Cool Girls, Inc. and Self-Concept: The Role of Social Capital. The Journal of Early Adolescence

Journal of Early Adolescence 2014, Vol. 34(6) 816 –836

© The Author(s) 2013 Reprints and permissions:

sagepub.com/journalsPermissions.nav DOI: 10.1177/0272431613511329

jea.sagepub.com

Article

Cool Girls, Inc. and Self-Concept: The Role of Social Capital

Jessica D. Thomason1 and Gabriel Kuperminc1

Abstract Social capital was examined as a mediator of the association between youth development program participation and gains in self-concept in a sample of 86 primarily African American female adolescents in the Cool Girls, program, and 89 comparisons. Two dimensions of social capital (the diversity of girls’ social networks and the number of life domains in which girls were able to access help) were examined. It was hypothesized that participation in Cool Girls would be associated with increases in social capital and that this would mediate the relationship between participation in Cool Girls and increases in self-concept. Social capital was associated with gains in two dimensions of self-concept regardless of program participation: social acceptance and behavioral conduct. Finally, there was a significant indirect effect of participation on social acceptance and behavioral conduct, mediated through social capital. Implications are discussed.

Keywords social capital, youth development, self-concept, adolescents

The neighborhood and school conditions in which youth are raised can create barriers or opportunities for their development (Bronfenbrenner, 1979). Many inner-city neighborhoods are isolated from social resources

1Georgia State University, Atlanta, USA

Corresponding Author: Jessica D. Thomason, Department of Psychology, Georgia State University, 140 Decatur St., 11th floor, Atlanta, GA, 30303, USA. Email: [email protected]

511329 JEAXXX10.1177/0272431613511329Journal of Early AdolescenceThomason and Kuperminc research-article2013

at PENNSYLVANIA STATE UNIV on May 8, 2016jea.sagepub.comDownloaded from

Thomason and Kuperminc 817

that provide opportunities for social mobility and psychological or emotional support (Fitzpatrick, Wright, Bettina, & LaGory, 2005; Stanton-Salazar, 1997, 2001, 2005). Researchers have found a link between increased social support and certain dimensions of adolescent self-concept (Demaray, Malecki, Rueger, Brown, & Summers, 2009), suggesting the importance of having access to supportive social networks. As a result, some programs attempt to provide youth with access to resources that might not otherwise be available to them. One example of this is Cool Girls, Inc., which notes in its mission statement that the program will empower girls and increase their self-esteem by increasing access to social resources and teaching girls how to access resources when in need. Thus, increasing social resources is seen as a mechanism through which youth eventually achieve desired youth develop- ment outcomes, including positive self-concepts.

Accessing Social Resources

Theories of social capital offer a useful framework for conceptualizing how youth programs, like Cool Girls, can help increase young peoples’ social net- works in ways that might increase opportunities for social mobility. Researchers use the term social capital to describe the extent of an individu- al’s access to people with diverse resources and view it as a resource that can provide opportunities and provide psychological and emotional support. One way to increase youth’s social capital is through participation in Youth Development Programs (YDPs). Drawing on research showing that the tran- sition to adolescence often poses risks for the development of girls’ self-con- cepts, some YDPs work exclusively with girls (LeCroy, 2004). Such programs often focus on increasing self-esteem (Royse, 1998) by challenging gender role stereotypes and encouraging self-determination (Kuperminc, Thomason, DiMeo, & Broomfield-Massey, 2011; LeCroy, 2004). In addition, due to the underrepresentation of women in positions of political power and in the fields of mathematics, science, technology, and engineering (Else-Quest, Hyde, & Linn, 2010), as well as persistent gender stereotypes, young girls may not see these opportunities as available to them. A common strategy of YDPs is to expand young peoples’ exposure to adult role models who can provide encouragement and practical guidance (Eccles & Gootman, 2002). It has been argued that developmental relationships (Li & Julian, 2012) or youth– adult partnerships (Zeldin, Christens, & Powers, 2013) that form between youth and these adult role models can serve as the active ingredients that promote positive youth development because they engage youth in reciprocal (e.g., shared interests) rather than prescriptive (e.g., skill training) interac- tions that can grow to increasingly draw on the youth’s contribution over

at PENNSYLVANIA STATE UNIV on May 8, 2016jea.sagepub.comDownloaded from

818 Journal of Early Adolescence 34(6)

time. Thus, increasing the network of these types of relationships in the lives of young people (i.e., their social capital) has the potential to be a key media- tor in the link between program participation and desired youth outcomes. In the present study, we examined youth’s perceptions of the acquisition of social capital as a mediator of the link between participation in Cool Girls and increases in self-concept.

Self-Concept in Adolescence

Self-concept is defined by how individuals assess their strengths and weak- nesses in multiple life domains (e.g., scholastic competence, behavioral con- duct; Harter, 1999), and is differentiated from self-esteem, as defined by a general sense of self-worth, or how one feels about oneself. Research has shown that the development of a positive self-concept during adolescence assists in negotiating new roles and relationships with others (Harter & Monsour, 1992, 1999). For example, perceived social acceptance has been related to several youth outcomes such as peer-reported hostile and aggres- sive behavior, peer-reported companionship, and teacher-reported adjust- ment on internalizing and externalizing behaviors and social problems (McElhaney, Antonishak, & Allen, 2008; Waldrip, Malcom, & Jensen- Campbell, 2008).

Research has also found significant gender differences in dimensions of self-concept, including body image (Nishina, Ammon, Bellmomre, & Graham, 2006; Peterson et al., 2007) and self-esteem (Zimmerman et al., 1997). Whereas both genders struggle with body image issues (Bearman & Stice, 2008; Jones, 2004; McCabe & Ricciardelli, 2003), girls’ concerns about their bodies are more likely to contribute to depression (Bearman & Stice, 2008; Cambron, Acitelli, & Pettit, 2009). In addition, girls faced with negotiating traditional gender roles as females while maintaining academic achievement (Skelton, Francis, & Read, 2010) often develop poorer math attitudes and stress related to stereotype threat (Else-Quest et al., 2010; Huguet & Régner, 2009). These challenges can spill over to other aspects of the self-concept as well, contributing to concerns about relationships and declines in overall self-worth (Miller & White, 2003).

