Black high school students' participation in school-sponsored sports activities: Effects on school engagement and achievement

Myths abound about the role of sports in schooling, particularly for Black student athletes. The literature affords little guidance on the specific relationship between participation in schoolsponsored sports and various outcomes for Black adolescents. Using a nationally representative sample, this study examines: (a) the effects of sports participation on various school engagement and student self evaluative variables, controlling for important background characteristics such as SES and gender; (b) the potential differential effects of sports participation for Black students; and (c) the degree to which sports participation affects Black students' academic achievement. Sports participation was found to improve the school engagement and academic self confidence of alI student athletes. Moreover, a positive intervening relationship was found between sports participation and academic achievement.

INTRODUCTION

As a key part of the social and cultural fabric of human life, sports permeate social realities at many levels, both globally and individually (Snyder & Spreitzer,1989). Few institutions in U.S. society are unaffected by sports (Edwards, 1973). Values reflected in athletic competition-striving for excellence, fair play, sportsmanship, hard work, and commitment to a goal-are inextricably linked to our nation's mainstream cultural values. Although much has been writter_ about the role of sports in society in a broad sense (Blanchard,1995; Lapchick,1986; Rees & Miracle,1986), the social function of professional and collegiate sports has often been the primary focus of this inquiry.

The research literature affords little guidance on the specific relationship between participation in high school athletics and school-related outcomes for adolescents. Even less is known about specific benefits for African American students. To complicate matters, it is not uncommon for research on high school athletics to be combined with examination of other extracurricular activities in attempts to investigate their effects upon schooling (Crain,1981; Holland & Andre,1987; Otto,1982). The assumption that most extracurricular activities affect students similarly lacks empirical evidence. Although past research on sports and schooling has found small, positive effects for participation in athletics on various student outcome measures such as grade point averages, methodological problems have marred the results of many of these studies (Trent & Braddock, 1992).

Much of the recent scholarly attention paid to the topic of sports and schooling has been theoretical and philosophical in its focus (Arnold,1997; Sadovskii & Sadovskii,1993).

These studies have attempted to portray sports as valued human practices that have many analogues in other areas of social life. Further, they often examine the heuristic and moral qualities of sport, and focus on the degree to which sports reaffirm and break down social norms. Arnold contends, for example, that, like most social conventions, sport is valueneutral-that is, it can be viewed as either good or bad. On the one hand, sports teach perseverance and encourage respect for rules, fair play and teamwork. Indeed, Sadovskii and Sadovskii, in their study of the connections between sports and mathematics, assert that athletics can have a beneficial effect on students' intellectual growth, states of mind, and will power. However, these authors contend that sports can also breed rugged individualism, vanity, intense rivalries, intolerance, and a "win-at-all-costs" attitude.

Purpose of the Study

In light of the dearth of empirical research exploring the effects of participation in high school sports upon educational outcome measures for adolescents, the present study sets out to address three aspects of this topic. The first stage of the analysis explores the effects of participation in high school sports on various school engagement and student self-evaluative factors while holding constant important background characteristics such as socioeconomic status (SES) and gender. Second, this analysis examines differences across race/ethnicity, explicating specific results for African American high school students. The third stage examines the extent to which sports participation affects student achievement during high school while controlling for both student background characteristics (SES and gender), along with the outcome variables analyzed in the first stage.

Research Questions

Three research questions guided this study:

(1) What is the impact of participation in school-sponsored team and individual sports on students' high school grade point average (GPA), general self-concept, and academic self-confidence?

(2) What are the specific relationships between sports participation and these variables for African American 10th-grade students relative to students from other racial/ ethnic groups?

(3) What are the effects of participation in sports upon student achievement, taking into account students' GPAs, self-concepts, and academic self-confidence-the dependent variables of the first stage of the analysis-as precursors to academic achievement?

