American Educational Research Journal

October 2016, Vol. 53, No. 5, pp. 1376–1410

DOI: 10.3102/0002831216667579
Ó 2016 AERA. http://aerj.aera.net

Extracurricular Activity Participation in

High School: Mechanisms Linking
Participation to Math Achievement
and 4-Year College Attendance
David S. Morris
Tulane University

Extracurricular activity participation (EAP) has been positively linked with

increased academic achievement and college attendance. However, the
mechanisms linking EAP to educational outcomes are poorly understood.
Using the Education Longitudinal Study of 2002 (ELS), this study contributes
to our understanding of the relationship between EAP and educational suc-
cess by (1) examining the ability of nine educational, social, and develop-
mental mechanisms to explain the link between EAP and high school
math achievement gains and the chances of 4-year college attendance
and (2) examining the ability of family income to moderate the influence
of these mechanisms. Results suggest that educational expectations, noncog-
nitive skills, and social capital in the form of communication among paren-
tal groups are meaningful mediators, regardless of family income. These
findings extend the literature concerned with understanding how EAP is
related to academic outcomes, a connection that is not necessarily intuitive.

KEYWORDS: achievement, college, extracurricular activities

T he benefits of extracurricular activity participation (EAP) on educational

outcomes have been widely acknowledged through empirical research
(Covay & Carbonaro, 2010; Gardner, Roth, & Brooks-Gunn, 2008;
Mahoney, Harris, & Eccles, 2006). Although some types of extracurricular
activities (EAs) have been associated with negative developmental
outcomes—the link between sports participation and increased alcohol con-
sumption and the link between intense participation and decreased educa-
tional outcomes offer examples (Denault, Poulin, & Pedersen, 2009;

DAVID S. MORRIS is a visiting assistant professor of sociology at Tulane University, 220

Newcomb Hall, New Orleans, LA 70118, USA; e-mail: [email protected]. His
research focuses on family and the schooling process, school security and student
behavior, inequality, and the political system.
 Extracurricular Activity in High School
Eccles, Barber, Stone, & Hunt, 2003; Fredricks, 2012)—the overall landscape
largely suggests that EAP is beneficial for young people in a variety of ways
(Barber, Eccles, & Stone, 2001; Fredricks & Eccles, 2006; Hansen, Larson, &
Dworkin, 2003; Stanton-Salazar, 2011). When it comes to crucial educational
outcomes, involvement in EAP is associated with higher math achievement
test scores for elementary school students and increased math scores for
high school students (Broh, 2002; Covay & Carbonaro, 2010; Dumais,
2006; Morris, 2015). EAP is also linked to an increase in the chances of col-
lege attendance (Barber et al., 2001; Gardner et al., 2008). These are two crit-
ical educational outcomes, as math skills are a predictor of academic success
in other academic domains and a predictor of educational attainment and
labor market productivity (Adelman, 2006; Duncan et al., 2007; Rose &
Betts, 2001), while obtaining a 4-year college degree is closely linked with
increased income, social status, health, and marital stability, among a bevy
of other social issues (Cockerham, 2012; Fry, 2013; Martin, 2004).
It is unclear, however, why EAP is associated with academic achieve-
ment and the likelihood of graduating from high school and going to college
and whether these factors vary based on important family background char-
acteristics. Few studies have attempted to answer this question by examining
the mediating factors that help to explain the benefits of EAs, prompting
scholars to call for further examination into moderators and mediators
(Feldman & Matjasko, 2005; Lawson & Lawson, 2013). Prior research has
attempted to identify mediating factors such as social capital accumulation,
self-concept, and work ethic (Barber et al., 2001; Broh, 2002). However,
these studies primarily attempt to explain the link between EAP and psycho-
logical and developmental outcomes like substance abuse, self-esteem, and
ambition (Barber et al., 2001; Guest & Schneider, 2003) or explain the link
between one specific type of EAP and academic achievement (Broh, 2002).
To address this gap in the literature, this study examines how nine
potential mechanisms mediate the relationship between varying forms of
EAs (e.g., time spent in EAP, academic clubs, arts activities) and math
achievement growth in high school and the chances of attending a 4-year
college. This study also examines the ability of an important family back-
ground characteristic—family income—to moderate the relationship
between EAP and math achievement and college attendance.
By doing so, this study provides an important addition to the body of
research on the educational effects of EAP by shedding light on which activ-
ities and mechanisms are responsible for math achievement and pursuit of
a 4-year degree, something that has generally been lacking in EAP research.
Addressing this gap in our understanding is key since parents, K–12
schools, postsecondary institutions, and students themselves have placed
increased importance upon EAs as a training ground for educational
improvement and a signifier of social and academic competency (Lareau,
2003; Lawson & Lawson, 2013). Furthermore, work explaining the manner

