discussion posted 12

Who’s Calling the Shots? Women Coaches in Division I Women’s Sports n

Susan Welch, The Pennsylvania State University

Lee Sigelman, The George Washington University

Objective. During the past 30 years, women have become an increasingly small proportion of coaches of women’s sports. We test several explanations for why some institutions have women coaches and others do not. Methods. Taking the 329 NCAA Division I institutions with women’s sports programs as the cases in point, we explore possible explanations for the dearth of women coaches: the resources and prestige of an institution or sport; the gender of the athletic director; and institutional traditionalism. Results. Within Division I schools, women coaches are more frequently found in more prestigious, resource-richer institutions and those that devote more resources to women’s sports. Conclusions. Although the analysis is confined to Division I schools, it appears that more institutional and organizational efforts need to be made to increase the number of women in the eligible pools for head coaching jobs.

Title IX of the Educational Amendments of 1972 mandated equality between men’s and women’s college sports. Since then, as women’s sports have skyrocketed in participation and popularity, the proportion of women’s teams coached by women has plummeted. In 1972, approximately 90 per- cent of the members of the coaching staffs for women’s teams nationwide were women, but by 2006 that figure had dropped all the way to 42 percent (Carpenter and Acosta, 2006; Wilson, 2007). This trend runs directly counter to the progress that women made during the same period in many traditionally masculine occupations (see Wharton, 2000; Padavic and Res- kin, 2002, for a recent review). Although a glass ceiling still exists in many occupations, women’s share of professional and leadership positions in business, politics, and religion has grown markedly since the 1970s and women have made significant inroads into prestigious professions such as medicine and law. Women have also increased their representation and clout in academia. They now constitute the majority of undergraduate students, and their share of graduate and professional students, faculty members,

nDirect correspondence to Susan Welch, Office of the Dean, College of the Liberal Arts, The Pennsylvania State University, 110 Sparks Bldg., University Park, PA 16803 [email protected]. Upon request, the first-named author will share all data and coding information with those wishing to replicate this study. The authors appreciate the assistance of Nick Stark and the useful suggestions of Barry Lee, Robert Drago, and Sandra Hanson.

SOCIAL SCIENCE QUARTERLY, Volume 88, Number 5, December 2007 r 2007 by the Southwestern Social Science Association

senior administrators, and college and university presidents has mush- roomed.

So in contrast to their earlier numerical dominance as coaches of college women’s teams, women now occupy only about four of these positions in 10. How can the relative paucity of women coaches be explained? Although useful data on patterns of growth in women’s sports and women’s sports leadership have been compiled (see especially Carpenter and Acosta, 2006) and gender differences in coaches’ ambitions and motivations have often been examined (see, e.g., Hasbrook, 1988; Sagas and Cunningham, 2004; Marback et al., 2005), explanations of the gender patterning of coaching positions remain largely untested.1 Moreover, findings about the situation of women college coaches have rarely made their way into broader analyses of occupational gender stratification (but see Knoppers, 1989).

To enhance understanding of occupational gender stratification in general and gender equity in collegiate sports in particular, we begin by examining longitudinal trends in the presence of women coaches in NCAA Division I (DI) colleges and universities in light of what is already known about changing gender representation in occupations. Then, focusing on four DI women’s sports, we test explanations of why women are more likely to occupy coaching positions in some settings than in others. Although DI teams are a subset of all college athletic teams, the proportion of women coaches in DI (44 percent) is similar to those in DII (36 percent), and DIII (44 percent) (Carpenter and Acosta, 2006:16), and all three proportions have declined over the years.

The Gender Composition of Occupations

Studies of occupational gender stratification have established that most occupations are dominated by one sex or the other, that male-dominated occupations pay better, on average, than do female-dominated ones, and that even within the same occupation, men and women are often differ- entiated by task and function (Bielby and Baron, 1984:27–55; Padavic and Reskin, 2002).

Gender and Occupations

The occupational stratification system is shaped by gender relationships inside and outside the workplace, especially by the traditional role of women as family caregivers (Hartmann, 1981; Okamoto and England, 1999).

1Theberge (1990, 1993) has explored the perceptions of women coaches about their jobs, and several studies have focused on women athletic directors (Whisenant, 2003; Grappendorf and Lough, 2006).

1416 Social Science Quarterly

Although that role has undergone change in recent decades, historically it has shaped women’s educational opportunities and their fit within the world of work (Reskin and Hartmann, 1986; Padavic and Reskin, 2002) as well as the attitudes of employers and co-workers about appropriate ‘‘jobs for women.’’

Underlying some explanations of occupational gender stratification are the ideas that for many women the family is a higher priority than the job (Polachek, 1981; Reskin and Padavic, 1994:75–79; Padavic and Reskin, 2002:ch. 3) and that this value system limits both their attention to pre- paring for high-responsibility jobs and the time they spend in the labor force paving the way to job advancement. Many women seek only part-time positions that enable them to spend time with their families but also limit their job responsibilities and income. The idea that women are socialized into wanting ‘‘women’s jobs’’ also persists, though it has been undermined by studies of changing career preferences among young girls (Okamoto and England, 1999).

Changes in ‘‘Women’s Jobs’’

Because cases of formerly female occupations becoming largely male are nonexistent to our knowledge (Padavic and Reskin, 2002:72),2 research on occupational gender stratification has focused on the movement of women into previously male-dominated occupations. Such movement can occur for several reasons. When labor is in short supply, employers that have always relied on men become more open to hiring women, a tendency personified by ‘‘Rosie the Riveter’’ during World War II. Another possibility is that men may come to see a particular occupation as less desirable, in which case women will be called on to fill the gap—as has occurred among elementary and secondary teachers (Reskin and Roos, 1990:42), insurance adjusters (Hartmann, Kraut, and Tilly, 1986), and bank employees (Strober and Arnold, 1987). Employers also may reconsider the desirability of hiring women because of women’s increasing skills (Reskin and Roos, 1990:111– 27) or their own increased knowledge of women’s productivity (Bielby and Bielby, 1988), or in response to the growing number of women in markets served by an industry (Baron and Bielby, 1985; Reskin and Roos, 1990:129–44). Changes in hiring patterns can also reflect new leadership at the top of companies and public organizations (Cohen, Broschak, and Haveman, 1998; Reskin and Roos, 1990:57, 93–111), lawsuits against cer- tain kinds of discrimination, and, more broadly, lessened tolerance for gender discrimination in education and employment.

