Rooth

Dan-Olof

. 2014. “Do Employers Use Unemployment as a Sorting Criterion when Hiring? Evidence from a Field Experiment.” American Economic Review 104(3):1014–39.

20.

Festinger

Leon

. 1954. “A Theory of Social Comparison Processes.” Human Relations 7(2):117–40.

21.

Finkelstein

Sydney

Hambrick

Donald C.

1990. “Top-Management-Team Tenure and Organizational Outcomes: The Moderating Role of Managerial Discretion.” Administrative Science Quarterly 35(3):484–503.

22.

Foreman

Jeremy J.

Turick

Robert M.

2021. “The effect of race on lateral moves to coach central positions.” Sport Management Review 24(1):69–91.

23.

Foschi

Martha

Ndobo

Andre

Faure

Alice

. 2019. “Assessing and Blocking Double Standards for Competence.” Advances in Group Processes 36:19–45.

24.

Foy

Steven L.

Ray

Rayshawn

. 2019. “Skin in the Game: Colorism and the Subtle Operation of Stereotypes in Men’s College Basketball.” American Journal of Sociology 125(3):730–85.

25.

Giuliano

Laura

Levine

David I.

Leonard

Jonathan

. 2009. “Manager Race and the Race of New Hires.” Journal of Labor Economics 27(4):589–631.

26.

Gomer

Justin

Ossei-Owusu

Shaun

. 2022. “Coaching While Black: Race, Leadership, and the National Football League.” Journal of Sport and Social Issues.

27.

Greenhalgh

Leonard

Rosenblatt

Zehava

. 1984. “Job Insecurity: Toward Conceptual Clarity.” Academy of Management Review 9(3):438–48.

28.

Greenhaus

Jeffrey H.

Parasuraman

Saroj

. 1993. “Job Performance Attributions and Career Advancement Prospects: An Examination of Gender and Race Effects.” Organizational Behavior and Human Decision Processes 55(2):273–297.

29.

Hartmann

Douglas

. 2000. “Rethinking the Relationships Between Sport and Race in American Culture: Golden Ghettos and Contested Terrain.” Sociology of Sport Journal 17(3):229–253.

30.

Hill

Terrence D.

Davis

Andrew P.

Roos

J. Micah

French

Michael T.

2020. “Limitations of Fixed-Effects Models for Panel Data.” Sociological Perspectives 63(3):357–369.

31.

Huffman

Matt L.

Cohen

Philip N

. 2004. “Racial Wage Inequality: Job Segregation and Devaluation across U.S. Labor Markets.” American Journal of Sociology 109(4):902–36.

32.

Ibarra

Herminia

. 1995. “Race, Opportunity, and Diversity of Social Circles in Managerial Networks.” The Academy of Management Journal 38(3):673–703.

33.

Jackall

Robert

. 1988. Moral Mazes: The World of Corporate Managers. New York: Oxford University Press.

34.

James

Erika Hayes

. 2000. “Race-Related Differences in Promotions and Support: Underlying Effects of Human and Social Capital.” Organization Science 11(5):493–508.

35.

Kalev

Alexandra

. 2014. “How You Downsize Is Who You Downsize: Biased Formalization, Accountability, and Managerial Diversity.” American Sociological Review 79(1):109–135.

36.

Kalleberg

Arne L.

2009. “Precarious Work, Insecure Workers: Employment Relations in Transition.” American Sociological Review 74(1):1–22.

37.

Kanter

Rosabeth Moss

. 1977. Men and Women of the Corporation. New York: Basic Books.

38.

Katz

Andy

. 2008. “Williams, Pitino Have Their Powerhouse Programs Back on Track.” ESPN. 28 March. Accessed on February 9, 2014 (https://sports.espn.go.com/ncb/ncaatourney08/columns/story?colunist=katz_andy&id=3318348).

39.

Lapchick

Richard

Agusta

Robert

Kinkopf

Nathaniel

McPhee

Frank

. 2013. The 2012 Racial and Gender Report Card: College Sport. Orlando: The Institute for Diversity and Ethics in Sport (TIDES). Retrieved March 9, 2014 (https://www.tidesport.org/racialgenderreportcard.html).