CONCEPTUAL FRAMEWORK

Why should participation in athletics matter within the context of high school? The underlying assumption of the present research is that participation in sports can benefit adolescents in at least two ways: (a) by enhancing their interests and feelings of connectedness to their schools, and (b) by virtue of the intrinsic value of the athletic participation itself. Adolescents who are able to find structured activities within schools that capture their attention to the degree that they willing to spend voluntary time on campus interacting with school staff and their fellow students might be more likely to buy into the philosophy of schooling. Participation in extracurricular activities such as sports reinforces school norms and the goals of education. As a result, student athletes are often extrinsically motivated, as demonstrated by those students who strive to avoid course failure so that they can remain eligible to participate in extracurricular athletic activities. Schoolsponsored sports participation in this regard can be viewed as holding-power mechanisms because of their ability to keep students coming to school each day prepared to accept the normative codes governing their behavior and motivated to take part in their own education.

Within this conceptual framework the potential for enhancing extrinsic motivation includes not only participation in sports but also participation in band, yearbook club, school newspaper staff, and student government activities-indeed, any school-sponsored activity in which adolescents enjoy taking part. This is one reason why research on school sports is often combined with research on other extracurricular activities.

Previous research suggests that involvement in extracurricular activities such as sports increases students' overall interest and commitment to school as well as their engagement in more personal student-teacher contact, more positive attitudes about school, and greater parent-school contact (Grain, 1981; Trent & Braddock, 1992). According to Eidsmore (1964), high school athletes enjoy several advantages over non-athletes, including academic performance. Research also suggests the existence of a positive relationship between student participation in extracurricular activities and intergroup relations on school campuses. In a study of two hundred desegregated schools in the South, Grain found that in schools where students have high levels of participation in extracurricular activities, significantly less racial/ethnic conflict occurs than in schools where participation was minimal. Similarly, Slavin and Madden (1979) found that various school practices such as sports can facilitate positive racial/ethnic relations as well as positive intergroup attitudes and behaviors among students in northern and southern desegregated schools. Analyzing survey data from the Educational Testing Service (ETS), Slavin and Madden found that Black and White student athletes who were members of multiracial teams adapted better to diverse schooling environments as indicated by effects on six outcomes. Specifically, students who played together were significantly more likely to: (a) have three or more friends of another race/ethnicity, (b) have talked on the telephone with someone of a different race/ethnicity, (c) want to have more friends of different racial/ethnic groups, (d) have positive attitudes toward integration, (e) feel they were as smart as others of different racial/ethnic groups, and (f) feel comfortable around people of different racial/ethnic groups.

Several authors have addressed how sports participation and, more generally, participation in extracurricular activities can provide extrinsic rewards to students and help them form social bonds and relationships within school (Cram, 1981; Slavin & Madden, 1979; Trent & Braddock, 1992). However, in addition to this research, there is evidence that sports participation also has intrinsic value for students (Bandura, 1986; Kavussanu & McAuley, 1995). That is, physical exercise has advantages in its own right because athletes enjoy sports for the sake of sport. In this regard, participation in sports can be seen as an end in itself. In addition to the feelings of satisfaction that accompany athletic participation, building proficiency in athletics fosters a certain level of self-confidence in one's ability that can carry over into other aspects of students' lives, both at school and outside of school (Chickering,1969). Developing competence in an area is an essential part of becoming an adult. Moreover, there is evidence to suggest that a person's disposition and mental well-being are linked to the amount of physical activity in which he or she engages (Scheier & Carver, 1987). These aspects comprise the intrinsic value of sports because they can occur with or without accolades or rewards for the student athlete.

Sports participation has other intrinsic outcomes as well. According to Kavussanu and McAuley (1995), highly active individuals are significantly more optimistic overall and experience greater self-efficacy than do their non-active or low-active counterparts. This conclusion was based on their study of 188 men and women between the ages of 19 and 71 years old, drawn from a university and a local health club in a midwestern U.S. town. The study findings suggest that the positive feelings highly active individuals experience may be attributed to feelings of mastery that result from the accomplishment of an exercise workout-feelings that give individuals a sense of greater control over their environments and enhance their positive attitudes for the future. Again, these results are supported by previous research investigating the relationship between physical activity and psychological well-being that has yielded similar findings (Hamid, 1990; Scheier & Carver, 1989; Thayer, 1987).

