Printer Friendly

College graduation rates of student athletes and students attending college male basketball games: a case study.

The relationship between college sport involvement as either an athlete or a "fan" and the 6-year graduation rate was studied for 3,145 1983 Ball State University freshmen. Eighty-nine of the students were identified as varsity athletes, 564 as having attended a sample of two male basketball games, and 2,492 as neither being an athlete nor having attended the sample of games. Gender, race, local residence, location of hometown, academic major, college academic preparation, attained grade point average, and Myers-Briggs Type Inventory characteristics were used as control variables in analyses of the relationship between sport involvement and graduation rate. Tinto's contention that "social integration" is an important factor in retention of college students served as the basis for predicting that the graduation rates would be higher for athletes and "fans" than for "other" students and this was confirmed. While the sampling of basketball games was limited to many athletes had advantages not provided for most "other" students, the results were consistent with the predictions of Tinto's attrition model.

The present study examined the 6-year graduation rate of students from the 1983 Ball State University (BSU) freshman class who were identified as being a BSU varsity athlete, having attended a sample of BSU male basketball games (i.e., "direct sport consumers;" McPherson, 1975), or neither being a BSU athlete nor having attended a game. The study was a follow-up of two prior studies conducted at BSU (Schurr, Ruble & Ellen, 1985; Schurr, Wittig, & Ruble, 1988) that examined factors associated with college student attendance at male basketball games. Results of the prior BSU studies supported the theory of Edwards (1973) about characteristics of "sports fans" and extended his theory to include social factors and additional personal characteristics. Edwards proposed that fans have high interest in sports because they identify with characteristics of sport competition and participants; in particular, businessmen have high interest because " competition is a microcosm of their everyday secular lives" and blacks have high interest because of the special significance of sports in black society.

The BSU studies did find disproportionately high attendance at the basketball games for students having academic majors in business fields and for blacks. In addition, the studies linked game attendance to gender and a preference for action-oriented activities, with high attendance for males and a group identified as "sensing" types by the Myers-Briggs Type Indicator (MBTI; Myers & McCaulley, 1985). Most relevant for purposes of this study were results indicating that few students attended games alone and attendance was relatively high for students who lived in residence halls and/or were from hometowns located farther distances from the university. These findings suggested that students attending the games had developed a stronger identification with the university and were more involved in its social system than students not attending the games. Generally, results reported in the BSU studies were consistent with results of other studies that included the same types of variables (e.g., Doyle, Lewis, & Malmisur, 1980; Lowe & Harrold, 1972; McPherson, 1975).

The possible implication of the BSU studies with respect to graduation is found in the three general facets of Tinto's (1975) well-known college attrition model: (a) students enter college with different levels of academic preparation and other attributes; (b) they develop different levels of integration into an institution's academic social system, including the attainment of different grade point averages (GPA) and the development of different attitudes about academic programs; and (c) they develop different levels of integration into an institution's social system, including the establishment of different levels of interaction with peers through formal, semi-formal, and informal groups and with adult members of the academic community. Tinto proposed that if academic and social integration are positive, commitment and motivation to persist in attaining a degree are enhanced, and high levels of either type of integration might offset low levels of the other type for determining persistence. Thus, according to the model, it seemed tenable that the graduation rate for students identified as direct sport consumers or varsity athletes might be higher than the rate for other students because sport consumers ostensibly had developed higher levels of social integration by identifying with the team, while athletes have relatively high levels of social integration implicit in their ongoing contact with other athletes, coaches, and university personnel.

Although no prior research was identified that examined the graduation rate of undergraduate student direct sport consumers, several studies compared the academic achievement and graduation rates of college athletes and nonathletes (e.g., Brede & Camp, 1987; Henschen & Fry, 1984; Jacobs, 1985; Kleiber & Malik, 1989; Purdy, Eitzen & Hufnagel, 1985; Shapiro, 1984; Snyder, 1985; Weber & Sherman, 1987). Brede and Camp reported that results of such studies were inconsistent and "...failed to show unequivocally that college athletes differ in any important way from other college students in terms of their |college academic achievement~...". A possible explanation for the inconsistent results was suggested by Snyder and Spreitzer (1983) when they observed that studies such as those reviewed by Brede and Camp used only a single or a few institutions; consequently, findings from a study conducted at a particular institution might be meaningful only for that institution or institutions having similar institutional characteristics. However, a recent United States General Accounting Office (USGAO, 1989) study of NCAA Division I schools, which was not limited in scope and did not suffer from certain methodological problems associated with earlier attempts at a comprehensive study, reported that the 5-year graduation rate was higher for college athletes (56%) than for nonathletes (47%). The USGAO report also listed types of variables that should be controlled for comparisons of athlete and nonathlete graduate rates.