Youth Development Programs

YDPs emerged in the 1990s as a response to single-issue prevention pro- grams that had become prevalent at that time. For example, many programs were implemented in schools and communities to prevent teenage pregnancy, or drug and alcohol abuse, or to reduce school drop-out rates. Researchers

at PENNSYLVANIA STATE UNIV on May 8, 2016jea.sagepub.comDownloaded from

Thomason and Kuperminc 819

and practitioners began to recognize that focusing on the positive develop- ment of the “whole child,” rather than single-issue risk-behaviors, could pre- vent a wide-range of negative outcomes and prepare youth for the future. In order to help youth develop positively in terms of their physical and psycho- logical health, it is necessary to focus on building up youth’s internal and external resources (Eccles & Gootman, 2002). Internally, YDPs seek to pro- mote positive self-concept, self-determination, and positive future orienta- tion, among others. Externally, YDPs often seek to promote and expand youth’s access to positive adult role models. Such programs encourage the integration of family, school, and community efforts in helping to support and empower youth, broaden their horizons, and promote skill building (Eccles & Gootman, 2002; Roth & Brooks-Gunn, 2003). In essence, many YDPs seek to expand the social capital networks of youth by putting them in contact with people in their communities with various knowledge and skills.

What Is Social Capital and Why Is It Important?

According to Bourdieu (1983), social capital refers to membership in a net- work of people that contribute to an individual’s access to material, practical, symbolic, or socially instituted resources (such as a family name). The amount of social capital depends on the size of a person’s network, as well as the amount held by the other members. The connections one has with indi- viduals who have capital themselves can lead to increased social and eco- nomic status via further connections, advice, and financial loans or other resources (Bourdieu, 1983; Coleman, 1988a; Ferguson, 2006; Wall, Ferrazzi, & Schryer, 1998).

Stanton-Salazar (1997) measured social networks in terms of the help domains (e.g., nonfamily, school-based, friendships) and whether ties to fam- ily and institutions were weak or strong in their ability to provide youth with “institutional knowledge” that could aid in social mobility. Later, using quan- titative and qualitative methods to explore youth’s social networks, Stanton- Salazar (2001) identified several categories of adults as providing social support among the most resilient participants. He also expanded this work to explore the importance of peer social networks, including platonic relation- ships between males and females, on the positive development of Mexican- origin youth (Stanton-Salazar, 2005). These studies on the social networks of resilient youth highlight the importance of having a large, diverse social net- work comprised of family members and individuals in institutions such as churches or schools. With a diverse social network, youth are more likely able to find someone to support them in multiple areas of their lives. Parents cannot always help with schoolwork, and teachers may not always be

at PENNSYLVANIA STATE UNIV on May 8, 2016jea.sagepub.comDownloaded from

820 Journal of Early Adolescence 34(6)

available to help each student solve a life problem. Therefore, studies that go beyond a measurement of network size to include a range of life domains with helping resources and the diversity of networks may provide more use- ful descriptions of social capital and its role in development.

Youth Development Programs and Social Capital

Despite being one of the “active ingredients” in the success of YDPs, researchers rarely have directly studied social capital and its role in promot- ing youth development. However, Jarrett, Sullivan, and Watkins’s (2005) qualitative study of three urban youth programs found that youth had increased access to supportive adults who provided them with information, assistance, exposure to adult worlds, and support and encouragement. In addition, in Sullivan and Larson’s (2010) qualitative study of seven primarily urban youth programs, youth described receiving information, skills, and access to adult worlds, even if these relationships were brief or superficial. These studies highlight the potential for YDPs to play a critical role in increasing youth’s access to adults who provide valuable resources. The research on natural or informal mentoring provides some insight into these processes as well. For example, Hurd and Sellers (2013) found that Black youth who described having “connected” relationships with a natural mentor did better academically through the psychological and social benefits of these relationships. Furthermore, Mekinda and Hirsch (in press) describe the char- acteristics of informal mentoring relationships that occur in after-school pro- grams as an important ingredient to developing positive psychological outcomes for youth. These studies suggest that there are benefits of natural mentoring relationships, and that these relationships can be built through par- ticipation in after-school programs similar to Cool Girls.

Youth Self-Concept and Social Capital

Social capital is related to several youth outcomes such as academic achieve- ment (Kao & Rutherford, 2007), extracurricular participation (Glanville, Sikkink, & Hernandez, 2008), substance use (Curran, 2007), drop-out rates (Groninger & Lee, 2001), and indicators of health and well-being (Rosenfeld, Richman, & Bowen, 1998). Research on the association between social sup- port and self-concept is also relevant, given conceptual overlap of social sup- port with social capital. For example, Rosenfeld et al. (1998) examined the social support of African American and White youth involved in a drop-out prevention program. The authors found that youth who reported low social support had significantly lower self-esteem than others.

at PENNSYLVANIA STATE UNIV on May 8, 2016jea.sagepub.comDownloaded from

Thomason and Kuperminc 821

In the present study, we examined the role of social capital in the develop- ment of self-concept for youth attending an Atlanta-based YDP. Because the program seeks not only to provide access to social resources directly, but also to teach girls how to seek out the support they need, we hypothesized that girls who participate will report significantly more social capital than those who do not. We also hypothesized that social capital would be associated with increases in self-concept and that social capital would mediate the rela- tion between participation and increases in self-concept.

Method

Program Description

Cool Girls, Inc. is an Atlanta, Georgia–based YDP with the primary focus of empowering low-income girls to “. . . become self-assured and to break cycles of poverty, teen pregnancy, racism, and sexism” (Cool Girls, Inc., 2008; Kuperminc et al., 2011). This multi-component program includes an after-school club (Cool Girls Club) for life skills development, several pro- grams for academic support, a technology program, business program, and a one-on-one mentoring program in which girls are matched with female vol- unteers to provide personal support, friendship, and role modeling (Cool Girls, Inc., 2008). The Cool Girls, Inc. theory of change asserts that increas- ing access to social resources and teaching girls how to seek out help when in need will aid their positive development. The Cool Girls Club seeks to help each girl be “. . . able to identify when she needs resources and support and knows how to access them” (Cool Girls, Inc., 2008). All girls enrolled in the program are expected to attend Cool Girls Club every week. Additional pro- gram components, including technology workshops and summer camps that focus on entrepreneurship, money management, career choices and leader- ship skills further provide girls with exposure to women in several fields such as business, health care, and “non-traditional” fields such as players from the Atlanta Dream Women’s Basketball team. The present study focuses on the “social resources” aspect of the Cool Girls, Inc. theory of change and evalu- ates whether participation in the program is related to increased access to social resources and whether these resources are associated with increases in self-concept.

Participants

We used data from a 2005-2006 evaluation of Cool Girls, Inc. that used a quasi-experimental demographically matched comparison group design.

at PENNSYLVANIA STATE UNIV on May 8, 2016jea.sagepub.comDownloaded from

822 Journal of Early Adolescence 34(6)

Pre-test data were collected from 175 girls (86 Cool Girls, 89 comparisons) at four elementary schools (Grades K-5), two middle schools (Grades 6-8), and two schools that served elementary and middle school–aged students (Grades K-8). Eighty participants were in elementary school, and 95 were in middle school, with ages ranging from 9 to 15. Eighty-eight percent identi- fied as African American, 4% as Latina/Hispanic, 2% as Asian/Vietnamese, 1% as Caucasian, and 4% as “Other.” Ethnicity data were missing for 2 girls. Post-test data were available for 150 of the girls. Cool girls and comparisons shared similar demographic characteristics with respect to age and ethnicity; however, Cool Girls were significantly less likely to be from a two-parent family, χ2(3, N = 175) = 4.89, p < .05. Independent-samples t-tests computed for all study variables at pre-test found no significant differences between Cool Girls and comparisons. These findings suggest that the groups were demographically similar and were relatively well matched with regard to social capital and self-concept at pre-test.

Procedure

All Cool Girls participants in Grades 5 through 8 who attended Cool Girls Club were eligible for the study. Comparisons were identified in two ways: through school guidance counselor nominations and by asking program par- ticipants to name two to three friends who were not in Cool Girls but whom they thought would answer questions the same way they did. The research team sent letters to Cool Girls participants and nominated comparison girls explaining the study to families. The Georgia State University Institutional Review Board approved these procedures; signed parental consent was obtained for both groups, as well as youth’s assent to participate.

The research team collected pre-test data in the beginning of the 2005 fall semester (September-October) and post-test data at the end of the 2006 spring semester (April-May). Participants completed paper-and-pencil surveys at the pre- and post-test in-group sessions that lasted approximately 30 minutes. An evaluator read each question aloud to account for variations in reading level. Cool Girls, Inc. participants and comparisons completed identical questionnaires. The evaluator reminded participants that their answers were confidential and that they could withdraw their participation at any time.

Measures

Participation in Cool Girls, Inc. Program participation was dummy coded, with Cool Girls, Inc. coded “1” and comparisons coded “0.”

at PENNSYLVANIA STATE UNIV on May 8, 2016jea.sagepub.comDownloaded from

Thomason and Kuperminc 823

Involvement in extracurricular activities. Cool Girls and comparisons reported their level of participation in sports or games, taking lessons or attending classes outside of school, and going to meetings outside of school for a club or youth group (other than Cool Girls). Response options were on a 4-point scale ranging from 0 = never to 3 = almost every day. To assess the number of activities in which youth were actively involved, we created an index vari- able measuring the number of activities in which youth reported participating at least “once or twice per week” (range = 0-3).

Social capital. We assessed social capital eliciting information about helping resources in the domains of schoolwork, support and guidance, problem- solving, and goal-setting. Participants provided information about whether they had someone to help them when they experienced difficulty in each domain (e.g., “Everyone needs help with schoolwork some times. I have someone to go to for help with schoolwork.”). If they responded “yes,” they provided a name and their relationship to the person. We then used responses to construct measures of two aspects of social capital: the number of domains in which the participant reported having help and the diversity of her net- work. We calculated number of help domains by counting the total number of domains in which the participant reported having help (range = 0-4). We coded relationships of helping resources to the participant into five catego- ries: immediate family (1), extended family (2), nonfamilial adult (3), friend/ peer (4), and other (5), and calculated diversity of network as a count of the total number of different relationship types represented among each partici- pant’s social capital resources (range = 0-4). We averaged the z-scores of these two dimensions to create an overall social capital score (network diver- sity and number of help domains were significantly correlated at pre-test and post-test, rs = .44 and .42, respectively, ps = .00).

Self-concept. Self-concept was measured using an adaptation of the Self-Per- ception Profile for Children (Harter, 1985). The adapted scale used a 4-point Likert-type scale ranging from 1 = not true to 4 = always true. This measure has shown evidence of reliability and validity in past research with ethnically diverse children and adolescents (Kuperminc et al., 2011). Like the original, this adapted measure assesses self-concept in five domains (Harter, 1982; 1990), including social acceptance (α = .71; for example, “It is hard for me to make friends” [reverse scored]), behavioral conduct (α = .63; for example, “I act the way I am supposed to”), scholastic competence (α = .67; for example, “I am good at my schoolwork”), and body image (α = .69; for example, “I wish my body was different”). The scale also includes a separate measure of global self-worth (α = .77; for example, “I am happy with myself”).

at PENNSYLVANIA STATE UNIV on May 8, 2016jea.sagepub.comDownloaded from

824 Journal of Early Adolescence 34(6)

Results

Preliminary Analyses

Attrition analysis. Pre- and post-test assessments were completed by 85.7% (n = 150) of the participants. To examine potential contributors to attrition, indi- viduals who dropped out were compared with those who completed the post- test using t-tests or chi-squares on all study variables. Individuals who remained in the study had higher global self-worth, t(173) = 1.99, p < .05, and were more likely to be from a two-parent home, χ2(1, N = 175) = 8.45, p < .01. There were no significant differences between the groups on any other self-concept mea- sures, social capital measures, age, grade, or living with extended family.

Missing data. The percentage of missing data in the study variables ranged from 0.6% to 14.9%. We used Little’s MCAR test to determine whether miss- ing values were missing completely at random. The test was not significant χ2(14, N = 175) = 19.30, p = .15, suggesting that all missing values were miss- ing completely at random; thus, we imputed them using the expectation max- imization (EM) algorithm as suggested by Widaman (2006).