Figure I presents a schematic model of the conceptual framework guiding this investigation. As depicted in the figure, students' participation in high school sports is presumed to affect three intermediary dependent variables: GPA, self-concept, and academic selfconfidence. These three variables in turn are presumed to affect student achievement. They also serve as proxy measures for students' self-evaluative engagement in school and for students' self-perception of their own ability and worth.

Recently, I examined the impact of sports participation on several additional student outcome variables, including the degree to which students work hard in school, come to class prepared to work, and perceive their life chances (]ordan, 1999). Like the present study, this preliminary research analyzed nationally representative data to address these questions. Although the results were mixed, they provided some evidence that participation in sports has a positive relationship with these outcomes. The dotted line in Figure I connecting students' participation in high school sports and academic achievement suggests that although a direct link may exist between these two factors, it is not as strong or consistent as the indirect impact of sports on GPA, self-concept, and academic selfconfidence. Additionally, various student background characteristics such as SES, gender, and prior achievement have been shown to mediate each of the above relationships.

MOTHODS

Data Source

Data Source

Data for this study were drawn from the National Educational Longitudinal Study of 1988 (NELS:88), a longitudinal survey sponsored by the U.S. Department of Education's National Center for Education Statistics (NCES) (1996). NELS:88 researchers surveyed 25,000 students along with their teachers, parents, and principals, spread among a thousand schools. The present study uses base-year (grade 8) NELS:88 data for student background characteristics and first follow-up (grade 10) data, approximately 15,000 weighted cases in all, to model the effects of athletic participation on the dependent variables. These data constitute an inclusive sample containing students from both public and private schools from all regions and community types within the United States.

Study Variables

Independent Variables. The independent variables examined in the present study, participation in team sports and participation in individual sports, were a function of each sampled student's membership on a school-sponsored athletic team, coupled with the number of hours the student spent on extracurricular activities.' These variables were measured as a combination of a simple dummy variable that took a value of one if a student participated in a given sport, and zero if she or he did not participate. Participation in each sport was then multiplied by the number of hours each student spent engaged in sports. This combination enabled comparisons to be made between athletes and nonathletes relative to a sliding scale of the amount of time (from zero to 20 or more hours per week) students in each group spent participating in sports. Thus, the independent variables can be viewed as measures of students' levels or depth of involvement in schoolsponsored athletics. They were further used to detect differences in the levels of student commitment to athletics in ways that dichotomous measures cannot detect.

Studies comparing student athletes to non-athletes using dummy variables run the risk of treating adolescents who are only casually interested in sports, and who participate in them infrequently, the same as those who are very committed to sports activities. For instance, students who participate on a school-sponsored athletic team for one hour per week during the spring term may be lumped in with student athletes who are avid athletes and who spend more than 10 hours per week on team activities throughout the school year. The problem with such analysis is that, compared to avid players, students participating less often may appear more like non-athletes on some measures and may be qualitatively different as well. For this reason, simple dichotomous measures of participation versus nonparticipation in sports are imprecise at best. To address this limitation, the present study examined sports participation along the categorical continuum just described. A student who participated in no school-sponsored sports was assigned a value of zero (0). The remaining designations were as follows: less than one hour per week (assigned value of 1); 1 to 4 hours per week (assigned value of 2); 5 to 9 hours per week (assigned value of 3); 14 to 19 hours (assigned value of 4); and 20 or more hours per week (assigned value of 5). In this way, this study represents both a departure from and an enhancement of previous research on high school sports participation that primarily has examined similarities and differences between student athletes and non-athletes.