2Much is made of the growing number of males in the nursing profession in the United States, but only about 11 percent of nurses are men (Statistical Abstract of the United States, 2006:Table 604; see also Williams, 1992).

Women Coaches in Division I Women’s Sports 1417

Explaining the Decline in the Proportion of Women Coaches

Women’s college sports were not organized nationally until 1941 (Wom- en’s Sports Foundation, n.d.), when the first intercollegiate women’s golf championship was held. Growing interest and participation in women’s sports led to the formation in 1971 of the Association for Intercollegiate Athletics for Women (AIAW), which began to sponsor national tourna- ments in several sports, including basketball in 1972, the year Title IX was enacted. Participation in women’s collegiate sports immediately soared (see Table 1), and in 1981 the NCAA, which had long spurned women’s sports, began sponsoring national women’s tournaments. The AIAW rapidly lost members and revenue to the NCAA and in 1983 it ceased operation altogether.

As the number of women’s teams grew, the proportion coached by wom- en shrank (Carpenter and Acosta, 2006); the correlation between the mean number of women’s sports per school in a given year and the proportion of women coaches nationwide from 1978 (when the Department of Education began collecting these data) through 2006 is �0.81. What is especially striking is the pace at which the two trend lines diverged in the years immediately after the enactment of Title IX.3 Since 1984, the first year that divisional differences were tracked, the decline has been steepest in DII and DIII schools, and more modest in DI (Carpenter and Acosta, 2006).

Coaching women’s sports became a more desirable job after 1972, with higher pay, status, and visibility—the opposite of the situation in many fields where women have advanced, such as education, insurance, and banking. Moreover, the supply of women candidates, at least in the early years after Title IX’s passage, was outpaced by the growth in the number of women’s teams. With the average number of women’s sports per school growing rapidly during the 1970s, thousands of coaching jobs had to be filled, many in sports in which schools had not previously fielded teams. Under those conditions, and as gender norms were changing and demands for successful women’s teams were growing, men increasingly loomed as attractive candidates in an enterprise that no longer was seen as serving just a ‘‘women’s constituency,’’ but rather the larger base of students and alumni.

3The proportion of women’s team coaches in 1972 and before is an estimate by those familiar with women’s sports at that time. We spot checked 30 sports programs established before 1970 among the four sports on which we focus, including all the schools that had soccer, softball, and volleyball programs and a random group of those with basketball pro- grams. Our check turned up only one male coach in 1970 and 19 females (the rest were impossible to ascertain with the fragmentary documentation that has survived), a ratio sup- porting the estimates given in the text. In high school interscholastic sports, as the number of high school sports for girls increased after Title IX, the proportion of women coaches decreased by about the same magnitude as the collegiate decline (see the summary in He- ishman, Bunker, and Tutwiler, 1990). This decline persisted across states and sports and for women athletic directors, too (Whisenant, Pederson, and Obenour, 2002; Grappendorf and Lough, 2006).

1418 Social Science Quarterly

In fact, men have won nearly 75 percent of the coaching jobs in collegiate women’s sports since 1999 (Wilson, 2007).

Explanations for the Dearth of Women Coaches Today

U.S. colleges and universities have now had more than three decades to integrate women’s sports as full participants in their athletic programs. The expansion of women’s teams in the 1970s and 1980s produced a new gen- eration of former players, assistant coaches, and others with the potential to become coaches. That being the case, why do most women’s teams have male coaches? Drawing on previous work on gender and occupation, we consider several explanations in an institution-level analysis intended to complement previous individual-level analyses of the experiences and atti- tudes of female coaches and athletes.

Prestige and Resources

Until the 1970s, women’s athletics was a small enterprise, mostly shunted out of sight in third-rate facilities and shackled by bare-bones budgets and inferior equipment. Of course, schools without resources for either male or female athletics still exist, as do schools where women’s athletics are the poor stepchildren of men’s athletics, and sports that are in the limelight while others are invisible. Do factors such as these help explain the presence or absence of women’s coaches?

TABLE 1

Trends in the Percentage of Women Coaches and the Number of Women’s Teamsa

Year Women as a % of

Coaches Women’s Teams

per School

1970 901 2.5 1978 58 5.6 1980 54 6.5 1985 51 7.0 1990 47 7.2 1995 48 7.3 2000 46 8.1 2004 44 8.3 2006 42 8.5

aData from Carpenter and Acosta (2006). The proportion of coaches who were women in and before 1972 is an estimate. Systematic tracking appears to have started in 1978. This table includes all divisions. Data on the numbers of coaches not available.

Women Coaches in Division I Women’s Sports 1419

Resources Devoted to Women’s Sports. Some institutions are wealthy and others are financially strapped. Beyond this, some devote substantial re- sources to women’s athletics, with a full array of scholarships, a bevy of coaches, and a large operating budget, while others offer few scholarships and have part-time coaches running their programs on shoestring budgets.

Differentiation According to Sport and Institutional Resources. College sports are highly stratified from sport to sport and from institution to institution. Some women’s teams enjoy high prestige and visibility, with thousands of people following them and basking in the reflected glory of their athletic successes (Cialdini et al., 1976). Others struggle to maintain financial viability and are little known outside their home communities. DI status is itself an indicator of such prestige, but as we have already noted, differences in the proportion of women coaches among the three divisions are small (Carpenter and Acosta, 2006).

If patterns in other occupations hold for the coaching profession, women should be differentially relegated to the lower levels of the sports-status hierarchy, just as historically women were found in the less prestigious and lower-paid echelons of many professions. In other words, women may still be disproportionately found in the coaching jobs most like those they held before Title IX.