Examination of potential covariates. We examined seven variables as potential covariates: ethnicity, age, grade level, family composition, living with extended family, length of time in Cool Girls, and participation in extracur- ricular activities. Due to the small groups in the breakdown of ethnicity, we did not examine specific ethnic differences. However, ethnicity was dummy coded in order to differentiate participants who identified as African Ameri- can (coded “1”; n = 155), and the rest of the sample (coded “0”; n = 20). This variable was included as a covariate in all analyses. A dummy code differen- tiating participants who lived with two parents (1) or in a single parent home (0) was entered in all analyses due to its importance in contributing to social capital, and because family composition was related to program participa- tion and attrition. Furthermore, during early adolescence, family composi- tion may be associated with perceptions of social support (Gayman, Turner, Cislo, & Eliassen, 2011). Participation in extracurricular activities at pre-test significantly predicted scholastic competence at post-test (β = .04, p = .03), and was therefore included as a covariate in analyses of scholastic compe- tence. No other variables predicted post-test levels of self-concept or social capital and therefore were not entered as covariates (Jaccard, Guilamo- Ramos, Johansson, & Bouris, 2006).

Mean differences in social capital by program participation. We calculated inde- pendent-samples t-tests to examine mean differences in social capital by

at PENNSYLVANIA STATE UNIV on May 8, 2016jea.sagepub.comDownloaded from

Thomason and Kuperminc 825

program participation. The t-tests were not significant for any of the pre-test measures, but reached significance for post-test measures of number of help domains, t(173) = −2.35, p = .02, network diversity t(173) = −2.37, p = .02, and overall social capital, t(173) = −2.82, p = .00. At post-test, Cool Girls (X

— = 3.74, SD = .48) reported that they could access help in more domains

than comparisons (X —

= 3.49, SD = .87). Similarly, Cool Girls (X —

= 2.13, SD = .66) reported greater network diversity than comparisons (X

— = 1.89, SD =

.66) and reported more social capital (X —

= 0.18, SD = .65) than comparisons (X

— = −0.17, SD = .97).

Correlations among variables. Correlations were obtained for all study vari- ables (see Table 1). Correlations among pre-test and post-test social capital variables were significant, suggesting moderate stability over time. Number of help domains at pre-test was significantly correlated with all pre-test self- concept variables except behavioral conduct and body image and with all post-test self-concept variables except body image. Number of domains at post-test was significantly correlated with all post-test self-concept variables.

Primary Analyses

Participation in Cool Girls and social capital. We used hierarchical multiple regression to examine the hypothesis that participation in Cool Girls contrib- utes to increases in number of help domains, network diversity, and overall social capital (Table 2). We entered pre-test social capital (number of domains, diversity of network, or total social capital) in Step 1 of the equation along with the covariates, followed by participation in Cool Girls, the pri- mary independent variable, in Step 2. The models predicting number of help domains and the diversity of participants’ network were not significant. However, the overall model predicting total social capital was significant at the trend level, suggesting that participation in Cool Girls was independently associated with a slight increase in overall social capital.

Social capital and self-concept. We used hierarchical multiple regression to examine whether reported access to social capital by post-test was associated with increases in social acceptance, behavioral conduct, scholastic compe- tence, body image, and global self-worth (see Table 3). For these analyses, we focused on overall social capital. For each model, we entered pre-test levels of self-concept, family composition, and ethnicity in Step 1, and post- test levels of total social capital in Step 2. For the model predicting scholastic competence, in extracurricular participation was also included in Step 1. As

at PENNSYLVANIA STATE UNIV on May 8, 2016jea.sagepub.comDownloaded from

826

T a b

le 1

. C

o rr

el at

io ns

A m

o ng

a ll

St ud

y V

ar ia

bl es

.

1. 2.

3. 4.

5. 6.

7. 8.

9. 10

. 11

. 12

. 13

. 14

.

1 .

So ci

al a

cc ep

ta nc

e (1

) —

.3 3

.5 4

.4 6

.4 2

.2 6

.0 3

.5 4

.2 7

.5 2

.3 2

.3 8

.0 6

− .0

5 2

. B

eh av

io ra

l c o nd

uc t

(1 )

— —

.4 5

.3 9

.1 9

.0 9

− .0

5 .1

0 .6

2 .2

6 .3

2 .0

8 .0

7 −

.0 8

3 .

G lo

ba l s

el f-

w o rt

h (1

) —

— —

.3 5

.6 3

.1 9

.1 2

.2 7

.4 0

.6 0

.2 8

.4 7

.0 8

− .0

1 4

. Sc

ho la

st ic

c o m

pe te

nc e

(1 )

— —

— —

.2 0

.2 6

.0 3

.4 6

.3 4

.2 2

.5 7

.0 6

.1 7

.0 4

5 .

B o dy

im ag

e (1

) —

— —

— —

.1 2

.0 7

.1 6

.1 6

.4 6

.1 2

.5 8

.0 9

.1 5

6 .

So ci

al c

ap it al

# d

o m

ai ns

( 1)

— —

— —

— —

.4 4

.1 9

.1 7

.1 6

.1 8

.0 2

.3 0

.1 8

7 .

So ci

al c

ap it al

d iv

er si

ty (

1) —

— —

— —

— —

.1 4

.1 0

.0 3

.0 3

− .0

3 .1

9 .2

8 8

. So

ci al

a cc

ep ta

nc e

(2 )

— —

— —

— —

— —

.3 5

.5 7

.4 7

.4 7

.2 7

.0 9

9 .

B eh

av io

ra l c

o nd

uc t

(2 )

— —

— —

— —

— —

— .3

9 .4

9 .2

4 .2

2 .0

6 10

. G

lo ba

l s el

f- w

o rt

h (2

) —

— —

— —

— —

— —

— .3

5 .6

8 .2

3 −

.0 0

11 .

Sc ho

la st

ic c

o m

pe te

nc e

(2 )

— —

— —

— —

— —

— —

— .2

3 .2

2 .0

6 12

. B

o dy

im ag

e (2

) —

— —

— —

— —

— —

— —

— .2

3 .0

0 13

. So

ci al

C ap

it al

# D

o m

ai ns

( 2)

— —

— —

— —

— —

— —

— —

— .4

2 14

. So

ci al

c ap

it al

d iv

er si

ty (

2) —

— —

— —

— —

— —

— —

— —

—

N ot

e. A

ll va

lu es

> .1

6 o r

< .1

6 ar

e si

gn ifi

ca nt

ly d

iff er

en t

fr o m

z er

o . T

he n

um be

rs 1

a nd

2 in

di ca

te e

it he

r pr

e- te

st (

1) o

r po

st -t

es t

(2 ).

at PENNSYLVANIA STATE UNIV on May 8, 2016jea.sagepub.comDownloaded from

Thomason and Kuperminc 827

shown in Table 3, social capital contributed to increases in two self-concept domains: social acceptance and behavioral conduct.

Mediation analyses. We used Preacher and Hayes’s (2004) bootstrapping method for testing indirect effects to assess the hypothesis that increases in social capital related to participation in Cool Girls would be associated with increases in self-concept. We examined mediation models only for social acceptance and behavioral conduct, given that only these dimensions of self- concept showed associations with social capital. We included pre-test levels of self-concept, family composition, and ethnicity as covariates. There was a significant indirect effect of participation on changes in social acceptance and behavioral conduct (see Table 4).

Table 2. Cool Girls Participation Predicting Increases in Social Capital From Pre- to Post-test.

b SE β ΔR2

Number in domains Step 1 .18*** Pre-test no. of domains .24 .06 .27*** Ethnicity .66 .16 .29*** Family composition .04 .10 .09 Step 2 .01 Participation in Cool Girls .13 .10 .09 Total R2 .19 Diversity of network Step 1 .13*** Pre-test diversity .23 .06 .26*** Ethnicity .32 .16 .15* Family composition −.13 .10 −.09 Step 2 .01 Participation in Cool Girls .15 .10 .12 Total R2 .14 Total social capital Step 1 .20*** Pre-test social capital .26 .06 .30*** Ethnicity .70 .19 .26*** Family composition −.06 .12 −.04 Step 2 .01†

Participation in Cool Girls .20 .12 .12† Total R2 .21

†p < .10. *p < .05. **p < .01. ***p < .001.

at PENNSYLVANIA STATE UNIV on May 8, 2016jea.sagepub.comDownloaded from

828

T a b

le 3

. So

ci al

C ap

it al

P re

di ct

in g

In cr

ea se

s in

S el

f- C

o nc

ep t.

So ci

al a

cc ep

ta nc

e B

eh av

io ra

l c o nd

uc t

Sc ho

la st

ic c

o m

pe te

nc e

B o dy

im ag

e G

lo ba

l s el

f- w

o rt

h

b( SE

) β

Δ R

2 b(

SE )

β Δ

R 2

b( SE

) β

Δ R

2 b(

SE )

β Δ

R 2

b( SE

) β

Δ R

2

St ep

1. .3

3* **

.3 8*

** .3

7* **

.3 6*

** .4

1* **

Pr

e- te

st s

el f-

co nc

ep t

.4 7(

.0 6)

.5 2*

** .5

4( .0

5) .6

1* **

.4 8(

.0 6)

** *

.5 6(

.0 6)

.5 7*

** .5

3( .0

5) .6

1* **

Et

hn ic

it y

.1 8(

.1 2)

.1 1

− .0

2( .1

2) −

.0 1

.2 2(

.1 0)

.1 4*

.2 3(

.1 2)

.1 3†

.3 1(

.1 0)

.2 0*

*

Fa

m ily

co

m po

si ti o n

.0 5(

.0 7)

.0 5

.0 7(

.0 7)

.0 6

.1 3(

.0 6)

.1 4*

.0 2(

.0 7)

.0 2

.0 6(

.0 6)

.0 6

Ex

tr a-

cu rr

ic ul

ar

ac ti vi

ti es

.0 7(

.0 3)

.1 4*

St ep

2 .

.0 2*

.0 2*

.0 1

.0 0

.0 0

Po

st -t

es t

so ci

al

ca pi

ta l

.1 3(

.0 6)

.1 5*

.0 9(

.0 4)

.1 4*

.0 4(

.0 4)

.0 7

.0 1(

.0 4)

.0 2

.0 3(

.0 4)

.0 6

T

o ta

l R 2

.3 5*

.4 0*

.3 8

.3 6

.4 1

† p <

.1 0.

* *p

< .0

1. *

** p

< .0

01 .

at PENNSYLVANIA STATE UNIV on May 8, 2016jea.sagepub.comDownloaded from

Thomason and Kuperminc 829

Discussion

In the present study, we examined the process by which participation in Cool Girls contributed to the development of self-concept via a focus on increasing participants’ access to social capital. As hypothesized, we found (a) that pro- gram participation was associated with increases in access to social capital (though only at a trend level when family composition and ethnicity were controlled), (b) that program-related increases in social capital would con- tribute to increases in self-concept, and (c) that social capital would mediate associations between program participation and indices of self-concept.

The finding that, regardless of participation in Cool Girls, girls who had increases in social capital from pre-test to post-test would experience increases in self-concept, points to the importance of social capital for promoting the development of self-concept especially in the areas of social acceptance and behavioral conduct. This is consistent with prior research documenting a link between social support and self-concept (Arslan, 2009; Johnson, 1996).

Mediation analyses focused on examining whether social capital–mediated associations between participation in Cool Girls and increases in self-concept. These results are particularly relevant for Cool Girls because they reveal a pro- cess by which the program may contribute to potential increases in participants’ self-concept. The results indicate that support and guidance, help with school- work, problem-solving, and goal-setting are critical areas in which having a person to go to may make a significant difference in self-concept.

Interestingly, the only two types of self-concept that were related to pro- gram participation and social capital were social acceptance and behavioral

Table 4. Indirect Effects of Participation on Self-Concept.

b(SE) CI

Social acceptance Total social capital .03(.02) [.00, .09] Behavioral conduct Total social capital .02(.02) [.00, .07] Scholastic competence Total social capital .01(.01) [−.01, .03] Body image Total social capital .00(.01) [−.02, .03] Global self-worth Total social capital .01(.01) [−.00, .04]

Note. Confidence intervals (CIs) that do not cross zero are considered to be significantly different from zero.

at PENNSYLVANIA STATE UNIV on May 8, 2016jea.sagepub.comDownloaded from

830 Journal of Early Adolescence 34(6)

conduct. The finding that having a diverse network of people to go to for help in multiple areas of one’s life contributed to increased feelings of social acceptance is consistent with past findings that perceived support buffers the psychological stress associated with rejection from peers (Beeri & Lev- Weisel, 2012). The finding that social capital was related to perceived behav- ioral conduct is consistent with theorizing that social capital can reinforce social norms among youth about appropriate behavior (Coleman, 1988b).

Social capital was not associated with any of the other domains of self- concept, including global self-worth, scholastic competence, or body image. It could be that certain aspects of social capital that we did not measure in this study would have been more relevant to these domains. For example, feelings of self-worth and body image are closely tied, and may be more related to the quality of relationships within one’s network. Programmatic activities designed to engage young people with peers and adults to raise awareness about issues like unrealistic body ideals that can negatively affect youth’s body image and overall self-esteem (Steiner-Adair et al., 2002) may be one way that programs can help enrich social capital networks and use those net- works to empower youth. Furthermore, ethnic identity is closely tied to feel- ings of self-worth for ethnic minority populations, including African American females, who made up the majority of the sample. For example, Gaylord- Harden, Ragsdale, Mandara, Richards, and Petersen (2007) found an indirect effect of perceived social support on a reduction in depression through posi- tive racial identity. When studying self-esteem or self-concept in this popula- tion, future research should include measures of ethnic identity. Scholastic competence may be more related to social capital specifically designed to aid in social mobility and achievement, such as tutors and school personnel who can teach youth how to apply for college and expose youth to career paths. Although participants in our study were asked about having someone to go to for help with schoolwork, we did not specifically focus on this domain and its contribution to feelings of academic competence.

Finally, there is some debate in the literature about the cultural validity of the Self-Perception Profile for Children (Harter, 1985) and with other similar measures. Some studies have failed to find support for validity of the measure when utilized in a sample of African American female adolescents (Stewart, Roberts, & Kim, 2010). Others have found similar factor structures across eth- nic groups (including the ones in this study), although not strict levels of mea- surement equivalence (Michaels, Barr, Roosa, & Knight, 2007), suggesting that while such measures may assess self-concept in similar ways, the absolute scores may not be comparable across groups. Thus, while not ideal, the mea- sures of self-concept used in this study appear to demonstrate sufficient validity given that cross-ethnic comparisons were not the focus of the current study.

at PENNSYLVANIA STATE UNIV on May 8, 2016jea.sagepub.comDownloaded from

Thomason and Kuperminc 831

Limitations and Future Directions

Our ability to measure long-term programmatic effects was limited by the study’s pre-test, post-test design. According to Cole and Maxwell (2003), measuring mediation over time with only two time points can lead to mis- leading results due to an inability to accurately measure the stationarity of the variables. The addition of a follow-up assessment 6 months or a year beyond the post-test would have strengthened the study and allowed for better assess- ment of trajectories of change over time and longer term program effects.

In addition, the measurement of social capital was limited in several ways. In keeping with the program’s theory of change, survey items were devel- oped to assess the extent to which girls were able to access needed help in domains directly addressed by Cool Girls programs: problem-solving, help with schoolwork, and goal-setting. The fourth domain, general support and guidance, was worded slightly differently, in order to assess whether partici- pants had an adult natural mentor. For that item, the possible responses were limited to adults, whereas respondents were free to report friends or peers for the other three domains.

Furthermore, two important aspects of social capital were not measured in this study: the quality or helpfulness of social resource (Ferguson, 2006) and intergenerational closure or the extent to which members of an individual’s network know one another (Coleman, 1988b; Glanville et al., 2008; Kao & Rutherford, 2007). It could be argued that being able to report someone you can go to for help is meaningless unless that person actually is able to deliver the help you need. Intergenerational closure may be an important aspect of social capital that can address specific dimensions of self-concept such as behavioral conduct. The measures developed for this study offered some insight into how participation in a YDP might contribute to expanded resources for accessing help in solving everyday problems; however, more refined measures that reliably capture additional aspects of social capital from youth’s point of view are clearly needed. Other methods, such as quali- tative interviews with parents and youth participants might also provide valu- able information about the structure of social networks that exist between parents and other adults in the girls’ lives, such as teachers, program staff, and volunteers. The use of social network analysis (e.g., Blakeslee, 2012) can also provide richer information about how programs such as Cool Girls can influence youngsters’ access to social resources.

This study is also limited in its ability to account for the extent to which individual girls were exposed to the full Cool Girls program (e.g., “dosage”), as attendance data were not available. In addition to obtaining program

at PENNSYLVANIA STATE UNIV on May 8, 2016jea.sagepub.comDownloaded from

832 Journal of Early Adolescence 34(6)

records of participant attendance, subsequent evaluation surveys have included a self-report question asking each girl how often she attends Cool Girls Club. However, a strength of this study is that we were able to measure the extent to which girls participants were involved in a range of extracur- ricular activities and control for this potential confound, isolating the contri- butions of Cool Girls participation.

Implications

The results of this study have several implications for YDPs for girls in urban settings. YDPs like Cool Girls, Inc. seek to empower youth by promoting inter- nal and external resources. The present study found evidence that Cool Girls is promoting external resources, as program participation was linked to modest increases in social capital. Cool Girls, Inc. brings girls together, attracts com- munity volunteers to help support their program, and teaches participants how to seek out social support when needed. Programs that seek to increase the available social capital of their participants may look to the techniques of Cool Girls as an example. In addition, regardless of participation in Cool Girls, girls who experienced increases in social capital also experienced increases in their sense of social acceptance from peers and their behavioral conduct. Although self-concept was not directly linked to participation in Cool Girls, Inc., the findings suggest that a programmatic focus on helping youth develop and expand their social capital resources may offer an important pathway toward achieving valued youth development outcomes such as positive self-concepts. Youth who experience positive self-concept, in turn, are more likely to achieve academically, have hope for the future, and avoid risk-behaviors (Cedeno, Elias, Kelly, & Chu, 2010; Huang, 2011; Peterson, Buser, & Westburg, 2010). The study of social capital networks in the context of YDPs is an important direction in understanding how such programs play a role in promoting posi- tive developmental outcomes, particularly for adolescent girls growing up in socially and economically disadvantaged communities.

Acknowledgment

Thanks to Lawanda Cummings, Duane House, and Natalie McCoy Wilkins for their feedback on previous versions of this manuscript.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

at PENNSYLVANIA STATE UNIV on May 8, 2016jea.sagepub.comDownloaded from

Thomason and Kuperminc 833

Funding

The author(s) received no financial support for the research, authorship, and/or publi- cation of this article.

References

Arslan, C. (2009). Anger, self-esteem, and perceived social support in adolescence. Social Behavior and Personality, 37, 555-564.

Bearman, S., & Stice, E. (2008). Testing a gender additive model: The role of body image in adolescent depression. Journal of Abnormal Child Psychology, 36, 1251-1263.

Beeri, A., & Lev-Weisel, R. (2012). Social rejection by peers: a risk factor for psycho- logical distress. Child and Adolescent Mental Health, 17, 216-221.

Blakeslee, J. (2012). Expanding the scope of research with transition-age foster youth: Applications of the social network perspective. Child & Family Social Work, 17, 326-336.

Bourdieu, P. (1983). The forms of capital. Soziale Ungleichheiten, 2, 183-198. Bronfenbrenner, U. (1979). The ecology of human development: Experiments by

nature and design. Cambridge, MA: Harvard University Press. Cambron, M., Acitelli, L., & Pettit, J. (2009). Explaining gender differences in

depression: An interpersonal contingent self-esteem perspective. Sex Roles, 61, 751-761.

Cedeno, L., Elias, M., Kelly, S., & Chu, B. (2010). School violence, adjustment, and the influence of hope on low-income, African American youth. American Journal of Orthopsychiatry, 80, 213-226.

Cole, D., & Maxwell, S. (2003). Testing mediational models with longitudinal data: Questions and tips in the use of structural equation modeling. Journal of Abnormal Psychology, 112, 558-577.

Coleman, J. (1988a). The creation and destruction of social capital: Implications for the law. Journal of Law, Ethics, & Public Policy, 3, 375-404.

Coleman, J. (1988b). Social capital in the creation of human capital. The American Journal of Sociology, 94, 95-120.

Cool Girls, Inc. (2008). Dedicated to the self-empowerment of girls. Retrieved from http://www.thecoolgirls.org/about/cool-girls-mission/

Curran, E. (2007). The relationship between social capital and substance use by high school students. Journal of Alcohol & Drug Education, 51, 59-73.

Demaray, M., Malecki, C., Rueger, S., Brown, S., & Summers, K. (2009). The role of youth’s ratings of the importance of socially supportive behaviors in the relation- ship between social support and self-concept. Journal of Youth and Adolescence, 38, 13-28.

Eccles, J., & Gootman, J. (2002). Community programs to promote youth develop- ment. Washington, DC: National Academy Press.

Else-Quest, N., Hyde, J., & Linn, M. (2010). Cross-national patterns of gender dif- ferences in mathematics: A meta-analysis. Psychological Bulletin, 136, 103-127.

at PENNSYLVANIA STATE UNIV on May 8, 2016jea.sagepub.comDownloaded from

834 Journal of Early Adolescence 34(6)

Ferguson, K. (2006). Social capital and children’s wellbeing: A critical synthesis of the international social capital literature. International Journal of Social Welfare, 15, 2-18.

Fitzpatrick, K., Wright, D., Bettina, F., & LaGory, M. (2005). Depressive symp- tomatology, exposure to violence, and the role of social capital among African American adolescents. American Journal of Orthopsychiatry, 75, 262-274.

Gaylord-Harden, N. K., Ragsdale, B. L., Mandara, J., Richards, M. H., & Petersen, A. C. (2007). Perceived support and internalizing symptoms in African American adolescents: Self-esteem and ethnic identity as mediators. Journal of Youth and Adolescence, 36, 77-88.

Gayman, M., Turner, R., Cislo, A., & Eliassen, A. (2011). Early adolescent family experiences and perceived social support in young adulthood. Journal of Early Adolescence, 31, 880-908.

Glanville, J., Sikkink, D., & Hernandez, E. (2008). Religious involvement and educa- tional outcomes: The role of social capital and extracurricular participation. The Sociological Quarterly, 49, 105-137.

Groninger, R., & Lee, V. (2001). Social capital and dropping out of high school: Benefits to at-risk students of teachers’ support and guidance. Teachers College Record, 103, 548-581.

Harter, S. (1982). The Perceived Competence Scale for Children. Child Development, 53, 87-97.

Harter, S. (1985). The self-perception profile for children: Revision of the perceived competence scale for children. Denver, CO: University of Denver.

Harter, S. (1999). The construction of the self: A developmental perspective. New York, NY: Guilford.

Harter, S., & Monsour, A. (1992). Developmental analysis of conflict caused by opposing attributes in the adolescent self-portrait. Developmental Psychology, 28, 251-260.

Huang, C. (2011). Self-concept and academic achievement: A meta-analysis of longi- tudinal relations. Journal of School Psychology, 49, 505-528.

Huguet, P., & Régner, I. (2009). Counter-stereotypic beliefs in math do not protect school girls from stereotype threat. Journal of Experimental Social Psychology, 45, 1024-1027.

Hurd, N. M., & Sellers, R. M. (2013). Black adolescents’ relationships with natu- ral mentors: Associations with academic engagement via social and emotional development. Cultural Diversity & Ethnic Minority Psychology, 19, 76-85.

Jaccard, J., Guilamo-Ramos, V., Johansson, M., & Bouris, A. (2006). Multiple regres- sion analyses in clinical child and adolescent psychology. Journal of Clinical Child & Adolescent Psychology, 35, 456-479.

Jarrett, R., Sullivan, P., & Watkins, N. (2005). Developing social capital through par- ticipation in organized youth development programs: Qualitative insights from three programs. Journal of Community Psychology, 33, 41-55.

Johnson, E. (1996). The role of social support and gender orientation in adoles- cent female self-development (Doctoral dissertation). Dissertation Abstracts International: Section B. The Sciences and Engineering, 56, 5196.

at PENNSYLVANIA STATE UNIV on May 8, 2016jea.sagepub.comDownloaded from

Thomason and Kuperminc 835

Jones, D. (2004). Body image among adolescent girls and boys: A longitudinal study. Developmental Psychology, 40, 823-835.

Kao, G., & Rutherford, L. (2007). Does social capital still matter? Immigrant minor- ity disadvantage in school-specific social capital and its effects on academic achievement. Sociological Perspective, 50, 27-52.

Kuperminc, G., Thomason, J., DiMeo, M., & Broomfield-Massey, K. (2011). Cool Girls, Inc.: Promoting the positive development of urban pre-adolescent and early adolescent girls. Journal of Primary Prevention, 32, 171-183.

LeCroy, C. (2004). Experimental evaluation of “Go Grrrls” preventive intervention for early adolescent girls. Journal of Primary Prevention, 25, 457-473.

Li, J., & Julian, M. M. (2012). Developmental relationships as the active ingredient: A unifying working hypothesis of “what works” across intervention settings. The American Journal of Orthopsychiatry, 82, 157-166.

McCabe, M., & Ricciardelli, L. (2003). Sociocultural influences on body image and body changes among adolescent boys and girls. The Journal of Social Psychology, 143, 5-26.

McElhaney, K.B., Antonishak, J., & Allen, J.P. (2008). “They Like Me, They Like Me Not”: Popularity and Adolescents’ Perceptions of Acceptance Predicting Social Functioning Over Time. Child Development, 79, 720-731.

Mekinda, M. A., & Hirsch, B. J. (in press). After-school programs. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 221-232). Thousand Oaks, CA: Sage.

Michaels, M., Barr, A., Roosa, M., & Knight, G. (2007). Self-esteem: Assessing measurement equivalence in a multiethnic sample of youth. Journal of Early Adolescence, 27, 269-295.

Miller, J., & White, N. (2003). Gender and adolescent relationship violence: A con- textual examination. Criminology, 41, 1207-1249.

Nishina, A., Ammon, N., Bellmomre, A., & Graham, S. (2006). Body dissatisfaction and physical development among ethnic minority adolescents. Journal of Youth and Adolescence, 35, 189-201.

Peterson, C., Buser, T., & Westburg, N. (2010). Effects of familial attachment, social support, involvement, and self-esteem on youth substance use and sexual risk taking. The Family Journal, 18, 369-376.

Peterson, S.H., Wingood, G.M, DiClemente, R.J., Harrington, K., & Davies, S. (2007). Images of Sexual Stereotypes in Rap Videos and the Health of African American Female Adolescents. Journal of Women’s Health, 16, 1157-1164.

Preacher, K., & Hayes, A. (2004). SPSS and SAS procedures for estimating the indirect effects in simple mediation models. Behavior Research Methods, Instruments, & Computers, 36, 717-731.

Rosenfeld, L., Richman, J., & Bowen, G. (1998). Low social support among at-risk adolescents. Social Work in Education, 20, 245-260.

Roth, J., & Brooks-Gunn, J. (2003). What Exactly is a Youth Development Program? Answers from Research and Practice. Applied Developmental Science, 7, 94-111.

Royse, D. (1998). Scouting and Girl Scout curriculum as interventions: Effects on adolescents’ self-esteem. Adolescence, 33, 159-169.

at PENNSYLVANIA STATE UNIV on May 8, 2016jea.sagepub.comDownloaded from

836 Journal of Early Adolescence 34(6)

Skelton, C., Francis, B., & Read, B. (2010). “Brains before ‘beauty’?” High achieving girls, school and gender identities. Educational Studies, 36, 185-194.

Stanton-Salazar, R. (1997). A social capital framework for understanding the social- ization of racial minority children and youths. Harvard Educational Review, 67, 1-40.

Stanton-Salazar, R. (2001). Manufacturing hope and despair: The school and kin sup- port networks of U.S.-Mexican Youth. New York, NY: Teachers College Press.

Stanton-Salazar, R. (2005). Adolescent peer networks as a context for social and emo- tional support. Youth & Society, 36, 379-417.

Steiner-Adair, C., Sjostrom, L., Franko, D., Seeta, P., Tucker, R., Becker, A., & Herzog, D. (2002). Primary prevention of risk factors for eating disorders in ado- lescent girls: Learning from practice. International Journal of Eating Disorders, 32, 401-411.

Stewart, P., Roberts, M., & Kim, K. (2010). The psychometric properties of the Harter self-perception profile for children with at-risk African American females. Journal of Child and Family Studies, 19, 326-333.

Sullivan, R., & Larson, R. (2010). Connecting youth to high-resource adults: Lessons from effective youth programs. Journal of Adolescent Research, 25, 99-123.

Waldrip, A.M., Malcom, K.T., & Jensen-Campbell, L.A. (2008). With a Little Help from Your Friends: the Importance of High-quality Friendships on Early Adolescent Adjustment. Social Development, 17, 832-852.

Wall, E., Ferrazzi, G., & Schryer, F. (1998). Getting the goods on social capital. Rural Sociology, 63, 300-322.

Widaman, K. F. (2006). Missing data: What to do with or without them. In R. Bakeman, J. M. Gottman, D. Brewer, K. Bub, M. Burchinal, F. P. Graham, & K. McCartney (Eds.), Best practices in quantitative methods for developmentalists (pp. 42-65). New York, NY: Wiley-Blackwell.

Zeldin, S., Christens, B. D., & Powers, J. (2013). The psychology and practice of youth-adult partnership: Bridging generations for youth development and com- munity change. American Journal of Community Psychology, 51, 385-397.

Zimmerman, M., Copeland, L., Shope, J., & Dielman, T. (1997). A longitudinal study of self-esteem: Implications for adolescent development. Journal of Youth and Adolescence, 26, 117-141.

Author Biographies

Jessica D. Thomason is a doctoral candidate in the community psychology program at Georgia State University. Her research focuses on positive youth development, specifically community initiatives to promote adolescent well-being and sexual health.

Gabriel Kuperminc, professor of psychology at Georgia State University, directs the doctoral program in community psychology and serves on the faculty of the develop- mental psychology program. His research focuses on resilience and positive youth development, particularly among culturally diverse and economically disadvantaged youth.

at PENNSYLVANIA STATE UNIV on May 8, 2016jea.sagepub.comDownloaded from