Control Variables. Several student background characteristics were used as control variables throughout the analysis. All background variables were measured at grade 8, the NELS:88 base year. These variables included SES, gender, risk factors for dropping out of school (including single-parent-headed household, limited parental education, having siblings who dropped out of school, excessive number of hours spent home alone, limited English proficiency, and low family income), prior GPA (grade 8), standardized reading achievement, preparedness for class, student self-concept, and whether or not the student believed he or she was seen as athletic in grade 8. SES, risk factors, prior self concept, and prior GPA are composite measures constructed by NCES researchers. Student gender, grade 8 reading scores, and "seen as athletic by others" are single-item measures drawn directly from the base-year NELS:88 sample. "Preparedness for class" was a scale constructed to measure the degree to which NELS:88 students came to class ready to learn during the 8th grade. This three-item scale asked: "How often do you come to class and find yourself WITHOUT these things?" and presented the following responses: "(a) pencil or paper (when needed); (b) books (when needed); and (c) your homework done (when assigned)." Students' responses were coded on a four-point, Likert-type scale from 1 (usually) to 4 (never).

Dependent Variables. Student achievement was the primary dependent variable examined in this study. It was measured as the mean of four standardized NELS:88 cognitive tests taken during grade 10. The tests covered the following four core subject areas: reading, mathematics, history, and science. Rather than examining the direct effects of sports participation on student achievement, these effects were first examined on three dependent/mediating variables: (a) student GPA, (b) student self-concept, and (c) student academic self-confidence. For the purposes of this study, these three variables served equally as both intervening and dependent variables.

Similar to the independent variables, the dependent variables were grade 10 items. GPA was the initial dependent variable presented in the study. It was measured by averaging students' self-reported grades in mathematics, English, social studies, and science; and then calibrating this average to fit a conventional four-point scale (e.g., A = 4.0, B = 3.0, and so forth). Students' self-concept was derived from an eight-item composite scale created by NCES researchers. This scale includes items such as "I feel good about myself," "I feel I am a person of worth, equal to other people," and "I am able to do things as well as most other people."

Students' academic self-confidence was derived from an eight-item scale specifically constructed for the study to measure the degree to which students feel confident and competent in their ability to handle fundamental high school courses, namely English and mathematics. The scale had a reliability coefficient (Cronbach's alpha) of .79. It included items such as "I learn things quickly in English classes," "Mathematics is one of my best subjects," and "I have always done well in mathematics."

Data Analysis

Cross-tabulations and ordinary least squares (OLS) regression techniques were used to analyze the effects of participation in team sports and individual sports on student outcomes. Cross-tabulations were used to illustrate zero-order differences in the dependent and independent variables across students' race/ethnicity. The first set of regression equations modeled the impact of sports participation on students' GPA, general selfconcept, and academic self-confidence. Another set of equations modeled the impact of sports participation on student achievement at grade 10, and shed light on the net impact of the initial dependent variables and students' background characteristics.

RESULTS

One of the more striking findings of this study is that an overwhelming percentage of students do not participate in either team or individual sports (see Table I). Table I presents the cross-tabulations of weighted frequencies for the independent variables and background characteristics of the NELS:88 sample across students' race/ethnicity. The sports participation variables are 10th-grade measures, while the student background variables are base-year measures (grade 8). As the table shows, the percentage of males and females was roughly even. The cross-tabulations also reveal that sports participation was strikingly low across the board for this NELS:88 sample. Indeed, the overwhelming majority of students participated in no school-sponsored sports at all. Taken together, about 21% of all 10th-grade students participated in team sports; 15% were involved in individual sports.

Of the African American students in the NELS:88 sample, 82% reported that they had not participated in any team sports by 10th grade, while nearly 90% reported not participating in individual sports. Modest racial/ethnic differences were noted in sports participation. For both team and individual sports, White students' rates of participation were slightly higher than those of Black students, followed by Hispanic students, who rated last in this category.z Conversely, White students overwhelmingly indicated that they were seen by others as being less athletically inclined than Black or Hispanic students.

This first set of analyses reveals that participation in both individual and team sports appears to have a small, but consistent, positive impact upon students' GPA, self-concept, and academic self-confidence. The effects for African American students were not fundamentally different from those noted for students of other racial/ethnic groups. The pattern of results was stable across race/ethnicity when controlling for SES, prior achievement, and other potentially important social-class factors. Racial/ethnic differences in the relationship between sports participation and student outcomes were negligible and in some cases found not to exist.