1377
Morris
in which social and educational inequality is reproduced points to parental
attitudes about EAP related to family income as a primary player (Lareau,
2003). Understanding exactly how EAP is linked to achievement and college
attendance—a connection that is not necessarily intuitive—and whether the
explanations vary by family income are key to understanding the role of EAP
in school success for various students.

Extracurricular Activity Participation, Academic

Achievement, and College Attendance
Although there have been concerns that too much participation may
result in diminishing returns or a drop in academic achievement
(Fredricks, 2012; Mahoney et al., 2006), studies overwhelmingly indicate
that EAP benefits achievement. Using data from the Early Childhood
Longitudinal Study-Kindergarten Class of 1998 (ECLS-K), Dumais (2006)
found that the intensity of participation in EAs in early elementary school
predicted gains in reading achievement in subsequent years. Using the
same dataset, Covay and Carbonaro (2010) found that participation in school
clubs was associated with gains in math and reading achievement between
first and third grade. Focusing on high school students, Broh (2002) exam-
ined the impact of sports participation on math and reading achievement,
and her results show that EAP in interscholastic sports activities predicts
math and reading achievement in 12th grade. Looking at both intensity of
participation and specific types of activity participation, Morris’s (2015) find-
ings indicate that the amount of time spent in EAP and participation in aca-
demic activities are related to gains in math achievement for high school
students.
Although math achievement is an important predictor of success in other
academic areas and is related to how far in school a student goes, achieve-
ment and college attendance are not as predictive of each other as many
assume (Jencks & Brown, 1975; Rumberger & Palardy, 2005). Because of
this, studies examining the academic effects of EAP have also looked at its
impact on college attendance. Using longitudinal data, Mahoney, Cairns,
and Farmer (2003) found that consistent participation in EAs in middle
and high school was related to the chances of attending college. Similarly,
the results from the study completed by Zaff, Moore, Papillo and Williams
(2003) indicate that in-school and out-of-school EAP between 8th and 12th
grade predicted college attendance after high school. Gardner et al. (2008)
found that greater duration and intensity of EAP was associated with
increased chances of attending college. A large body of research strongly
indicates that EAP is generally beneficial for both the chances of attending
college and academic achievement—two crucial indicators of educational
and social accomplishment—but the question remains as to why this is so,
and for whom.

1378
 Extracurricular Activity in High School
problem. It is more likely that participation in such activities affects a young
person’s overall academic self-knowledge and performance expectations. By
considering a general academic self-concept, it makes the possibility of
transfer from a sports or music domain to an academic domain more
plausible.

The Current Study and Research Questions

Past studies have demonstrated that academic achievement and 4-year
college attendance are affected by participation in EAs in high school.
However, there is little empirical explanation as to why. Further, it is unclear
whether the mechanisms linking EAP with achievement and 4-year college
attendance are moderated by a critical family background characteristic:
family income. This study attempts to address these gaps in the literature
by posing the following research questions:

1. How do educational outlook, noncognitive skills, social capital, school attach-

ment, and academic self-concept/efficacy mediate the relationship between
extracurricular activity participation and math achievement growth in high
school?
2. How do educational outlook, noncognitive skills, social capital, school attach-
ment, and academic self-concept/efficacy mediate the relationship between
extracurricular activity participation and the chances of attending a 4-year
college?
3. Does family income moderate the mediating effects of educational outlook,
noncognitive skills, social capital, school attachment, and academic self-
concept/efficacy?

Data and Methods

The first three waves of the Education Longitudinal Study of 2002 (ELS)
were used for the analysis. The ELS base-year study was completed in 2002
when nearly all respondents were in 10th grade and initially consisted of
15,362 students from 752 schools. Students were retained for the analysis
if they had a math item response theory (IRT) score for 12th grade and pro-
vided a response about how far in school they had gone as part of the sec-
ond follow-up 2 years after expected high school graduation in 2006. This
resulted in a final sample of 12,058 students from 748 schools. Students
included in the sample were compared with students removed from the
analysis, and the two samples are quite similar. However, those included
in the final sample are more likely to be White, come from a more educated
family with higher income, and attend a private school with fewer peers on
free or reduced price lunch (see Appendix A).