There were 3,145 BSU students included in the study: 89 male and female varsity athletes, 564 students who attended one or both of two BSU basketball games used to sample student attendance at games, and 2,492 students who were neither varsity athletes nor attended the games. One game used to sample student attendance was played during the first year students were enrolled at BSU; the second game was played during the following year. Attendance was determined by recording student identification numbers as students obtained reserved seating tickets preceding the games; all students attending games had to obtain such tickets. The necessity of showing identification upon arrival at the game assured accuracy of the sample.


The principal method used for analyzing the data was the multiple regression statistical procedure (Cohen & Cohen, 1983). Thirteen variables were used in the analyses: (a) a "graduation status" variable indicating if students had or had not graduated six years after entering BSU; (b) a "sport involvement" variable that identified if students were athletes (Athletes), had attended a game (Fans), or neither were athletes nor had attended a game (Other Students); and (c) 11 control variables.

The 11 control variables were identified in other studies conducted at BSU as being related to attendance at basketball games and/or attrition and, except for family income, included the types of control variables listed in USGAO report (1989). The control variables were: (a) living or not living in a residence hall during the term in which games were played; (b) gender; (c) race; (d) 22 groupings of academic majors based on groupings used by the College Entrance Examination Board (1987); (e) three groupings of student hometowns based on distances between the hometowns and the university; (f) the four personality characteristics measured by the MBTI Extraversion-Introversion, Sensing-Intuitive, Thinking-Feeling, and Judging-Perception scales; (g) a predicted first-year GPA (predicted GPA) based on Scholastic Aptitude Test (SAT) scores and high school academic achievement; and (h) the cumulative GPA attained (college GPA) while enrolled at BSU.

The interpretations for results of the study were based on statistics yielded by two regression models that specified graduation status as the criterion variable. The statistics were "effect parameters" and "effect size" (ES) values (Cohen & Cohen, 1983). The effect parameters identified relative differences in graduation rates for Athletes, Fans, and Other Students after relationships for other predictor variables had been controlled. The ES values identified magnitudes of relationships between the sport involvement and graduation status variables after other predictor variables were controlled; that is, the values were partial ES values. Cohen (1988) evaluates the magnitudes of partial ES values as small if around .14, medium-sized if around .36, and large if around .51. However, Cohen also indicates that magnitudes of relationships between predictor and criterion variables should be evaluated in terms of magnitudes of relationships for other predictor variables.

The two regression models employed in the study differed: in the first, the sport involvement variable and all control variables except for college GPA were specified as predictor variables, while in the second, college GPA was also specified as a predictor variable. College GPA is clearly recognized as being the most important factor associated with student graduation/attrition, partly because a student cannot continue pursuing a degree unless the GPA exceeds a specified value. Although relationships have been found between graduation/attrition and a number of variables, such relationships typically varied from study to study and vanished or were reduced considerably if the variables were analyzed along with the dominant influence of GPA; that is, when GPA was used as a control variable. It is commonly thought, however, that students are more likely to graduate (for residential type institutions) if they had at some time lived in residence halls (see for reviews of attrition studies, Bean & Metzner, 1985; Pantages & Creedon, 1978; Tinto, 1975). The SPSSX MANOVA statistical program (SPSS, 1988) was used to perform the analyses.

Results and Discussion


The analyses of the first regression model that specified the sport involvement variable and all control variables except for college GPA as predictors yielded a statistically significant relationship between sport involvement and graduation rate (F(2,3099) = 18.18, p|is less than~.001). The ES value for the sport involvement variable was .13, which was a small ES value according to Cohen's (1988) guidelines. This value, however, was among the four largest ES values obtained in the analyses, lower than ES values for predicted GPA (ES = .24) and academic major (ES = .26), but equivalent to the ES value for the residence hall variable (ES = .13).