Table II depicts regression results for the effects of participation in sports on GPA. It shows that, among all students, sports participation in both team and individual sports had a significant positive relationship to students' grades ((3 = .05, t = 7.14 and P _ .06, t=8.93 for team and individual sports, respectively; p

Table III shows the regression results for the effects of participation in sports on students' self-concept, controlling for student background characteristics. The best predictor of general self concept at grade 10 was prior self-concept measured at grade 8 (a=.45, t=58.34). As in the results discussed above, the effect of sports participation on self-concept was also small, positive, and significant. Among all students, the results were positive and significant (pi = .03, t = 3.84 and (3 = .03, t = 3.04 for team and individual sports, respectively; p

Table IV depicts the effects of sports participation on students' academic self-confidence. Comparing tables III and IV, it is apparent that the pattern of sports participation effects on general self-concept was similar to that of sports participation on academic selfconfidence. However, among student background characteristics, one notable difference was that 8th-grade GPA was found not to predict academic self-confidence as well as it predicted self-concept. Among the background controls, GPA, prior achievement on the grade 8 reading test, and general self concept were important predictors of academic self confidence. For all students, the coefficient for 8th-grade GPA was three times larger than any other variable in the equation ((beta) = .32, t = 35.53). Relative to the magnitude of other effects examined in this analysis, the standardized coefficients for sports participation depicted in Table IV were slightly larger. For all students, the following effects of sports participation were noted: for team sports ((beta) = .05, t = 5.79) and for individual sports ((3 = .03, t = 4.12). The results for European American students closely resembled those for all students; however, for African American students, the results were strikingly dissimilar. That is, as depicted in the first two columns of this table, African American students who participated in team sports rated their academic self-confidence significantly higher than did African American non-athletes ( (beta) = .07, t = 2.51). The regression results for Hispanic American students were also not significant but positive for both individual and team sport participation.

Finally, Table V presents the regression results for the effects of sports participationalong with GPA, self concept, and academic self-confidence-on grade 10 test scores across racial/ethnic groups, while controlling for student background characteristics. This analysis revealed a small but consistently positive effect of sports participation on academic achievement when the other variables in the equation were held constant. As shown in this table, the pattern of the effects of sports participation varied little across the different racial/ethnic groups. For all students, the regression coefficients for team and individual sports participation were (beta)3 = .02, t = 3.93 and (beta) = .03, t = 6.59, respectively (p

Several other interesting findings should be noted. First, as shown in Table V, among the student background controls, base-year risk factors revealing overall negative effects on achievement ((beta)=-.04, t=-7.03) were most negative for Hispanic American students ((beta)+-07, t=-3.83). The effects of these risk factors for African American students were also negative but not as strong, and this variable had no significant negative effect on achievement for European American students. Interesting gender differences were also noted. Though, among all students, males ((beta) = .13, t = 26.83) outperformed females on standardized tests, the gender gap was not as great among African American students. The regression coefficients for African American males (p3 = .07, t = 3.98) were nearly half the magnitude of the results for all students ( (beta) = .13, t = 26.83). The largest gender gap in achievement, even larger than that for all students, was found for White students ((3 =.15, t=24.97). Additionally, some differences between racial/ethnic groups were also evident within the intervening block of variables. For example, though 10th-grade GPA was a positive predictor of achievement for all students ((beta) = .09, t =14.02), its effects were weakest for Black students ((beta) = .01, t = 0.44) and strongest for White students ((beta) = .12, t =15.01).

As shown in Table V, the effects of sports participation on student achievement (as measured by composite grade 10 test scores) for European American students were almost identical to those for the full sample. The standardized regression coefficients ((beta)s) were the same, but due to the smaller sample size, the t-values were slightly lower for White students. For African American students, the size of the coefficients were similar; however, the results were significant only for team sports ((beta)=.04, t=2.19; p

DISCUSSION

Sports and physical activity can help adolescents who participate in them develop self-efficacy, self-confidence, and feelings of competence by virtue of the mastery of skills and talents that such participation engenders. Like adults, adolescents' sense of pride, along with their self concept, is connected to the degree to which they feel competent performing certain actions. As Chickening (1996) asserts,

. . . experiences encountered in athletics provoke reactions sharply relevant to the development of competence and sense of competence, and to the development of increased ability to manage them productively. Athletics offers a context in which concrete, unequivocal, and public performance provides clear evidence of developmental progress. (p. 29)

The extent to which sports participation bolsters the schooling experiences of student athletes and positively affects student learning is an important question for educational policy, practice, and research. As discussed at the outset, much of the research has addressed the role of athletics in education and in within the broader society, but the extant literature affords little guidance for those seeking a better understanding of the specific relationship between adolescents' participation in school-sponsored sports and various outcomes. Additionally, much of the previous research on sports participation and schooling is limited, primarily as a result of the voluntary nature of extracurricular activities and the inability to eliminate or control for self-selection biases. Analysis of differential effects of sports participation across racial/ethnic groups is also lacking.

To address these gaps in the literature, this study investigated three aspects of the topic: (a) the effects of sport participation on student GPA, self-concept, and academic self-confidence, while holding constant important background characteristics such as SES and gender; (b) racial/ethnic differences, particularly whether or not African American students benefit more from sports participation than do students of other racial/ethnic groups; and (c) the degree to which sports participation affects student achievement, beyond its impact on GPA, self-concept, and academic self confidence. The empirical evidence obtained from the study suggests that participation in sports, regardless of various student background characteristics, offers several benefits to adolescent students.

The finding that most high school students do not participate in school-sponsored sports is perhaps one of the most striking revelations of this study. About 21% of all lOthgrade students were found to participate in team sports of some kind while only 15% were involved in individual sports. For African American students,18% reported having participated in team sports by 10th grade, with nearly 10% having participated in individual sports. This suggests that whatever positive outcomes might have accrued from sports participation, few students realized those benefits. Another major finding was that participation in high school sports had a positive relationship to GPA, self-concept, and academic self-confidence for all students. However, these variables were examined as intermediate dependent variables or precursors, which were affected by sports participation and in turn affected student achievement.

The results of this study provide evidence that participation in high school sports is associated not only with higher grades, better self concept, and greater academic selfconfidence but also with higher academic achievement. Across each racial/edu-tic group analyzed in this study, the effects of sports participation were positive. Among 10th-grade students in particular, the effects were significant at p

The implications of the above findings are twofold. First, although high school reform has rightly focused on improving the social organization of schools and enhancing curricular development (McPartland, Balfanz, Jordan, & Legters,1998), ongoing and future reform efforts should also take into account the importance of extracurricular activities such as sports. The results of this study support the position that participation in sports can increase students' interest in school and improve their achievement. In light of these findings, educational policymakers and practitioners need to consider both improving the quality of high school athletic programs and seeking creative ways to increase the number of boys and girls who participate in such programs. Coleman (1961) cautions, however, that the resources and time schools place on athletics are resources taken away from bolstering educational programs. To that end, it is important that the potential benefits of developing better athletics departments be weighed against the more fundamental goal of educating students for productive lives as adults. Given the current state of educational reform, there appear to be sufficient opportunities for improving the quality of teaching and learning as well as those aspects of schools that enrich students' lives in other ways.

Second, like extracurricular activities in general, sports participation was found to increase high school students' personal investments in education by providing them with additional opportunities for interaction with adults. Within schools, for example, teachers, school administrators and staff, educational paraprofessionals, parent volunteers, and other adults who serve as coaches and related sports personnel often encourage students not only to excel in their extracurricular activities but also to focus on their total educational experience. Thus, sports participation often means exposing student athletes to a supportive environment that enriches both their love for their respective sports as well as their interest in school. Ultimately, this may help reduce the degree of alienation students sometimes experience while attending comprehensive high schools, which are oftentimes large and impersonal.

Limitations of the Study

At least two methodological limitations of the present study must be explained: (a) problems associated with analyzing grades, and (b) the absence of contextual variables in the analysis. Regarding the first of these, many high schools across the country have adopted some form of "no-pass/no-play" rule for student athletes. According to such rules, students whose grades fall below a specified level may be barred from participation in sports (and sometimes any other extracurricular activities). Such students may have been characterized as non-athletes in the NELS:88 sample. Additionally, various researchers have pointed out that student athletes are sometimes steered away from challenging courses, in which they are likely to earn poor grades, and into easy classes that help "pad" their GPAs (Holland & Andre,1987). Furthermore, it is conceivable that high school coaches sometimes pressure other teachers to give their players good grades (Jordan, 1999). Taken together, these conditions create a different set of circumstances for academic achievement for student athletes and non-athletes, to the greater benefit of the former.

As a result, the distribution of grades for student athletes is somewhat constrained to a range between A and C, while that for non-athletes covers the full distribution from A to F. Thus, in comparison to all students in a given school, the average grades for student athletes are almost always higher.

Prior research suggests that students who attend larger high schools have fewer opportunities to participate in extracurricular activities such as varsity sports, unless the larger school has sufficient human and financial resources to facilitate broader participation (Lindsay,1984). It is conceivable that contextual factors such as school size, poverty level, student racial/ethnic composition, regional and community location and sector each play a role in the relationship between sports participation and student achievement. These factors, however, were not considered in this analysis. Though the primary purpose of the present study was to investigate student-level predictors, future research must take the effects of contextual variables into account. For example, it only makes sense that school size mediates student participation in sports. In large schools where many talented students vie for a few spots on athletic teams, the chances of participation are most likely to be less than in smaller schools with fewer students and equivalent numbers of positions on sports teams. Notwithstanding, a school's financial and human resources can affect the availability of opportunities to participate in sports and other extracurricular activities. Large schools having greater resources can address the problem of inadequate opportunities for participation in various ways such as expanding their athletic departments to maintain many different sports teams and/or enhancing their intramural sports programs.

The advantages of possessing resources extend beyond athletics, however. Schools having more resources can provide greater opportunities for their students than those schools with fewer resources, not only in terms of access to sports but also regarding the quality of instruction. Impoverished high schools serving African American and Hispanic American students tend to offer fewer advanced placement/college preparatory courses than do their predominantly European American suburban counterparts. They also often have weaker music and arts programs, student governments, and athletic programs. Moreover, the more exclusive sports such as crew, fencing, golf, and tennis are seldom available for students attending poor, urban comprehensive high schools. Although contextual factors such as school size and the availability of school resources are potentially important explanatory variables, these too were beyond the scope of the present study.

CONCLUSION

Sports and schooling are inextricably linked. This study bears out the existence of a positive relationship between students' participation in high school sports and various student outcomes, including academic achievement, that is small but consistent. However, as is often the case in educational research, the findings generate as many questions as they answer. The present study was affected by methodological limitations similar to those of other studies that analyze programs or activities voluntary in nature. Moreover, in focusing on student-level variables, this study ignored potentially important contextual factors that might have mediated between sports participation and student outcomes within high schools. Further research is needed to more fully address issues of selfselection biases and also to analyze contextual variables such as school size, student poverty levels, financial resources, and staffing. Knowledge about these issues can be informed by further study that takes into account the specific conditions under which students are likely to benefit from participation in sports as well as studies examining the impact of sports on the psychological orientation of adolescents.

Notwithstanding these limitations, this study offers several important contributions to the research literature. It provides empirical evidence of a relationship between sports participation in certain school-related outcomes such as GPA and students' academic self confidence. It also shows that Black, White, and Hispanic American adolescents, with minor variations, appear to benefit equally from sports participation. Lastly, it provides evidence of a link between sports participation and academic achievement. The challenge for educators, policymakers, and researchers is to apply this knowledge to educational reform practices that enhance the schooling experiences of all adolescents.