1385
Morris
Table 4
Math Achievement Gains/4-Year College Attendance Regressed
on Interactions Between Family Income (in thousands) and
Mechanisms and Individual Mechanisms Regressed on Interactions
Between Family Income (in thousands) and EAP

Math Achievement 4-Year College Attendance

Model 1 Model 2
Interactions B SE Log Odds SE

Income*Importance of good grades .001 (.002) –.001 (.001)

Income*Educational expectations .001 (.002) .000 (.001)
Income*Noncognitive skills –.001 (.002) .000 (.001)
Income*Talks to teacher –.003 (.002) –.001 (.001)
Income*Talks to parents –.002 (.002) .000 (.001)
Income*Parents contact school –.000 (.001) .000 (.001)
Income*Parents talk to parents –.001 (.002) .001 (.001)
Income*School attachment .001 (.002) .000 (.001)
Income*Self-concept –.000 (.002) .000 (.001)
EAP P P
Mechanisms P P
Controls P P
N 12,058 12,058

Statistical significance of mechanisms No statistical

regressed on interactions between significance
EAP and family income

Note. OLS regression with ‘‘cluster’’ option is used to analyze interactions related to math
achievement. Logistic regression is used to analyze interactions related to the log-odds of
attending a 4-year college. Controls include race, gender, family income, parental educa-
tion, age, curriculum track, family structure, percentage students in school on free lunch,
school sector, school urbanicity, and school region. Main effects were included in all mod-
els. Weights (F2BYWT) were applied to all models. Interactions were also run between
mechanisms and tertiles of family income/quartiles of family income/parental educational
attainment with no substantive difference in the results. The bottom portion of the table
indicates that separate OLS regression models were also run to analyze interactions
between each form of EAP and family income with each of the nine mechanisms (models
included controls) as an outcome variable with no statistically significant interactions.
*p \ .05. **p \ .01. ***p \ .001.

Discussion
With an increased emphasis on involvement in EAs over the last several
decades by families, students, and schools, the potential benefits—and possible
detriments—of EAP have been widely studied. However, we know very little as
to why EAP is beneficial, especially when it comes to academic success and

1398
 Extracurricular Activity in High School
postsecondary attendance. Nor do we know how potential explanations of the
link between EAP and educational success might differ based upon important
family background characteristics like family income. Using longitudinal data,
this study set out to address these gaps in the literature by investigating the man-
ner in which nine educational, social, and developmental mechanisms mediate
the relationship between EAP and two critical educational outcomes: math
achievement and 4-year college attendance. Additionally, this study set out to
investigate the ability of family income to moderate the influence of these mech-
anisms on achievement and college attendance.
Based on the findings, three conclusions stand out. First, educational
expectations and noncognitive skills appear to be the most meaningful
mediating factors. Second, accumulation of social capital—in particular,
social connections and communication among groups of parents—appears
to mediate the link between EAP and college attendance. Third, family
income does not appear to moderate the relationship among EAP, mediating
mechanisms, and math achievement/4-year college attendance, lending no
support to compensation theory in this instance. EAP has been shown to dif-
ferentially aid lower income students (Dumais, 2006; Morris, 2015), but it
appears that mediation of these links does not differ by family income, at
least in terms of the mediating mechanisms under consideration here.
Figure 2 depicts these results in two different conceptual diagrams.
These findings extend previous research demonstrating the ability of non-
cognitive skills to partially mediate the relationship between EAP and academic
achievement in elementary school (Covay & Carbonaro, 2010) to the high school
realm. It is likely that involvement in EAs provides students a forum to practice
promptness, organization, delayed gratification, self-control, and attention to
detail. This would explain why students from various levels of family income
appear to benefit from the noncognitive skill growth provided by EAP since
young people from all walks of life need practice at such behaviors.
Additionally, the findings corroborate previous research that demonstrates the
importance of social capital accumulation from EAP for academic gains (Broh,
2002), especially in the form of connection among parents. It appears that
EAP can provide an important arena for social network closure, bringing chil-
dren and parents together to create relationships, organize, share information,
and build shared value systems related to educational success and furtherance.
This study’s findings underscore the importance of participation in EAs
that stress the possibility of and pathways to higher education, opportunities
to practice behaviors related to the hidden curriculum, and bringing children
and parents together to interact and share information about school. It
appears that more time spent in EAs overall and participation in academic
activities, school clubs, and junior varsity and varsity sports offer these
assets, but if other EAP (e.g., community activities) were to incorporate
such components as part of their process, they too may positively affect aca-
demic achievement and college attendance. It may be helpful for

