Due in 5 and half hrs Journal Article review no outside sources other than the article itself/ Sociology of Sports Terminology must be used. Enclosed as well

Leisure Sciences, 32: 299–317, 2010 Copyright C© Taylor & Francis Group, LLC ISSN: 0149-0400 print / 1521-0588 online DOI: 10.1080/01490400.2010.488189

The Impact of Socialization on Youth Program Outcomes: A Social Development Model Perspective

MAT D. DUERDEN PETER A. WITT

Department of Recreation, Park and Tourism Sciences Texas A&M University College Station, TX, USA

This study employs a mixed-methods design and a social development model (SDM) to examine the role of socialization processes within an international immersion program for adolescents. Longitudinal data from 108 participant and 49 comparison group members are analyzed using structural equation modeling procedures. Qualitative data are also used to assess participants’ perceptions of these processes. The quantitative findings indicate that the SDM fits the data well and partially mediates the growth of observed recreation program outcomes. The qualitative findings offer additional insights into the role of within program socialization processes.

Keywords mixed-methods, structured recreation settings, youth development

The individuals with whom adolescents interact across the contexts of their lives exert a powerful developmental influence. These people include parents (Baumrind, 1991), peers (Hartup, 1996), teachers (Hughes, Cavell, & Willson, 2001; Hughes & Kwok, 2007), and other nonparental adults including mentors (Beier, Rosenfeld, Spitalny, Zansky, & Bontempo, 2000). Although interpersonal relationships and socialization play major de- velopmental roles during adolescence, their influence is often overlooked when evaluating the impact of programs offered by youth serving agencies (Grossman & Bulle, 2006) and recreation organizations.

The social developmental model (SDM) provides a theoretical approach to understand- ing the impact of relationships and socialization on behavior (Catalano & Hawkins, 1996; Hawkins & Weis, 1985). The model posits that individuals develop bonds to groups and organizations when they experience opportunities for involvement, possess the skills neces- sary for involvement, and receive positive feedback regarding their involvement (Hawkins, Catalano, & Arthur, 2002). Bonding influences individuals to adopt and act in accordance with the norms and expectations of the group (Catalano & Hawkins). The SDM has proven efficacious in a variety of settings (Catalano, Haggerty, Oesterle, Fleming, & Hawkins, 2004; Catalano, Oxford, Harachi, Abbott, & Haggerty, 1999) but has yet to be tested in a traditional recreation context such as an out-of-school-time program.

Therefore, the purpose of this study was to employ a SDM to assess the mediating influ- ence of socializing processes on outcomes associated with participation in an international immersion/environmental education program for middle and high school-aged youth. The

Received 24 June 2009; accepted 4 December 2009. Address correspondence to Mat D. Duerden, Department of Recreation, Park and Tourism Sciences, 2261

TAMU, Texas A&M University, College Station, TX 77843. E-mail: [email protected]

299

300 M. D. Duerden and P. A. Witt

findings from this study provide insights regarding the influence of interpersonal processes on program outcomes and represent a positive youth development application of the SDM in an out-of-school-time program context.

Literature Review

The SDM (see Figure 1) resulted from efforts to explain the origins and processes asso- ciated with adolescent deviance (Hawkins & Weis, 1985). Social learning theory, social control theory, and differential association theory were used to create a framework to ex- plain both deviant and prosocial behavior from a social development perspective. Social learning theory suggests that behaviors, especially repeated behaviors, result in part from positive reinforcement (Akers, 1977; Bandura, 1977). Social control theory highlights the importance of bonds to socializing units (e.g., family, school, peers) in the development of behavior (Hirschi, 1969). Differential association theory proposes that both deviant and prosocial behaviors share similar developmental pathways (Matsueda, 1982).

The SDM identifies key constructs that influence the development of social bonds and beliefs in societal norms which in turn affect behavior (Catalano & Hawkins, 1996). For socialization to occur, youths need to be aware of opportunities to become involved within a socializing unit and subsequently engage in these opportunities. To successfully participate in an activity, adolescents also need to possess and apply appropriate skills. Actual involvement and skill levels influence the type and degree of reinforcement an individual receives. Perceived positive reinforcement leads individuals to form social bonds. These bonds consist of attachment to the socializing unit as well as a commitment to act according to the unit’s associated beliefs and norms (Catalano & Hawkins, 1996). This commitment, in turn, influences future behavior.

The SDM has been effective in predicting negative behaviors. In a study of alcohol misuse at ages 14 and 16, the SDM explained 45% of the variance in alcohol misuse at age 16 and significantly mediated the relation between age 14 and 16 drinking (Lonczak et al., 2001). In a study to address antisocial behavior among elementary school children, the SDM was able to explain 25–35% of the variance in behavior for children whose parents modeled deviant behaviors and those who did not (Catalano et al., 1999). Similar findings have also shown that children’s drug use is significantly related to parents’ drug use (Fleming,

Perceived Rewards

Skills for Involvement

Social Bonding

Belief in Social Norms

Behavior

Perceived Opportunities

FIGURE 1 Social development model (adapted from Hawkins & Weis, 1985).

The Impact of Socialization 301

Perceived Rewards

Skills for Involvement

Bonding within GEx

Belief in GEx

Norms

Program Outcomes

Involvement in GEx

FIGURE 2 Proposed conceptual model.

Brewer, Gainey, Haggerty, & Catalano, 1997). Additional research has also validated the effectiveness of interventions based upon the SDM to promote positive behavior (Catalano et al., 2004).

Although untested in recreational contexts, the SDM has the potential to provide insights into the role of social processes in youth recreation programs and their impact on targeted outcomes. Researchers have addressed important socialization related issues, including the role of autonomy (Hill & Sibthorp, 2006), identity development (Shaw, Kleiber, & Caldwell, 1995), and intrinsic motivation (Kleiber, Larson, & Csikszentmihalyi, 1986) in recreation settings, but additional work specifically focused on socialization is needed. Structured leisure activities clearly offer youth access to the socializing influence of peer groups (Barber, Stone, Hunt, & Eccles, 2005), but the exact mechanisms behind this socialization remain unclear. For example, Persson, Kerr, and Stattin (2007) found that older adolescents were more likely to continue participating in structured recreation programs in which their friends were involved but called for further research on the role of bonding to better understand these participation patterns. The SDM provides a tested theoretical lens to address the components, processes, and impacts of socialization as they pertain to youth recreation program contexts. Therefore, this study tested the mediating affect of a conceptual SDM (see Figure 2) on outcomes associated with an international immersion/environmental education program. The following hypotheses were tested:

1. Program participants will experience a significantly greater increase in self-reported levels of program outcomes in comparison to the controls.

2. The conceptual SDM will partially mediate the development of program outcomes that occur during both the preparatory (T2) and international workshop (T3) portions of the program.

The qualitative portion of the study was guided by the following research questions:

1. From the participants’ perspective, what role did socialization processes play in the overall program experience?

2. What influence did these processes have across the program components?

Methods

The study employed a quasi-experimental, concurrent nested mixed-method design (Hanson, Creswell, Clark, Petska, & Creswell, 2005) to address the hypotheses and re- search questions. This design involved the simultaneous collection and analysis of both

302 M. D. Duerden and P. A. Witt

qualitative and quantitative data. Emphasis was given to the quantitative data and hypothe- ses and the qualitative data were used to gain additional insights.

Program Description

The program studied was offered by Global Explorers (GEx), a nonprofit organization that provides international immersion experiences for middle school and high school students and teachers. GEx programs focus on four core disciplines (i.e., science, culture, leadership, and service) with the overall goal of helping students develop into responsible global citizens (Global Explorers, 2008). Each program includes three stages: a preparatory program, an international field workshop, and a post-trip service project. During the preparatory program youth participate in 9–12 sessions, which take place during the after-school free-time portion of the student’s day, ranging in length from one to three hours specific to each groups’ travel destination.

The international field workshop lasts between 7 and 14 days. Each group consists of students, teachers, and optional adult chaperones. All aspects of the international field workshop are arranged and supervised by GEx staff. In addition to local guides, GEx provides each group with a volunteer field scientist and a GEx staff member. During this portion of the program, students and teachers take part in cultural, scientific, and service activities led by GEx staff and local contracted guide services. Locations include Peru, Costa Rica, and Tanzania. Upon returning from the field workshop, participants design and implement a service project directed either toward the needs of their own community or the international community they visited.

Population

Data for this study were collected from seven different groups who traveled with GEx during 2008. Youth in the program were recruited by teachers at each participating school. For our study, each participating teacher was asked by the researchers to recruit students to serve as members of a nonequivalent comparison group (Babbie, 2005). Parental consent and student assent were obtained for all students involved in the study.

The participant group consisted of 108 students (females = 51; males = 57) from seven different groups and 49 students (females = 29; males = 20) who served as comparisons. We originally hoped to collect data from a larger portion of the 215 youth who participated in a GEx program during 2008, but a number of the groups either declined participation or had low response rates primarily due to lack of support for the evaluation from some teachers. In addition, only three of the seven teachers were able to recruit comparison group students. At the beginning of the study, participating and comparison students had a mean age of 14.5 years (SD = 1.65) and 13.6 (SD = .89) respectively. Eighty-two percent of the participants and 90% of the comparisons were White.

Quantitative Methodology

Data collection. Several procedures were employed to collect questionnaire data from the participant and comparison groups. At the completion of the preparatory program (T2), participants completed a questionnaire containing both traditional and retrospective pretest items. The traditional items addressed issues related to SDM components (e.g., bonding, opportunities for involvement) within the preparatory program. The retrospective pretest items assessed pre (T1) and post preparatory program (T2) levels of self-reported program outcome variables (i.e., environmental knowledge, attitude, and behavior). Students in the

The Impact of Socialization 303

comparison group also completed the retrospective pretest items during approximately the same timeframe as their participating counterparts.

Retrospective pretests were employed due to both logistical limitations that did not allow for data collection before all groups began their participation and to guard against self- report bias. Retrospective pretests occurred at the conclusion of the preparatory program and required respondents to indicate their current perception of the degree to which they possessed a specific trait, attitude, or attribute previous to their participation in the prepara- tory program (Sibthorp, Paisley, Gookin, & Ward, 2007). Use of this approach guarded against response-shift bias, which occurs between pre- and posttests when individuals’ internal scale of measurement changes as a result of an experience (Pratt, McGuigan, & Katzev, 2000; Sibthorp et al., 2007). For example, youth participants might rate themselves high on a pretest skills inventory as a result of inaccurate perceptions of the difficulty of the required tasks. After completing the tasks, even though individuals gained a greater degree of competence from their experience, they might rate themselves lower on the posttest than the pretest due to a more accurate perception of task difficulty.

After completing the international field workshop (T3), participants completed all items from the T2 questionnaire to provide information regarding the impact of this program component. To follow up with both the participant and comparison groups, a final round of data collection occurred during the fall of 2008 (T4). T4 data collection took place after the groups had completed their posttrip service projects. Thus, time between posttravel and follow-up data collection periods ranged from three to seven months. Because logistical difficulties associated with collecting data from the comparison group during the summer, data were only gathered from the comparisons at T1, T2, and T4.

Table 1 contains a breakdown of the responses collected at each of four data collection periods. Although the number of questionnaires collected across the first three data collec- tion occasions remained static, attrition occurred at T4 despite efforts to maintain a high response rates.

An attrition analysis was conducted to identify potential differences between those individuals with and without complete data. One-way ANOVA statistics revealed no signif- icant differences within both the participant and comparison groups. The assumption that the data were missing at random was supported by these findings. This finding, along with the low rate of missing data (i.e., < 5%) from individuals who completed at least a portion of the survey at each time wave, provided justification for imputing some of the missing data. Imputation was conducted using the LISREL 8.8 multiple imputation procedure to address missing values at each time wave for individuals who completed at least some portion of the questionnaire. Data were not imputed if no response was collected from an individual for a particular wave of data collection.