'Unfortunately, within the NELS:88 database, popular high school sports such as track and field, volleyball, and tennis are not individually listed; however, these sports are captured within broader, nonspecific categories. For example, the team sports participation variable incorporated students' participation in baseball, basketball, football, and/or soccer, plus an inclusive category called "other team sport." Likewise, the individual sports participation variable included swimming and the category, "individual sports."

2Results for Asian American and American Indian students and students from other racial/ethnic groups were not delineated in the NEIS:88 data, but these results are nested in the category called "all students."

REFERENCES

Arnold, P. J. (199. Sport, ethics and education. London: Cassell.

Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory. Englewood Cliffs, NJ: Prentice-Hall.

Blanchard, K. (1995). The anthropology of sport: An introduction. Westport, CT: Bergin & Garvey. Chickering, A. W. (1969). Education and identity. San Francisco: Jossey-Bass.

Coleman, J. S. (1961). The adolescent society. New York: The Free Press.

Crain, R. L. (1981). Making desegregation work: Extracurricular activities. Urban Review, 13, 121-126.

Edwards, H. (1973). Sociology of sport. Homewood, IL: Dorsey.

Eidsmore, R. M. (1964). High school athletes are brighter. Journal of Health, Physical Education and Recreation, 35, 53-54.

Hamid, P. N. (1990). Optimism and the reporting of flu episodes. Social Behavior and Personality, 28, 225-234.

Holland, A., & Andre, T. (1987). Participation in extracurricular activities in secondary school: What is known, what needs to be known? Review of Educational Research, 57(4), 437-466.

Jordan, W. J. (1999). Sports and schooling: Athletics matter in comprehensive high schools. Paper presented at the annual meeting of the American Educational Research Association, Montreal, Canada.

Kavussanu, M., & McAuley, E. (1995). Exercise and optimism: Are highly active individuals more optimistic? Journal of Sport and Exercise Psychology, 17, 246-248.

Lapchick, R. E. (1986). Fractured focus: Sport as a reflection of society. Lexington, MA: Lexington Books.

Lindsay, P. (1984). High school size, participation in activities, and young adult social participation: Some enduring effects on schooling. Educational Evaluation and Policy Analysis, 6(1), 73-83.

McPartland, J., Balfanz, J., Jordan, W. J., & Legters, N. E. (1998). Improving climate and achievement in a troubled urban high school through the talent development model. Journal on the Education of Students Placed At Risk, 3(4), 337-361.

National Center for Education Statistics. (1996). NELS:88/94 public use data files and electronic codebook (CD-ROM). Chicago: National Opinion Research Center at The University of Chicago. Otto, L. B. (1982). Extracurricular activities. In H. J. Walberg (Ed.), Improving educational

standards and productivity (pp. 217-227). Berkeley, CA: McCutchan.

Reel, C. R., & Miracle, A.W. (1986). Sport and social theory. Champaign, IL: Human Kinetics. Sadovskii, L. E., & Sadovskii, A. L. (1993). Mathematics and sports. Providence, RI: American Mathematical Society.

Scheier, M. F., & Carver, C. S. (1987). Dispositional optimism and physical well-being: The influence of generalized outcome expectancies on health. Journal of Personality, 55, 169-210. Slavin, R. E., & Madden, N. A. (1979). School practices that improve race relations. American Educational Research Journal, 16(2), 169-180.

Snyder, E. E., & Spreitzer, E. A. (1989). Social aspects of sport (3rd ed.). Englewood Cliffs, NJ: Prentice-Hall.

Thayer, R E. (198. Problem perception, optimism, and related states as a function of time of day (diurnal rhythm) and moderate exercise: Two arousal systems in interaction. Motivation and Emotion, II, 19-36.

Trent, W. T., & Braddock, J. H., 111. (1992). Extracurricular activities in secondary schools. In A C. Alkin, Encyclopedia of educational research (Vol. 2, 6th ed.; pp. 476-481). New York: Macmillan.

Will j. Jordan, The johns Hopkins University

Copyright Howard University Winter 1999
Provided by ProQuest Information and Learning Company. All rights Reserved