At larger schools, women’s basketball typically draws larger crowds and more television revenue than other women’s sports, and many schools offer women full scholarships and prospects for advancement into careers in pro- fessional sports. On the other hand, athletes participate in sports like lacrosse, gymnastics, swimming, and field hockey far from the limelight, many on partial scholarships or none at all. The popularity and prestige of a sport may influence whether women are hired as coaches. If discrimination is at work, women coaches should be clustered in less prestigious sports. However, whereas the few women coaches of male teams are indeed more often found in less prestigious sports (Kane and Stangl, 1991), in women’s sports more women are coaches in basketball, volleyball, soccer, and softball than on less prestigious and less visible teams (Stahura and Greenwood, 2002).

Of course, the presence of women’s coaches also reflects the distinctive historical development of each sport. For example, although women’s soccer is now the fourth most widely offered women’s sport, with 89 percent of all NCAA schools fielding a team, in 1977 fewer than 3 percent of these in- stitutions had women’s soccer teams. The development of a women’s coach- ing cadre in soccer therefore lagged behind that of women’s basketball or volleyball, which in 1977 were played by 90 percent and 80 percent, respec- tively, of these schools (Carpenter and Acosta, 2006). Similarly, because only 22 schools currently sponsor varsity volleyball for men, prospective volleyball coaches must turn largely to women’s teams, producing substantial male- female competition for these jobs.

1420 Social Science Quarterly

This array of resource-related considerations led us to hypothesize that women coaches would be found disproportionately in (1) institutions that provide more modest support for women’s athletics, (2) smaller and poorer institutions, and (3) less prestigious sports.

Male-Dominated Athletic Departments

Potential or current women coaches may be disadvantaged as a conse- quence of the deeply ingrained male culture of athletic departments. Al- though the notion of a ‘‘gendered institution’’ provokes vigorous debate (see Britton, 2000), its applicability to college sports is undeniable. The great majority of DI athletic directors are men (Carpenter and Acosta, 2006; Wilson, 2007), and it seems certain that their personal and professional networks are disproportionately male. Whisenant, Peterson, and Obenour (2002) describe athletic departments as one of the purest manifestations of ‘‘hegemonic masculinity.’’ In such an environment, hiring men as head coaches could well seem natural, for women’s teams as well as men’s. Moreover, many women coaches do not feel supported in such an envi- ronment (Wilson, 2007). Thus, Stangl and Kane (1991) attribute the de- cline in women high school coaches to the dearth of women athletic directors. Male athletic directors, they speculate, are more likely to hire males because they perceive a shortage of qualified women coaches and a tendency for women to have family obligations (Acosta and Carpenter, 1988, cited by Stangl and Kane, 1991). That interpretation seems consistent with the finding, referred to above, that leadership at the top of an orga- nization can be a key to the hiring and advancement of women within an organization (Cohen, Broschak, and Haveman, 1998; Reskin and Roos, 1990). We therefore hypothesized that women coaches would be more likely to be found in institutions with women athletic directors.

Traditional Environments

It also seems possible that women coaches would be more likely to be found in institutions and parts of the country where more traditional gen- der-role attitudes dominate. For example, Drago and his colleagues (2005) suggest that the anti-lesbian bias found in many institutions deters the development and hiring of women coaches, a perception that many coaches seem to share (Sweet and Malec, 1988; Redden, 2007).4 Although the evidence is mixed (Weinberg, Reveles, and Jackson, 1984; Frankl and Bab-

4Drago et al. (2005) cite Census estimates that as many as 6 percent of full-time women coaches are lesbians. It is not clear, however, whether this figure is high or low relative to other occupational categories.

Women Coaches in Division I Women’s Sports 1421

bitt, 1998), some have suggested that women athletes themselves prefer to be coached by men and that this preference in turn influences athletic admin- istrators. It follows that southern conservatism on a variety of gender-related issues could contribute to a dearth of women coaches (Ward, 2004:704). Whereas more traditional gender-role orientations are also more likely to dominate in religious institutions, public institutions might be subject to pronounced internal and external pressures toward openness to gender equality, for both legal and political reasons. Thus, we hypothesized that women coaches would be less likely to be found in more traditional institutions vis-a-vis gender—southern and religiously affiliated schools and in private schools.

Data and Methods

We drew data on four sports—basketball, volleyball, soccer, and soft- ball—in the 329 DI schools from the Department of Education files on equity in college athletics and from individual schools’ athletic websites. We focus on DI schools because they have the most visible women’s athletic programs and because doing so enables us to control for several character- istics; all these schools actively recruit athletes, most offer scholarships, and all have made substantial investments in sports. At the same time, DI schools vary widely in overall resources, size, visibility, resources devoted to women’s sports, institutional control (public, private secular, or religious), and location.

Except as noted, all institutional data are for 2004–2005. The four sports that we examine are among the top five in terms of the proportion of schools offering them (ranging from 98 percent for basketball to 87 percent for softball) (Carpenter and Acosta, 2006)5 and they are the most popular in terms of attendance; the top women’s basketball team averages over 15,000 spectators per game, volleyball nearly 7,000, soccer 3,400, and softball 1,800.6 The sports differ in another way, too. The median dates of devel- opment of varsity basketball and volleyball at DI schools were soon after the adoption of Title IX, 1974 and 1976, respectively. The typical DI school adopted softball as a varsity sport in 1980 but did not launch women’s soccer until 1994.

5Although cross-country (89 percent) currently ranks fourth in participation, several fac- tors led us to substitute softball in its place. Cross-country is often merged with track in record keeping (as in the federal Title IX reports). Often, there are dual coaches in track and cross-country. Attendance records for cross-country and track were not available on the NCAA website, suggesting the low visibility of the sport. Basketball and volleyball were also the top women’s sports in 1977, but soccer (in 3 percent of the institutions) and softball (in 48 percent) were much less popular then.

6These are the latest data reported by the NCAA at its home page, but not all are for 2004–2005.

1422 Social Science Quarterly

Variables

The dependent variables are (1) the proportion of all of a school’s wom- en’s teams coached by women, (2 and 3) the proportions for women’s teams in the four ‘‘major’’ sports and in ‘‘minor’’ sports, and (4) whether or not the head coach in a given sport was a woman, all as of 2005–2006. We drew the information about coaches in the four sports from each school’s website, supplemented for the other dependent variables by data available online at the Department of Education’s Office of Postsecondary Education Equity in Athletics website (http://ope.ed.gov/athletics/Search.asp) and periodically reported in the Chronicle of Higher Education.