40.

Lapchick

Richard

. 2020. The 2019 Racial and Gender Report Card: College Sport. Orlando: The Institute for Diversity and Ethics in Sport (TIDES) (Retrieved August 1, 2020).

41.

Huisi (Jessica)

Hausknecht

John P.

Dragoni

Lisa

. 2020. “Initial and Longer-Term Change in Unit-Level Turnover Following Leader Succession: Contingent Effects of Outgoing and Incoming Leader Characteristics.” Organization Science 31(2):458–476.

42.

Lieberson

Stanley

. 1997. “Modeling Social Processes: Some Lessons from Sports.” Sociological Forum 12(1):11–35.

43.

Lowe

Travis Scott

. 2018. “Perceived Job and Labor Market Insecurity in the United States: An Assessment of Workers’ Attitudes from 2002 to 2014.” Work and Occupations 45(3):313–45.

44.

Maume

David J.

Jr.

1999. “Glass Ceilings and Glass Escalators: Occupational Segregation and Race and Sex Differences in Managerial Promotions.” Work and Occupations 26(4):483–509.

45.

Meyer

Craig

. 2019. “Black head coaches rare at top tier of college basketball.” Associated Press. Retrieved July 25, 2020 (https://apnews.com/7b5ef1bb52e9406081e1481cb791a4a0).

46.

Morgenroth

Thekla

Kirby

Teri A.

Ryan

Michelle K.

Sudkamper

Antonia

. 2020. “The Who, When, and Why of the Glass Cliff Phenomenon: A Meta-Analysis of Appointments to Precarious Leadership Positions.” Psychological Bulletin 146(9):797–829.

47.

Murguia

Edward

Telles

Edward E

. 1996. “Phenotype and Schooling among Mexican Americans.” Sociology of Education 69(4):276–89.

48.

Norris

Davon

Moss-Pech

Corey

. 2022. “White Men Can’t Jump, but Does It Even Matter? Exit Discrimination in the NBA.” Social Forces 100(3):961–89.

49.

Oates

Thomas P.

2017. “Selling Streetball: Racialized Space, Commercialized Spectacle, and Playground Basketball.” Critical Studies in Media Communication 34(1):94–100.

50.

Ogden

David C.

Hilt

Michael L.

2003. “Collective Identity and Basketball: An Explanation for the Decreasing Number of African-Americans on America's Baseball Diamonds.” Journal of Leisure Research 35(2):213–227.

51.

Pedulla

David S.

2016. “Penalized or Protected? Gender and the Consequences of Nonstandard and Mismatched Employment Histories.” American Sociological Review 81(2):262–89.

52.

Pedulla

David S.

Pager

Devah

. 2019. “Race and Networks in the Job Search Process.” American Sociological Review 84(6):983–1012.

53.

Quillian

Lincoln

Pager

Devah

Hexel

Ole

Midtbøen

Arnfinn H

. 2017. “Meta-Analyses of Field Experiments Shows No Change in Racial Discrimination in Hiring over Time.” Proceedings of the National Academy of Sciences 114(41):10870–5.

54.

Rabe-Hesketh

Sophia

Skrondal

Anders

. 2012. Multilevel and Longitudinal Modeling Using Stata, Volume II: Categorial Responses, Counts, and Survival. 3rd ed. College Station, TX: Stata Press.

55.

Rankin-Wright

Alexandra J.

Hylton

Kevin

Norman

Leanne

. 2016. “Off-colour Landscape: Framing Race Equality in Sport Coaching.” Sociology of Sport Journal 33(4):357–368.

56.

Ray

Victor

. 2019. “A Theory of Racialized Organizations.” American Sociological Review 84(1):26–53.

57.

Salaga

Steven

Juravich

Matthew

. 2020. “National Football League Head Coach Race, Performance, Retention, and Dismissal.” Sport Management Review 23(5):978–91.