The analyses of the second regression model that specified the sport involvement variable and all control variables as predictors yielded a statistically significant relationship between sport involvement and graduation rate (F(2,3098) = 12.47, p|is less than~.001). The ES value for the sport involvement variable was .09, again in the range of small ES size values. This value, however, was among the four largest ES values obtained in the analyses, lower than the ES values for college GPA (ES = .54) and academic major (ES = .21), equivalent to the ES value for predicted GPA (ES = .10), and higher than the ES value for the residence hall variable (ES = .05).


The statistics required for interpreting the results are shown in Table 1. The effect parameters identify differences in the relative graduation rates of Athletes, Fans, and Other Students after controlling for other variables. The effect parameters yielded by the first model which specified all control variables except college GPA as predictors indicated that the graduation rates for Athletes (b = .05) and Fans (b = .04) were about .14 higher than the rate for Other Students (b = -.09). The effect parameters yielded by the second model which specified all control variables as predictors indicated that the graduation rate for Fans (b = .05) was about .10 higher than the rate for Other Students (b = -.05) and that the graduation rate for Athletes (b = .00) was about .05 higher than the rate for Other Students.

Essentially, the data shown in Table 1 indicate that students identified as Athletes, Fans, and Other Students entered college with equivalent levels of academic preparation for college; that is, high school achievement and SAT scores yielding about the same predicted first-year GPA means (respectively, M = 2.39, M = 2.37, and M = 2.40). The 6-year-graduation rates, however, were higher for Athletes (P = .62) and Fans (P = .64) than for Other Students (P = .48). Moreover, the relative graduation rates for Athletes and Fans were still higher than the rate for Other Students after academic preparation for college, academic achievement in college, and variables indicating a variety of personal characteristics and college experiences were controlled.


Although a limitation for the study was the few number of basketball games used to sample student attendance behavior, the results were consistent with Tinto's (1975) contention about the role of social integration in student decisions to persist in attaining a college degree. Furthermore, they were in line with writers who have posited that sports play a role in the social integration of people into society (see for a brief overview, McPherson, 1975). This is not to suggest, however, that the particular results found for students identified in this study as direct sport consumers would be found at all types of institutions, as the role that sport plays in the social systems of institutions varies from one to another. BSU is a mostly 4-year residential university that draws over 90% of its undergraduate students from a state with a very strong basketball tradition. Many BSU students are likely to identify with the basketball program, and attending basketball games is likely to reflect behavior associated with the sensing, attaining, or expressing integration into the social system.

The media often center on selected cases that portray a negative image of college athletes and sports programs (e.g., Asher, 1986; Hersch, 1989). The results of this study, however, agreed with findings of the 1989 USGAO study, indicating that college sports programs can coexist with and possibly contribute positively to college academic programs. It is true that many athletes in the study received financial and special academic support not received by typical students, but this does not rule out the possible contribution that sport involvement made to decisions of athletes to persist in seeking a degree. Students who were involved in sports via direct sport consumption had a graduation rate equivalent to the rate of athletes, but typically did not receive the special financial and academic assistance provided for athletes.

At very least, the results indicated that the BSU sports program served a disproportionately high percentage of 1983 BSU freshmen who would attain a degree. Although the magnitude of the relationship between sport involvement and graduation rate was small, considering that attrition studies typically find small if any relationships of predictor variables with graduation/attrition after relationships for traditional predictors of graduation/attrition are controlled (e.g., SAT scores and college GPA), the relationship seemed meaningful. The difference in the graduation rates for students involved and not involved in the sports program of around 10% (depending on which analysis was used) would be an exciting finding for university administrators if it represented the results of an experimental program designed to improve retention or an admission program designed to select students having a higher probability of graduating.


Asher, M. (1986). Abuses in college athletics. In R. Lapchick (Ed.), Fractured focus: Sport as a reflection of society (pp. 5-20). Lexington, MA: Lexington.

Bean, J. P., & Metzner, B. S. (1985). A conceptual model of nontraditional undergraduate student attrition. Review of Educational Research, 55, 485-540.

Brede R. M., & Camp, H. J. (1987). The education of college student-athletes. Sociology of Sport Journal, 4, 245-257.

Cohen, J. (1988). Statistical power analysis for the behavioral sciences. Hillsdale, NJ: Lawrence Erlbaum.