Status and Resources. The independent variables include measures of sta- tus and resources. We distinguish among schools with DI-A, DI-AA, and DI-AAA athletic programs. DI-AA schools (e.g., most historically black schools, Bucknell, Youngstown State, and Montana) generally have smaller stadiums, field fewer teams, and generate less revenue from athletics than their DI-A counterparts; DI-AAA schools (e.g., Bradley, George Washing- ton, and Xavier) do not field football teams but often do have high-visibility teams in basketball or other sports. Within the DI-A category, we further distinguish between schools that belong to a ‘‘power conference,’’ that is, a conference that receives an automatic bid to the Bowl Championship Series (the Atlantic Coast, Big 10, Big 12, Southeastern, Pacific 10, and Big East, plus Notre Dame) and those that do not. The power schools’ football- generated bowl and television revenues enable them to support other sports and generally to dominate the major sports rankings. DI-A schools not in power conferences include many smaller state schools, such as Akron, Boise State, Colorado State, and Southern Mississippi.

These divisional differences reflect different institutional sizes and resource bases. For example, the mean undergraduate enrollment at power conference schools is 17,000; at other DI-A schools it is 13,000, while DI-AA and DI- AAA schools average 7,000. Schools in power conferences spend an average of more than $6 million on women’s athletics, compared to less than $4 million in other DI-A schools and somewhat over $2 million in the other DI categories.

Support for Women’s Sports. We assembled seven measures of actual or potential institutional support for women’s sports: the number of sports in which a school fielded a team, its total scholarship support for women athletes, its overall operating expenses for women’s athletics, the proportion of its total recruiting costs devoted to women’s sports, the proportion of its total coaching salaries devoted to women’s sports, the female proportion of its undergraduate student body, and the ratio of its women sports partic- ipants to its women students (the latter being a key measure of compliance

Women Coaches in Division I Women’s Sports 1423

with Title IX (U.S. Department of Education, 2005)).7 These measures resolved into two underlying dimensions. The first dimension, on which the first, second, third, and (albeit less strongly) seventh measures were the high loaders, primarily represents the absolute amount of institutional resources devoted to women’s athletics; the second dimension, on which the fourth, fifth, and sixth measures loaded highly, represents the relative share of resources devoted to women’s athletics.8 We generated scores on each dimension by standardizing each underlying variable (mean 5 0, SD 5 1), weighting each standardized value by the factor score coefficient for that variable, and then summing over the seven variables.

These data—and the financial data in particular—have various limita- tions. For example, whereas some schools report an exact balance between revenues and expenditures for women’s sports, others report much larger expenditures than revenues; because the great majority of women’s teams are not moneymakers, most of those reporting budgets that are balanced or in the black must have been receiving and reporting university subventions that made up the difference, whereas other schools either did not receive or did not report such subventions. Sometimes, reports of revenues and expendi- tures from similarly placed schools differ wildly, prompting suspicion that the revenue and expenditure bases for calculations differed. Even then, anomalies remain. The number of coaches does not always match the number of women’s sports; a few schools have coaching vacancies and sev- eral have more than one head coach per sport, particularly in swimming and diving and in track and field, but occasionally in other sports.9

Gender of the Athletic Director

Data on whether an athletic director was male (0) or female (1) again came from the school’s websites. In the handful of cases where the website did not give the name of the athletic director or where a name left the gender of the person ambiguous, we followed up by contacting the athletic de- partment. Only 8 percent of DI schools had women athletic directors, so there was not much variance on this measure. The data at hand did not permit us to determine whether women athletic directors were more likely to hire women coaches than were male athletic directors because the hiring of some coaches may have occurred before, or at the same time as, the ap- pointment of the athletic director. What we could test here was whether

7hhttp://ope.ed.gov/athletics/Search.aspi. 8Extraction was via principal component analysis with an oblique rotation with Kaiser

normalization. The rotation converged in three iterations using SPSS. The two factors are correlated at 0.01. The eigenvalue of the first component is 2.25, the second 2.05, and the two explain 61 percent of the variation. Other components have eigenvalues below 1.0.

9There were no instances of shared coaching duties in the four major sports we examined.

1424 Social Science Quarterly

some athletic departments were likely to be more women-friendly, as in- dicated by having a woman in this key administrative post.

To represent gender-role traditionalism, we include in our models whether (1) or not (0) a school is northern and whether its control is public, private secular, or religious. Public schools are the omitted category throughout, and the other two types are represented by two dummy variables. Despite ex- pecting more traditional schools to have fewer women coaches, we were also open to the possibility that leaders of schools in the South and religious schools might be reluctant to have men coach young women.

Findings

Table 2 shows the distribution of women coaches across DI categories and across sports. In the four sports considered here, women coaches were most plentiful in power conference schools, followed in order by other DI-A, DI- AA, and DI-AAA schools. In 83 percent of schools in power conferences, but in only 58 percent in DI-AAA schools, half or more of the head coaches of women’s teams were women.

In the power conferences, women coaches were most prevalent in bas- ketball, the most visible of the four sports. Women coaches also predom- inated in softball, except in DI-AAA schools. (The cross-divisional differences are statistically significant in basketball and softball, but not in volleyball or soccer.) Women were less likely to be coaches in less visible sports such as field hockey, lacrosse, swimming, and track than in the major sports, but there were also significant divisional differences in the proportion of women coaches in these sports (cf. Stahura and Greenwood, 2002).

The analyses summarized in Table 3 show the results for the first, second, and third dependent variables. These models had very limited overall pre- dictive power, with R2 ranging from 0.06 to 0.14. Nonetheless, the results point toward some interesting differences. Across all sports, women coaches were, other factors being equal, most prevalent in DI-A schools, including those in power conferences, and least so (behind the power schools by nine percentage points) in DI-AAA schools. The same pattern held, albeit in more pronounced form, for just the four ‘‘major’’ sports considered here; as a group, coaches of these sports were 17 percentage points more likely to be women in power conference schools than in DI-AAA schools, with the counterpart figures for other DI-A schools and DI-AA schools being 14 and 12 percentage points, respectively. In the ‘‘minor’’ sports, no significant divisional differences emerged.