58.

Sagarin

Jeff

. 2020. “Jeff Sagarin Ratings: NCAAB.” USA Today. Retrieved (https://www.usatoday.com/sports/ncaab/sagarin/).

59.

Savage

Scott V.

Seebruck

Ryan

. 2016. “Race, Supervisorial Change, and Job Outcomes: Employability Resilience in NCAA Division I College Basketball Coaching.” Sociological Quarterly 57(3):415–36.

60.

Seebruck

Ryan

Savage

Scott V.

2020. “To Promote or Hire? How Racial Processes and Organizational Characteristics Affect Internal Promotions in NCAA Division I College Basketball Coaching.” Social Problems 67(3):546–564.

61.

Shapiro

Debra L.

Hom

Peter W.

Shen

Wei

Agarwal

Rajshree

. 2016. “How Do Leader Departures Affect Subordinates’ Organizational Attachment? A 360-Degree Relational Perspective.” Academy of Management Review 41(3):479–502.

62.

Shoss

Mindy K.

2017. “Job Insecurity: An Integrative Review and Agenda for Future Research.” Journal of Management 43(6):1911–39.

63.

Smith

Ryan A.

1999. “Racial Differences in Access to Hierarchical Authority: An Analysis over Time, 1972-1994.” The Sociological Quarterly 40(3):367–395.

64.

Stainback

Kevin

. 2008. “Social Contacts and Race/Ethnic Job Matching.” Social Forces 87(2):857–886.

65.

Tomaskovic-Devey

Donald

. 2014. “The Relational Generation of Workplace Inequality.” Social Currents 1:51–73.

66.

Tomaskovic-Devey

Donald

Thomas

Melvin

Johnson

Kecia

. 2005. “Race and the Accumulation of Human Capital across the Career: A Theoretical Model and Fixed-Effects Application.” American Journal of Sociology 111(1):58–89.

67.

Turick

Robert

Bopp

Trevor

. 2016. “A Current Analysis of Black Head Football Coordinators at the NCAA DI-FBS Level.” Journal of Intercollegiate Sport 9(2):282–302.

68.

Valletta

Robert G.

1999. “Declining Job Security.” American Journal of Economics 17(4):S170–S197.

69.

Wangrow

David B.

Schwartz

Evan

Hughes-Moran

Margaret

. 2021. “Executive Dismissal or Retention? A Study of Performance, Power, and Survival for College Basketball Coaches.” Journal of General Management 47(1):56–68.

70.

Washington

Jesse

. 2021. “More Black head coaches doesn’t mean college basketball’s problem is solved.” Retrieved November 9, 2021 (https://theundefeated.com/features/more-black-head-coaches-doesnt-mean-college-basketballs-problem-is-solved/).

71.

West

Brady T.

2008. “A Simple and Flexible Rating Method for Predicting Success in the NCAA Basketball Tournament: Updated Results from 2007.” Journal of Quantitative Analysis in Sports 4(2):8.

72.

Wilson

George

. 2009. “Race and Exits from White Collar Jobs: Differences in Downward Mobility Processes among African American, Latino, and White Males.” Research in Social Stratification and Mobility 27(3):143–156.

73.

Wilson

George

McBrier

Debra Branch

. 2005. “Race and Loss of Privilege: African American/White Differences in the Determinants of Job Layoffs from Upper-Tier Occupations.” Sociological Forum 20(2):301–321.

74.

Wilson

George

Roscigno

Vincent J.

2010. “Race and Downward Mobility from Privileged Occupations: African/American/White Dynamics across the Early Work-Career.” Social Science Research 39(1):67–77.

75.

Wolfe

Richard A.

Weick

Karl E.

Usher

John M.

Terborg

James R.

Poppo

Laura

Murrell

Audrey J.

Dukerich

Janet M.

Core

Deborah Crown

Dickson

Kevin E.

Jourdan

Jessica Simmons

. 2005. “Sport and organizational studies: Exploring synergy.” Journal of Management Inquiry 14(2):182–210.