Cohen, J., & Cohen, P. (1983). Applied multiple regression/correlation analysis for the behavioral sciences. Hillsdale, NJ: Lawrence Erlbaum.

College Entrance Examination Board (1987). Registration Bulletin 1987-88. New York: Author.

Doyle, R. C. II. Lewis, J. M., & Malmisur, M. (1980). A sociological application of Rooney's fan region theory. Journal of Sport Behavior, 3, 51-60.

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

Henschen, K. P., & Fry, D. (1984). An archival study of the relationship of intercollegiate athletic participation and graduation. Sociology of Sport Journal, 1, 52-56.

Hersch, H. (1989, September 18). The right to know. Sports Illustrated, p. 18.

Jacobs, K. J. (1985). A comparison of the graduation rates of student-athletes with overall student body who enrolled at the University of North Carolina at Chapel Hill from 1966 to 1975 (M. A. thesis, University of Oregon). Microform Publications, University of Oregon. U085 175.

Kleiber, D., & Malik, P. (1989). Educational involvement of college athletes and subsequent well-being in early adulthood. Journal of Sport Behavior, 12, 203-211.

Lowe, B., & Harrold, R. D. (1972). The student as sport consumer. Paper presented at the 75th annual meeting of the National College of Physical Education for Men, New Orleans, LA.

McPherson, B. D. (1975). Sport consumption and the economics of consumerism. In D. N. Ball & J. W. Loy (Eds.), Sport and social order: Contributions to the sociology of sport (pp. 243-275). Reading, MA: Addison-Wesley.

Myers, I., & McCaulley, M. H. (1985). Manual: A guide to the development and use of the Myers-Briggs Type Indicator. Palo Alto, CA: Consulting Psychologists.

Pantages, T. J., & Creedon, C. F. (1978). Studies of college attrition---1950-1975. Review of Educational Research, 48, 49-101.

Purdy, D., Eitzen, S., & Hufnagel R. (1985). Are athletes also students? The educational attainment of athletes. In D. Chu, J. Segrave, & S. Becker (Eds.), Sports and higher education (pp. 221-234). Champaign, IL: Human Kinetics.

Schurr, K. T., Ruble, V. E., & Ellen, A. (1985). Myers-Briggs Type Indicator and demographic characteristics of students attending and not attending a college basketball game. Journal of Sport Behavior, 8, 181-194.

Schurr, K. T., Wittig, A. F., & Ruble, V. E. (1988). Demographic and personality characteristics associated with persistent, occasional, and non-attendance of university basketball games. Journal of Sport Behavior, 11, 3-17.

Shapiro, B. J. (1984). Intercollegiate athletic participation and academic achievement: A case study of Michigan State University student-athletes, 1950-1980. Sociology of Sport Journal, 4, 46-51.

Snyder, E. E. (1985). A theoretical analysis of academic and athlete roles. Sociology of Sport Journal, 2, 210-217.

SPSS, Inc. (1988). SPSSX user's guide. Chicago: Author.

Tinto, V. (1975). Dropouts from higher education--A theoretical synthesis of recent research. Review of Educational Research, 45, 89-125.

United States General Accounting Office (1989). Student athletes: Information on their academic performance. Fact sheet for the Chairman, Subcommittee on Postsecondary Education, Committee on Education and Labor, House of Representatives. Washington, DC: General Accounting Office. (ERIC Document Reproduction Service No. 310 106).

Weber, L., & Sherman, T. M. (1987). Effects of a transition program on student athletes' academic success: An exploratory study. Journal of Sport Sociology, 4, 78-83.
COPYRIGHT 1993 University of South Alabama
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1993 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Schurr, K. Terry; Wittig, A.F.; Ruble, V.E.; Henriksen, L.W.
Publication:Journal of Sport Behavior
Date:Mar 1, 1993
Previous Article:Effect of goal difficulty on performance of aerobic, anaerobic and power tasks in laboratory and field settings.
Next Article:Conflict with baseball umpires: an observational study.

Related Articles
Study takes measure of college athletes.
arena Higher Learning?
The Influences of Peer and University Culture on Female Student Athletes' Perceptions of Career Termination, Professionalization, and Social...
New ranking shows high graduation rates for some UO sports.
UO athletes' graduation rates measured.
UO doesn't make the grade on graduation.
Graduation rates mixed bag for UO.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters