Religious Involvement and Educational Outcomes: The Role of Social Capital and Extracurricular Participation

Abstract

Previous research has observed that religious participation is positively related to a wide variety of adolescent outcomes, including academic achievement, but relatively little is known about why this is the case. We focus on a group of related potential explanations for why religious involvement improves educational outcomes. We examine whether religious participation enhances academic outcomes among teens by the way in which it shapes their social ties, or social capital, focusing on both intergenerational relationships and on relationships with peers. We also examine the potential intervening role of extracurricular participation. Using structural equation models to analyze data from the National Longitudinal Study of Adolescent Health (Add Health), we examine the potential role of social capital and extracurricular participation in mediating the relationship between religious participation and academic achievement, dropping out of high school, and attachment to school. We find that religious attendance promotes higher intergenerational closure, friendship networks with higher educational resources and norms, and extracurricular participation. These intervening variables account for a small part of the influence of adolescent religious participation on the educational outcomes in this study.

ACKNOWLEDGMENTS

This research uses data from Add Health, a program project designed by J. Richard Udry, Peter S. Bearman, and Kathleen Mullan Harris, and funded by a grant P01-HD31921 from the National Institute of Child Health and Human Development, with cooperative funding from 17 other agencies. Special acknowledgment is due Ronald R. Rindfuss and Barbara Entwisle for assistance in the original design. Persons interested in obtaining data files from Add Health should contact Add Health, Carolina Population Center, 123 W. Franklin Street, Chapel Hill, NC 27516-2524 (http://www.cpc.unc.edu/addhealth/contract.html). This research also uses data from Adolescent Health and Academic Achievement (AHAA). The AHAA study was funded by grants from the National Institute of Child Health and Human Development (01 HD40428-02) to the Population Research Center, University of Texas at Austin, Chandra Muller (PI), and the National Science Foundation (REC-0126167) to the Population Research Center, University of Texas at Austin, Chandra Muller and Pedro Reyes (Co-PI). We thank the Center for the Study of Latino Religion at the University of Notre Dame for research support and Robert Baller, Karen Heimer, Mary Noonan, Anthony Paik, and four anonymous reviewers for comments on an earlier version of this article.

NOTES

Notes

¹ The findings of Carbonaro (1999) and Morgan and Sørensen (1999) cast some doubt on the idea that intergenerational closure fosters educational achievement. However, Morgan and Sørensen (1999) examine school-level rather than family-level closure. In reanalyzing the data from his 1998 article, Carbonaro (1999) concludes that the influence he found for intergenerational closure was spurious because when number of friends in school is added to the model, the effect of parents knowing their child's friends' parents disappears. An alternative interpretation is the measure of “friends in school” is actually a dimension of closure since it is based on the number of friends that the parent could name. Moreover, the current study's measure of closure gauges communication among parents, rather than simply whether parents are known to one another, a potentially stronger measure of closure. See also Hallinan and Kubitschek (1999) for a discussion of the limitations of the measures used in the Carbonaro (1999) and Morgan and Sørensen (1999) analyses.

² Part of the reason for conflicting findings may be that the school context conditions the influence of various types of extracurricular participation (Guest and Schneider 2003).

³ Because the transcript data are only available for the high school years, the sample for this analysis is limited to respondents who were in high school one year after Wave I.

⁴ The chi-square differences were 4.42 with 6 degrees of freedom (p = .62), 3.51 with 6 degrees of freedom (p = .74), and 6.947 with 5 degrees of freedom (p = .22) for the GPA, dropout, and attachment models, respectively. Thus, constraining the coefficients to be equal does not result in a significant loss of fit.

⁵ Unfortunately, parents were not presented with a list of friends identified by the adolescent, which would lead to a more accurate measure of intergenerational closure. A reviewer suggested that this measure might reflect misplaced parental confidence in their knowledge of their child's friends and that this confidence would be correlated with requiring their children to attend church. To examine whether this explanation might account for the relationship between the Add Health closure measure and adolescent religious attendance, in auxiliary analyses we included controls for (1) how much freedom the teen's parents allow them in making decisions about what time they come home, what time they go to bed, the television they watch, the people they associate with, and what they wear; (2) activities the teen does with his or her mother, including shopping, playing sports, discussing dating or parties, going to movies or other entertainment events, discussing personal problems, and talking about school work; and (3) how much time the teen spends “just hanging out with friends.” Including these additional variables only slightly reduced the coefficient for religious attendance in the closure equation, a reduction that was not statistically significant.

⁶ Results are available upon request.

⁷ Religious teachings highlight the desirability of strong familial relationships (Pearce and Axinn 1998), and places of worship provide support for families by organizing activities such as parenting classes, retreats, camps, and counseling that enable families to bond by spending time together (Abbott, Berry, and Meredith 1990; Maton and Wells 1995; Wilcox et al. 2004). Previous research has suggested that religious participation improves parent–child relationships (Pearce and Axinn 1998; Wilcox 1998). However, it is also possible that higher functioning families attend religious services more often. In the earlier analyses with the Add Health data we found that initial family satisfaction positively predicted change in religious attendance, but initial attendance did not positively predict change in family satisfaction. Therefore, rather than conceptualize family satisfaction as an intervening social capital variable, we treat it as a control variable.

⁸ In additional analyses we estimated the models using listwise deletion. Both sets of results are very similar.

⁹ While research on the relationship between extracurricular participation and educational outcomes has generally focused on the influence of participation on educational achievement and attainment, it is important to recognize that prior achievement and engagement likely influences extracurricular participation (Finn 1989). Therefore, we include contemporaneous educational achievement and engagement in the extracurricular participation equations. An alternative would be to intercorrelate these control variables with the error terms for the extracurricular participation equations. The approach we employ is more conservative than this alternative; with the alternative approach the estimated influence of religious attendance on extracurricular participation is larger.

¹⁰ An alternative to including the respondent's GPA and truancy at time 1 in the friends' GPA and truancy equations would be to intercorrelate time 1 GPA and truancy with the error terms for friends' GPA and truancy. Not surprisingly, when modeled this way, the estimated influence of religious attendance on friendship characteristics is slightly larger. However, it makes little difference to the estimated direct and indirect effects of religious attendance, and therefore, we present the former approach as it is the more conservative of the two.

¹¹ Results are available upon request.

¹² Closure is, however, statistically associated with attachment when the lagged dependent variable is not included in the equation.

¹³ Some scholars have argued that intense involvement in some religious communities, particularly conservative Protestant groups, undermines academic accomplishments by encouraging withdrawal from the secular world. For example, Darnell and Sherkat (1997) and Lehrer (1999) find that educational attainment is lower among fundamentalist Protestants than it is among mainline Protestants and Jews. Therefore, in auxiliary analyses we examined whether conservative Protestant adolescents had worse educational outcomes and whether the influence of attendance on educational outcomes is less positive for this group. There was no significant interaction between conservative Protestant and attendance, and conservative Protestant teens did not differ from other teens on any of the academic outcomes, net of the control variables, except that they were somewhat more likely to drop out of high school.