1399
Morris

Educational
Expectations &
Noncognitive
Family Skills
Income

Time Spent in Math

EAP, Academic Achievement
Activities, & in High School
School Clubs

Educational
Expectations,
Noncognitive Skills,
& Communication
Family among Parents
Income

Time Spent in
EAP, Academic 4-Year College
Activities, Attendance
School Clubs, &
Letter Sports

Figure 2. Conceptual models of the results.

1400
 Extracurricular Activity in High School
educational and community leaders to stress the inclusion of expectations
about college attendance, noncognitive skill practice, and social network
closure in EAs. Furthermore, it may be in the best interest of parents and
young people to seek out EAs that offer such opportunities.
It is important to stress here that these findings do not indicate, nor is the
claim being made, that the battery of mechanisms under consideration in this
study fully explain the link between EAP and math achievement and 4-year
college attendance. Although educational expectations, noncognitive skills,
and a particular form of social capital explain some of the effect of EAP,
a sizeable portion of the relationship remains unexplained. Consequently,
other important mediating factors must be at play when it comes to EAP
and educational outcomes and should be explored in future research.
In addition to this, the current study suffers from other shortcomings. First,
the analysis is correlational and not causal. Using OLS and logistic regression
prevents this study from concluding that the relationship between EAP and edu-
cational outcomes is causal. This also prevents the study from sufficiently con-
cluding that the mediating mechanisms under investigation are indeed affected
by EAP and then, in turn, affect math achievement and the chances of attending
a 4-year college. The study attempts to account for this shortcoming by holding
constant prior math achievement in high school, thereby measuring achieve-
ment growth. The analysis also employs Sobel tests to divide the total effects
of EAP into direct and indirect effects and then checks those findings against
results gathered from analyses using the KHB method for testing mediation.
Still, the statistical models remain correlational. Second, the composite measures
for the educational, social, and developmental mechanisms are somewhat nar-
row due to data limitations and may lack particular dimensions of each concept
or academic domains. Measures with more domain specificity (math self-
concept, for example) may provide different results, as may variables that mea-
sure depth and detail of social interactions among students, parents, and teach-
ers. Third, the analysis was only able to include math achievement as an
indicator of academic achievement since other measures of achievement
were completed in 10th grade only. Fourth, the current study only examines
high school students; other factors may be in play for younger children.

Conclusion
The current study makes an important addition to the body of literature
on EAP by demonstrating why various forms of participation are positively
related to two critical educational outcomes: high school math achievement
gains and 4-year college attendance—a connection that is not necessarily
intuitive. Prior studies that do examine mediating mechanisms have primar-
ily focused on one form of EAP and have not examined both achievement
and 4-year college attendance. Moreover, this study shows that the mediat-
ing ability of the various mechanisms does not appear to vary based on

1401
Morris
family income, a key predictor of many educational outcomes. Still, since
this study only explains a portion of the link between EAP and two educa-
tional outcomes, future research is needed to examine other mediating fac-
tors that further explain this relationship. In particular, data that have a wider
variety of activities to include in composite measures of EAP would be espe-
cially beneficial, as would examination of other potential mechanisms and
measures of academic achievement and educational success.

Appendix Table A1
Demographic Characteristics of Students Kept in Sample
and Students Dropped From Sample

Students in Sample Students Dropped From Sample

Mean SD Min Max Mean SD Min Max

Parent high school 0.233 0.423 0 1 0.349 0.477 0 1

degree or less
Parent attended 0.324 0.468 0 1 0.338 0.473 0 1
some college
Parent has college 0.443 0.497 0 1 0.314 0.464 0 1
degree or more
Income 66k 49k 0 200k 54k 44k 0 200k
Age 16.1 0.482 14.3 19.2 16.4 0.723 14.3 19.3
Female 0.520 0.500 0 1 0.445 0.497 0 1
Student is White 0.594 0.491 0 1 0.494 0.500 0 1
Student is Black 0.125 0.330 0 1 0.157 0.364 0 1
Student is Hispanic 0.132 0.338 0 1 0.189 0.391 0 1
Student is other race 0.150 0.357 0 1 0.160 0.367 0 1
Private school 0.239 0.427 0 1 0.132 0.339 0 1
Percentage free lunch 22.6 23.5 3 88 29.2 25.9 3 88
Urban school 0.332 0.471 0 1 0.357 0.479 0 1
Suburban school 0.485 0.500 0 1 0.463 0.499 0 1
Rural school 0.183 0.386 0 1 0.180 0.384 0 1
Northeastern school 0.189 0.391 0 1 0.167 0.373 0 1
Midwestern school 0.258 0.437 0 1 0.223 0.416 0 1
Western school 0.199 0.399 0 1 0.223 0.416 0 1
Southern school 0.355 0.478 0 1 0.388 0.487 0 1

Note. Dataset used for analysis is the first three waves of the Educational Longitudinal
Study of 2002.

1402
 Extracurricular Activity in High School
References
Adelman, C. (2006). The toolbox revisited: Paths to degree completion from high
school through college. Washington, DC: U.S. Department of Education.
Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and interpreting inter-
actions. Thousand Oaks, CA: Sage.
Barber, B. L., Eccles, J. S., & Stone, M. R. (2001). Whatever happened to the jock, the
brain, and the princess? Young adult pathways linked to adolescent activity
involvement and social identity. Journal of Adolescent Research, 16(5), 429–455.
Blomfield, C. J., & Barber, B. L. (2009). Brief report: Performing on the stage, the field,
or both? Australian adolescent extracurricular activity participation and self-
concept. Journal of Adolescence, 32(3), 733–739.
Blomfield, C. J., & Barber, B. L. (2011). Developmental experiences during extracur-
ricular activities and Australian adolescents’ self-concept: Particularly important
for youth from disadvantaged schools. Journal of Youth and Adolescence,
40(5), 582–594.
Boardman, J. D., & Robert, S. A. (2000). Neighborhood socioeconomic status and
perceptions of self-efficacy. Sociological Perspectives, 43(1), 117–136.
Bong, M., & Skaalvik, E. M. (2003). Academic self-concept and self-efficacy: How dif-
ferent are they really? Educational Psychology Review, 15(1), 1–40.
Breen, R., Karlson, K. B., & Holm, A. (2013). Total, direct, and indirect effects in logit
and probit models. Sociological Methods & Research. 42(2), 164–191.
Brooks-Gunn, J., & Duncan, G. J. (1997). The effects of poverty on children. The
Future of Children, 7(2), 55–71.
Broh, B. A. (2002). Linking extracurricular programming to academic achievement:
Who benefits and why? Sociology of Education, 75(1), 69–95.
Catalano, R. F., Oesterle, S., Fleming, C. B., & Hawkins, J. D. (2004). The importance
of bonding to school for healthy development: Findings from the social devel-
opment research group. Journal of School Health, 74(7), 252–261.
Cockerham, W. C. (2012). Medical sociology. Upper Saddle River, NJ: Pearson
Prentice Hall.
Coleman, J. (1988). Social capital in the creation of human capital. American Journal
of Sociology, 94(1), 95–120.
Covay, E., & Carbonaro, W. (2010). After the bell: Participation in extracurricular
activities, classroom behavior, and academic achievement. Sociology of
Education, 83(1), 20–45.
Darling, N. (2005). Participation in extracurricular activities and adolescent adjust-
ment: Cross-sectional and longitudinal findings. Journal of Youth and
Adolescence, 34(5), 493–505.
Denault, A.-S., Poulin, F., & Pedersen, S. (2009). Intensity of participation in orga-
nized youth activities during the high school years: Longitudinal associations
with adjustment. Applied Developmental Science, 13(2), 74–87.
Downey, D. B., & Condron, D. J. (2016). Fifty years since the Coleman report:
Rethinking the relationship between schools and inequality. Sociology of
Education, 89(3), 207–220.
Downey, D. B., von Hippel, P. T., & Broh, B. A. (2004). Are schools the great equal-
izer? Cognitive inequality during the summer months and the school year.
American Sociological Review, 69(5), 613–635.
Dumais, S. A. (2006). Elementary school students’ extracurricular activities: The
effects of participation on achievement and teachers’ evaluations. Sociological
Spectrum, 26(2), 117–147.

1407