Over and above these divisional differences, women were also more likely to occupy coaching positions in institutions that devoted more resources to women’s sports. Across all sports and for minor sports, the more resources an institution devoted to intercollegiate athletics, the higher the percentage of women coaches. Thus, not just larger, wealthier institutions, but also

Women Coaches in Division I Women’s Sports 1425

T A

B L E

2

W o m

e n ’s

P ro

p o rt

io n

o f

D iv

is io

n I

B a s k e tb

a ll,

S o c c e r,

S o ft

b a ll,

a n d

V o lle

y b a ll

C o a c h e s

o f

W o m

e n ’s

T e a m

s

P ro

p o rt

io n

o f

C o a c h in

g P

o s it io

n s

H e ld

b y

W o m

e n

% o f

S c h o o ls

D iv

is io

n I-

A P

o w

e r

C o n fe

re n c e

S c h o o ls

D iv

is io

n I-

A N

o n p

o w

e r

C o n fe

re n c e

S c h o o ls

D iv

is io

n I-

A A

S c h o o ls

D iv

is io

n I-

A A

A S

c h o o ls

L e s s

th a n

h a lf

2 8

1 7

2 6

2 6

4 2

H a lf

3 1

3 5

2 7

3 6

2 8

M o re

th a n

h a lf

4 1

4 8

4 7

3 8

3 0

n

1 0 0 %

1 0 0 %

1 0 0 %

1 0 0 %

1 0 0 %

O v e ra

ll %

in th

e fo

u r

s p

o rt

s 5 5

6 3

5 9

5 4

4 7

n

B a s k e tb

a ll

6 6

8 5

6 3

5 7

6 5

n

V o lle

y b

a ll

5 5

5 1

5 8

5 7

5 3

S o c c e r

3 4

4 2

3 0

3 4

2 9

S o ft b

a ll

6 6

7 7

8 4

6 5

4 5

n

A ll

o th

e r

w o m

e n ’s

s p

o rt

s 3 9

4 7

4 1

3 7

3 5

n

N 3 2 9

6 7

6 2

1 1 3

8 7

n D

iff e re

n c e s

a c ro

s s

d iv

is io

n ty

p e s

a re

s ig

n ifi

c a n t

a t

0 .0

5 .

N O

T E :

T h e

ta b

le in

c lu

d e s

o n ly

s c h o o ls

w it h

a t

le a s t

th re

e o f

th e

fo u r

s p

o rt

s .

T h is

a n d

s u b

s e q

u e n t

ta b

le s

in c lu

d e

o n ly

D I in

s ti tu

ti o n s .

N s

b y

s p

o rt

: b

a s k e tb

a ll,

3 2 9 ;

v o lle

y b

a ll,

3 1 2 ;

s o c c e r,

3 0 2 ;

s o ft b

a ll,

2 7 0 ;

a ll

o th

e rs

, 3 2 9 .

1426 Social Science Quarterly

those that devoted more resources to women’s sports had more women coaches. In the aggregate, women coaches were much more likely to be found in higher-profile, resource-richer institutions.

Schools with women athletic directors were more likely to have women coaches, except in the minor sports. Women coaches in all sports and the four major sports were 10 and 14 percentage points, respectively, more likely to be found in institutions with women athletic directors than in other DI institutions.

The impact of gender-role traditionalism, at least as we have measured it, was small and inconsistent. Religious schools were less likely to have women coaches in all sports considered jointly. However, women coaches were more likely to be found in the South when looking at all sports and the four major sports.

Grouping together coaches of different sports can obscure patterns that hold only for particular sports. Thus, in Table 4 we report analyses that

TABLE 3

Predictors of the Percentage of Women’s Coaches in Division I Women’s Sports

Predictor All Sports Four Major

Sports Other Sports

DI-A power conference schools 0.09 0.17 0.08 (0.05) n (0.07) n (0.07)

Other DI-A schools 0.08 0.14 0.06 (0.04) n (0.05) n (0.06)

DI-AA schools 0.06 0.12 0.05 (0.03) n (0.06) n (0.05)

Women’s sports resources 0.04 0.04 0.05 (0.01) n (0.02) n (0.02) n

Women’s sports share 0.03 0.05 0.02 (0.01) n (0.02) (0.02)

Woman athletic director 0.10 0.14 0.07 (0.03) n (0.06) n (0.06)

Private secular school 0.04 0.05 0.01 (0.02) (0.03) (0.04)

Religious school �0.07 �0.06 �0.06 (0.03) n (0.05) (0.05)

North �0.05 �0.11 �0.00 (0.02) n (0.03) n (0.03)

Constant 0.42 0.48 0.34 (0.03) n (0.05) n (0.05) n

R 2 0.14 0.13 0.06

npo0.05, one-tailed. N 5 329.

NOTE: The entries are unstandardized regression coefficients, with standard errors in paren- theses. The omitted categories are DI-AAA for divisions and public for type of institution. Major sports include basketball, volleyball, soccer, and softball. To be included in the ‘‘four major sports’’ model, a school had to field teams in at least three of the four sports. Women’s sports resources and share are factor scores based on the seven items described in the text and Appendix (obtainable from the authors). The estimates are ordinary least squares regression coefficients, with standard errors in parentheses.

Women Coaches in Division I Women’s Sports 1427

parallel those summarized in Table 3 but break out the major sports one at a time. In these finer-grained analyses, each sport displayed a distinctive pat- tern. Women basketball coaches were significantly more likely to be found in power conference schools than anywhere else. Converting the logit co- efficients into probabilities brings the magnitude of this difference into focus. For a hypothetical power conference school that fell at the mean on both resource dimensions, did not have a woman athletic director, was public rather than private, and was located in the South, the probability of having a woman as the coach of the women’s basketball team was 0.87. By contrast, for a DI-A school that was otherwise identical but did not belong to a power conference, the probability was only 0.60, and for DI-AA and I-AAA schools the probabilities were 0.59 and 0.63, respectively.

