Interscholastic Sports Participation and School Based Delinquency: Does Participation in Sport Foster a Positive High School Environment?

Abstract

This article uses social control and cultural spillover theories to frame how sports programs might impact the school environment. Using a nationally representative sample of roughly 1,200 public high schools, the relationship between school sports participation rates and in-school delinquent behaviors are examined. Results indicate that schools with higher proportions of sports participants report significantly fewer serious crimes (i.e., violent crimes) and suspensions occurring on school grounds. However, the incidence of minor crimes is unrelated to the proportion of students engaging in school sports. The results of this study support social control theory and suggest that one way to reduce violence and delinquency in schools is through encouraging participation in sports programs.

Acknowledgments

The authors are grateful to the Women's Sport Foundation, the National Center for Education Statistics, and the anonymous reviewers of Sociological Spectrum to help guide the completion of this article.

Notes

¹In order to conform to IES restricted-use data publication policy, all data from the SSOCS unweighted sample sizes are rounded to the nearest 10.

²Serious crime incidence rate, minor crime incidence rate, suspension rate, and total crime rate are measures constructed from using both the SSOCS and the CCD. The number of detailed criminal incidents and suspensions within high schools are collected by the SSOCS for both the 2003–2004 and 2005–2006 school years. The total number of students within each high school is collected by the CCD annually and these enrollment figures were used for schools that were either sampled by the SSOCS during the 2003–2004 or 2005–2006 school year. It should also be noted that the SSOCS restricted data files also included enrollment figures within each data file. Bivariate correlations revealed that the enrollment figures between the CCD and SSOCS are quite similar (2003–2004 CCD and 2003–2004 SSOCS, r = .970, p < .001; 2005–2006 CCD and 2005–2006 SSOCS, r = .989, p < .001).

³Total number of students suspended was constructed by adding the number of students who were suspended for possession of firearms/explosives, possession of weapons other than firearms/explosives, distribution or use of illegal drugs, distribution or use of alcohol, physical attacks or fights, and insubordination within each school.

⁴Number of participants in school sponsored sports is provided in the 2004 and 2006 CRDC. Total number of students enrolled within each school is provided by the 2003–2004 and 2005–2006 CCD.

⁵Number of students enrolled in AP math courses is provided in the 2004 and 2006 CRDC. Total number of students enrolled within each school is provided by the 2003–2004 and 2005–2006 CCD.

⁶Current academic year for the 2004 CRDC refers to the 2004–2005 school year and current academic year for the 2006 CRDC refers to the 2006–2007 school year. Unfortunately, AP math is the only measure in the analysis that did not have information for high schools for the 2003–2004 and 2005–2006 school years. This variable remains in the analysis because it still serves as a good proxy for academic achievement within high schools.

⁷CCD collection for the 1999–2000 and 2005–2006 school years defined eight geographic regions where schools could be located. “Large city” and “Midsize city” were combined to represent high schools located in urban areas. “Urban fringe of a large city” and “Urban fringe of a midsize city” were combined to represent high schools located in suburban areas. “Large town” (population ≥ to 25,000 located outside a metropolitan CBSA or inside a micropolitan CBSA) and “Small town” (population 25,000 ≥ to 2,500 and located outside a metropolitan CBSA or inside a micropolitan CBSA) were combined to represent schools located in towns. “Rural, outside CBSA” and “Rural, inside CBSA” were combined to represent schools located in rural area.

⁸The strongest correlation between all of the independent variables is between the percent of students who are eligible for free or reduced price lunches within high schools and the percent of white students within high schools (r = −.647, p < .001). The correlation matrix can be provided upon request.

⁹Caution should be given to any independent variable that has a VIF of 10 or higher. Paul Allison (1998) suggests a more conservative VIF estimate of 2.5, but given the sample size and relatively low VIF scores for all the coefficients in the regression model, issues of multicollinearity can be ruled out.

¹⁰The negative binominal models (i.e., schools' total population was used as the exposure term) that estimated the relationship between schools' sport participation rates and schools' crime rates found a significant negative relationship between high schools' sport participation rates and serious crime rates (b = −.432, p < .05), and a significant negative relationship between high schools' sport participation rates and suspension rates (−.904, p < .01). Further, no statistically significant relationship was found between high schools' sport participation rates and schools' total crime rate (b = −.047, non sig.), or between high schools' sport participation rates and schools' minor crime rate (b = .149, non sig.). Tables for the negative binomial analyses can be provided upon request.

Note. National figures from the CCD for high schools that offer only Grades 9–12 are in parentheses.

¹¹National Center for Education Statistics (2009) reports that among various extracurricular activities provided by high schools, interscholastic sports had the highest percentage of seniors (38.8%) participating in this school based activity in 2004 followed by academic clubs (21.3%), band (21%), vocational clubs (15.6%), and hobby clubs (11.7%).The percentage of students who participate in extracurricular activities in public high schools was retrieved from the Digest of Education Statistics: 2009. These figures can be accessed at the National Center for Education Statistics website nces.ed.gov.

¹²The percentage of students who were enrolled in AP calculus was retrieved from the Digest of Education Statistics: 2009. These figures can be accessed at the National Center for Education Statistics website nces.ed.gov.

Note. Standardized coefficients were calculated without the ROBUST CLUSTER command in STATA. The values in bold are statistically significant values in the regression models.

^a Schools sampled in 2003–2004 serve as the reference category.

^b Schools located in Suburban areas serves as the reference category.

*p < .05; **p < .01; ***p < .001.