The model poorly predicted the employment of women coaches in volleyball and soccer; not a single predictor in the soccer model had a

TABLE 4

Predictors of Division I Women’s Coaches in the Specified Sports

Predictor Basketball Volleyball Soccer Softball

Power conference schools 1.33 0.51 0.22 1.50 (0.67) n (0.60) (0.60) (0.71) n

Other DI-A schools �0.12 0.60 0.35 2.14 (0.48) (0.48) (0.49) (0.62) n

DI-AA schools �0.14 0.46 0.43 1.49 (0.38) (0.39) (0.40) (0.47) n

Women sports resources 0.20 �0.10 0.24 0.28 (0.20) (0.18) (0.18) (0.22)

Women sports share 0.15 0.17 0.12 0.51 (0.18) (0.18) (0.19) (0.22) n

Woman athletic director 1.08 1.04 �0.24 0.10 (0.58) (0.55) (0.55) (0.55)

Private secular school 0.08 �0.12 �0.18 1.20 (0.31) (0.29) (0.30) (0.35) n

Religious school �0.55 �0.22 �0.11 �0.11 (0.43) (0.44) (0.45) (0.50)

North �0.22 �0.92 �0.47 0.11 (0.29) (0.28) n (0.29) (0.34)

Constant 0.64 0.46 �0.53 �1.36 (0.40) (0.40) (0.41) (0.51 n)

N 327 311 300 268 Nagelkerke R2 0.13 0.08 0.03 0.22

npo0.05, one-tailed. NOTE: The entries are logistic regression coefficients, with standard errors in parentheses. The omitted categories are DI-AAA for divisions and public for type of institution. Women’s sports’ resources and share are factor scores based on the seven items described in the text and Appendix (obtainable from the authors). The estimates are logit coefficients, with standard errors in parentheses. Each equation is significant according to the Hosmer-Lemeshow test. The N reflect the number of schools having the sport.

1428 Social Science Quarterly

significant effect, and only region significantly predicted the sex of volleyball coaches. The overall fits of these models were 0.03 and 0.08, respectively. However, the model for softball coaches performed much better. In softball, the predicted probability of having a woman coach, with the values of the other predictors held constant as specified above, was 0.53 in power con- ference schools, 0.68 in other DI-A schools, and 0.52 in DI-AA schools, all far above the 0.20 probability in DI-AAA schools. The proportion of an institution’s athletic resources devoted to women’s sports also significantly differentiated schools with women softball coaches from other schools; tak- ing power conference schools as a case in point, the probability of having a woman coach jumped from 0.53 to 0.65 if the school scored one standard deviation above the mean, instead of at the mean, on the women’s relative athletic share dimension. Women softball coaches were also more likely to be found in private, nonparochial schools than elsewhere; at the power conference level, the probability of having a woman softball coach was 0.79 for private secular schools versus 0.53 for otherwise identical public schools.

Thus, these individual sport analyses largely confirmed that across sports, women coaches were more likely to be found in larger and more visible institutions and those that devoted more resources to women’s sports, but patterns unique to each sport were also evident.

Conclusion

Women’s sports play an increasingly important role in intercollegiate athletics in schools at every level of the NCAA pecking order. Identifying and retaining skilled coaches are keys to success in women’s as well as men’s sports. The employment of women as coaches is also of interest outside the arena of intercollegiate athletics because, to our knowledge, no other oc- cupation has made a transition from being primarily female to primarily male.

The tremendous growth of women’s athletics during the past 30 years has greatly expanded the need for full-time coaches, but the supply has not kept up with the demand. Clearly, the larger, richer institutions have hired a disproportionate share of the available women coaches. Women coaches are most likely to be found in institutions with more resources deployed to women’s athletics, proportionally and in real dollars, and in larger schools with more visible sports programs.

This set of findings is contrary to our expectations and to patterns of women’s employment in other fields. Women are less likely to be partners in law firms, and where they are, it is in smaller, less prestigious firms; they are less likely to rise to the CEO positions in business firms, and when they do it is in smaller companies, not Fortune 500 ones; and women disproportion- ately populate the lower rungs of the civil service, the lesser elected political offices, and the lower ranks of academia, to pinpoint just a few areas.

Women Coaches in Division I Women’s Sports 1429

Perhaps Reskin and Roos’s argument in Job Queues, Gender Queues (1990) might shed light on this counterintuitive set of findings. They point out that as women have gained footholds in many occupations, those oc- cupations have become resegregated within, as men reach the higher-status positions within job families. So, for example, more women are bakers than before, but men dominate the best paying, most commercially successful baking jobs, and women, as the newcomers, are relegated to the less lucrative and prestigious parts of the job class. However, in coaching, it is the men who are the newcomers and the women who, at least in several sports, are dominating the most visible jobs.

That perspective implies that there is a demand for women coaches, that in women’s sports women coaches are valued as much as or even more than male coaches, and that, accordingly, it is the largest, richest institutions that can afford to hire them. Schools in power conferences, those with relatively and absolutely bigger budgets, enjoy an obvious competitive advantage.

But there may be other reasons, too, for women’s disproportionate share in larger institutions. Reskin (2000, 2003) and Reskin and McBrier (2000) have pointed out that understanding occupational inequality means under- standing what goes on within organizations. We can speculate that these larger, richer institutions may themselves be more diverse and more con- cerned about increasing the share of women in visible leadership positions. They may be more open to lesbians, an issue in several women’s sports (Wilson, 2007; Redden, 2007). They also have the resources to provide more extensive support services for gender and racial diversity, such as affirmative action offices, multicultural support services, and more aggressive human resources activities. Our study does demonstrate substantial inter- institutional differences, though we have no information with which to measure diversity support systems, activities, and policies within each uni- versity.

Institutions may also be differentially responsive to the competing de- mands of raising a family and succeeding as a coach, although this is an issue that extends across most institutions (Wilson, 2007). Family-unfriendly workloads and extensive time commitments, along with unclear career paths, deter many potential women coaches who must try to balance work and family responsibilities (Drago et al., 2005; Everhart and Chelladurai, 1998; Wilson, 2007). The heavy workloads of coaches provide little time for outside responsibilities: coaching is considerably more than an eight-hour- per-day commitment (Drago et al., 2005). Tellingly, only 30 percent of full- time women coaches are married, compared to 55 percent of other women working full time, and only 18 percent of women coaches have children at home, compared to 45 percent of other women in the full-time workforce (Drago et al., 2005:5).10

10These data include all full-time coaches, not just those coaching at the college level.

1430 Social Science Quarterly

Undoubtedly due in part to the family-unfriendly nature of the job, women coaches are less likely than men to say they want to remain in coaching as they age (Sagas, Cunningham, and Ashley, 2000; Cunningham and Sagas, 2002). Female assistant coaches express less desire and determi- nation to become head coaches than do their male counterparts (e.g., Cun- ningham, Sagas, and Ashley, 2003; Knoppers et al., 1991; Sagas and Cunningham, 2004; Cunningham and Sagas, 2002, 2003); reflecting that hesitancy, they also apply for head coaching positions less often (Sagas, Cunningham, and Ashley, 2000). Moreover, many women assistant coaches believe that they will be discriminated against in getting and keeping a head coaching job (Cunningham and Sagas, 2002; Cunningham, Sagas, and Ashley, 2003; Sagas and Ashley, 2001).

Colleges and universities could make coaching more family-friendly, at least at the margins, by providing additional and different types of staff support, and as Drago and his colleagues have pointed out, the NCAA and coaches associations might help expand the pool by more clearly defining career ladders, offering programs to encourage and train potential coaches, and the like. Clearly, though, the deficit of women coaches is not a simple problem. To the extent that certain perceptions remain widespread among athletic administrators and potential women coaches alike, they do not bode well for increasing the proportion of women coaches.

Our findings have implications for the larger study of gender and oc- cupations because of the unique nature of the succession pattern in the coaching profession. Women’s coaching is a clear case of linkage between rising prestige and rewards, on the one hand, and the declining employment of women, on the other. As such, coaching merits consideration alongside the professions that students of gender-based occupational patterns have traditionally considered—professions that have displayed succession patterns opposite from those observable among women coaches.

REFERENCES

Acosta, R. V., and L. J. Carpenter. 1988. ‘‘Perceived Causes of Declining Representation of Women Leaders in Intercollegiate Sports—1988 Update.’’ Unpublished manuscript. New York: Brooklyn College.

Baron, James, and William Bielby. 1985. ‘‘Organizational Barriers to Gender Equality.’’ Pp. 233–51 in Alice Rossi, ed., Gender and the Life Course. New York: Aldine.

Bielby, Denise D., and William T. Bielby. 1988. ‘‘She Works Hard for the Money.’’ Amer- ican Journal of Sociology 93:1031–59.

Bielby, William T., and James Baron. 1984. ‘‘A Women’s Place is with Other Women: Sex Segregation Within Organizations.’’ Pp. 27–55 in Barbara Reskin, ed., Sex Segregation in the Workplace. Washington, DC: National Academy Press.

Britton, Dana. 2000. ‘‘The Epistemology of the Gendered Organization.’’ Gender and Society 14:418–34.

Women Coaches in Division I Women’s Sports 1431

Carpenter, Linda Jean, and R.Vivian Acosta. 2006. Women in Intercollegiate Sport: A Longitudinal, National Study. Twenty-Nine Year Update. Available at hhttp://www. womenssportsfoundation.org/cgi-bin/iowa/issues/part/article.html?record=1107i. Cialdini, R. B. et al. 1976. ‘‘Basking in Reflected Glory: Three (Football) Field Studies.’’ Journal of Personality and Social Psychology 34:366–75.

Cohen, Lisa, Joseph Broschak, and Heather Haveman. 1998. ‘‘And Then There Were More? Sex Composition and the Hiring/Promotion of Women Managers.’’ American Sociological Review 63(5):711–27.

Cunningham, George, and Michael Sagas. 2002. ‘‘The Differential Effects of Human Capital for Male and Female Division I Basketball Coaches.’’ Research Quarterly for Exercise and Sport 73(4):489–95.

———. 2003. ‘‘Occupational Turnover Intent Among Assistant Coaches of Women’s Teams: The Role of Organizational Work Experiences.’’ Sex Roles 49:185–90.

Cunningham, George, Michael Sagas, and Frank Ashley. 2003. ‘‘Coaching Self-Efficacy, Desire to Become a Head Coach, and Occupational Turnover Intent: Gender Differences Between NCAA Assistant Coaches of Women’s Teams.’’ International Journal of Sport Psy- chology 34:125–37.

Drago, Robert, Lynn Hennighausen, Jacqueline Rogers, Teresa Vescio, and Kai Dawn Stauffer. 2005. Final Report for CAGE: The Coaching and Gender Equity Project. Available at hLsir.la.psu.edu/workfam/CAGE.htmi. Everhart, C. Bonnie, and Packianathan Chelladurai. 1998. ‘‘Gender Differences in Prefer- ences for Coaching as an Occupation: The Role of Self-Efficacy, Valence, and Perceived Barriers.’’ Research Quarterly for Exercise and Sport 69(2):188–200.

Frankl, Daniel, and Donald G. Babbitt III. 1998. ‘‘Gender Bias: A Study of High School Track and Field Athletes’ Perceptions of Hypothetical Male and Female Head Coaches.’’ Journal of Sport Behavior 21:396–407.

Grappendorf, Heidi, and Nancy Lough. 2006. ‘‘An Endangered Species: Characteristics and Perspectives from Female NCAA Division I Athletic Directors of Both Separate and Merged Athletic Departments.’’ Sport Management and Related Topics Journal 2:6–20.

Hartmann, Heidi. 1981. ‘‘The Family as Locus of Gender, Class, and Political Struggle.’’ Signs 6:366–94.

Hartmann, Heidi, Robert Kraut, and Louise Tilly, eds. 1986. Computer Chips and Paper Clips: Technology and Women’s Employment, Vol. 1. Washington, DC: National Academy Press.

Hasbrook, Cynthia. 1988. ‘‘Female Coaches—Why the Declining Numbers and Percent- ages.’’ Journal of Physical Education, Recreation and Dance 58(2):59–63.

Heishman, M., L. Bunker, and R. Tutwiler. 1990. ‘‘The Decline of Women Leaders (Coaches and Athletic Directors) in Girls’ Interscholastic Sport Programs in Virginia from 1972 to 1987.’’ Research Quarterly for Exercise and Sport 61:103–07.

Kane, Mary Jo, and Jane Marie Stangl. 1991. ‘‘Employment Patterns of Female Coaches in Men’s Athletics: Tokenism and Marginalization as Reflections of Occupational Sex-Segre- gation.’’ Journal of Sport and Social Issues 15(1):21–41.

Knoppers, A. 1989. ‘‘Coaching: An Equal Opportunity Occupation?’’ Journal of Physical Education, Recreation, and Dance 60:38–43.

Knoppers, A., B. Meyer, M. Ewing, and L. Forest. 1991. ‘‘Opportunity and Work Behavior in College Coaching.’’ Journal of Sport and Social Issues 15:1–20.

1432 Social Science Quarterly

Marback, Tracy, Sandra Short, Martin Short, and Philip Sullivan. 2005. ‘‘Coaching Con- fidence: An Exploratory Investigation of Sources and Gender Differences.’’ Journal of Sport Behavior 28:18.

Okamoto, Dina, and Paula England. 1999. ‘‘Is There a Supply Side to Occupational Sex Segregation?’’ Sociological Perspectives 42:557–82.

Padavic, Irene, and Barbara Reskin. 2002. Women and Men at Work, 2nd ed. Thousand Oaks, CA: Pine Forge Press.

Polachek, Solomon W. 1981. ‘‘Occupational Self-Selection: A Human Capital Approach to Sex Differences in Occupational Structure.’’ Review of Economics and Statistics 63: 60–69.

Redden, Elizabeth. 2007. ‘‘ ‘Dirty Little Secrets’ in Women’s Sports.’’ Inside Higher Ed. Available at hinsidehighered.com/news/2007/04/30/sportsi. Reskin, Barbara. 2000. ‘‘Getting it Right: Sex and Inequality in Work Organizations.’’ Annual Review of Sociology 26:707–09.

———. 2003. ‘‘Including Mechanisms in Our Models of Ascriptive Inequality: 2002 Pres- idential Address.’’ American Sociological Review 68:1–21.

Reskin, Barbara, and Heidi Hartmann. 1986. Women’s Work, Men’s Work: Sex Segregation on the Job. Washington, DC: National Academy Press.

Reskin, Barbara, and Debra Branch McBrier. 2000. ‘‘Why Not Ascription? Organizations’ Employment of Male and Female Managers.’’ American Sociological Review 65:210–33.

Reskin, Barbara, and Irene Padavic. 1994. Women and Men at Work. Thousand Oaks, CA: Pine Forge Press.

Reskin, Barbara, and Patricia A. Roos. 1990. Job Queues, Gender Queues. Philadelophia, PA: Temple University Press.

Sagas, Michael, and Frank Ashley. 2001. ‘‘Gender Differences in the Intent to Leave Coaching: Testing the Role of Personal, External, and Work-Related Variables.’’ Interna- tional Journal of Sport Management 2:297–314.

Sagas, Michael, and George Cunningham. 2004. ‘‘Does Having ‘the Right Stuff’ Matter? Gender Differences in the Determinants of Career Success Among Intercollegiate Athletic Administrators.’’ Sex Roles 50:411–21.

Sagas, Michael, George Cunningham, and Frank Ashley. 2000. ‘‘Examining the Women’s Coaching Deficit Through the Perspective of Assistant Coaches.’’ International Journal of Sport Management 1:267–82.

Stahura, Kurt, and Michael Greenwood. 2002. ‘‘Sex of Head Coach as a Function of Sport Type Prestige and Institutional Structure.’’ Applied Research in Coaching and Athletics Annual 17:1–25.

Stangl, Jane Marie, and Mary Jo Kane. 1991. ‘‘Structural Variables that Offer Explanatory Power for the Underrepresentation of Women Coaches Since Title IX: The Case of Ho- mologous Reproduction.’’ Sociology of Sport Journal 8(1):47–60.

Strober, Myra, and Carolyn Arnold. 1987. ‘‘The Dynamics of Occupational Segregation Among Bank Tellers.’’ Pp. 107–48 in Clair Brown and Joseph Pechman, eds., Gender in the Workplace. Washington, DC: Brookings.

Sweet, K., and M. Malec. 1988. ‘‘Attitudes of Male Coaches of Female Basketball Players.’’ Paper presented at the World Congress of the International Association of Physical Edu- cation. Madrid, Spain.

Women Coaches in Division I Women’s Sports 1433

Theberge, Nancy. 1990. ‘‘Gender, Work, and Power: The Case of Women in Coaching.’’ Canadian Journal of Sociology/Cahiers canadiens de sociologie 15:59–75.

———. 1993. ‘‘The Construction of Gender in Sport: Women, Coaching, and the Nat- uralization of Difference.’’ Social Problems 40:301–13.

U.S. Department of Education, Office of Postsecondary Education. Various Years. Equity in Athletics Data. Available at hhttp://ope.ed.gov/athleticsi. Ward, Russell E. 2004. ‘‘Are Doors Being Opened for the ‘Ladies’ of College Sports? A Covariance Analysis.’’ Sex Roles 51:697–708.

Weinberg, R., M. Reveles, and A. Jackson. 1984. ‘‘Attitudes of Male and Female Athletes Towards Male and Female Coaches.’’ Journal of Sport Psychology 6:448–53.

Wharton, Amy S. 2000. ‘‘Feminism at Work.’’ Annals of the American Academy of Political and Social Science 571:167–82.

Whisenant, Warren. 2003. ‘‘How Women Have Fared as Interscholastic Athletic Admin- istrators Since the Passage of Title IX.’’ Sex Roles 49:179–84.

Whisenant, Warren, P. M. Pederson, and B. L. Obenour. 2002. ‘‘Success and Gender: Determining Advancement in Intercollegiate Athletic Directors.’’ Sex Roles 47:485–91.

Williams, Christine. 1992. ‘‘The Glass Escalator: Hidden Advantages for Men in the ‘Female’ Professions.’’ Social Problems 9:253–67.

Wilson, Robin. 2007. ‘‘Where Have All the Women Gone?’’ Chronicle of Higher Education May 4:A40ff.

Women’s Sports Foundation. n.d. Available at hwomenssportsfoundation.orgi.

1434 Social Science Quarterly