Measures. Items adapted from the Seattle Social Development Project (Hawkins et al., 2003) were used to measure opportunities for involvement, skills for involvement, rewards and recognition, and bonding at T2 and T3. Sample items for the various scales included statements such as: “I had lots of chances to participate in GEx activities” (opportunities

TABLE 1 Participant and Comparison Group Data Collection Overview

Group T1 T2 T3 T4

Participants 106 106 108 75 Comparisons 49 49 — 30

304 M. D. Duerden and P. A. Witt

for involvement), “I had difficulty following directions during GEx activities” (skills), “My GEx teachers praised or complimented me when I worked hard” (rewards), and “I liked the GEx program” (bonding). These scales have produced adequate estimates of internal consistency in previous studies (e.g., .64 for opportunities for involvement, .68 for rewards and recognition, and .76 for bonding; personal communication Karl G. Hill, October 25, 2007). No statistics were available from previous research for the skills scale. A six-item scale to measure GEx students’ beliefs about service, science, culture, and leadership (e.g., “I believe that learning about science can help us reduce our impact on the environment”) was developed by the authors after a review of GEx curriculum and was evaluated for content validity by GEx administrators.

Environmental attitudes (EA) and pro-environmental behaviors (EB) were measured using the affect and behavioral commitment subscales from the Children’s Environmental Attitude and Knowledge Scale (CHEAKS, Leeming, & Dwyer, 1995). These subscales consisted of 12 items each. The EA subscale contained statements such as: “I get angry about the damage pollution does to the environment” and “I am frightened to think people don’t care about the environment.” Items from the EB subscale included statements such as: “I have asked my family to recycle some of the things that we use” and “I do not let a water faucet run when it is not necessary.” Previous testing of the CHEAKS subscale from which this study’s EA and EB measures were drawn suggested acceptable levels of reliability and validity (Leeming & Dwyer, 1995). Finally, a five-item scale (e.g., “I can explain what the term ecology means”) to measure environmental knowledge (EK) was developed by the authors through a review of GEx curriculum and was evaluated for content validity by GEx administrators. All items employed in this study were assessed using a 5-point Likert response format (1 = very untrue to 5 = very true). All measures produced adequate levels of internal consistency (Table 2).

Analysis procedures. To guard against a family-wise error rate that would have arisen if multiple analyses had been conducted for each of the program outcomes, and due to the study’s relatively low sample size, a composite outcome score (KAB) was created by taking the mean of the EK, EA, and EB mean scores at each time period. The latent constructs associated with these scales were theoretically linked and provided further justification for this combination. Providing empirical support for the creation of a composite score, the EK, EA, and EB scores were significantly correlated (.30 to .58) across all time periods and the new KAB variable produced adequate reliability coefficients (T1 = .68; T2 = .68;

TABLE 2 Reliability Coefficients for all Measures

Alpha Coefficients

Pre- Post Post Follow- Scale Program Preparatory Travel up

Opportunities — .67 .61 — Skills — .75 .79 — Rewards — .75 .70 — Bonding — .75 .87 — Beliefs — .63 .61 — Environmental knowledge .78 .83 .77 .88 Environmental attitude .85 .85 .84 .86 Environmental behavior .75 .71 .65 .75

The Impact of Socialization 305

T3 = .71; and T4 = .74). Accordingly, the KAB as a more parsimonious representation of the program’s targeted outcomes was employed in the study’s analyses.

The first hypothesis, that the participants would experience a significantly greater increase in program outcomes in comparison to the controls, was tested using a repeated measure, ANOVA. Covariance structure analyses using LISREL 8.80 were employed to test the second hypothesis regarding the mediating role of the SDM on the development of pro-environmental behavior. Two separate structural test models were analyzed with one for the preparatory and one for the international workshop portions of the program. The analyses of two separate models allowed for the investigation of the contribution of the socializing processes from each program context to the overall development of outcomes.

The first model included KAB measured at T1 and T2 and all T2 SDM variables. The second model included KAB measured at T2 and T3 as well as all T3 SDM variables. Several analysis adaptations were necessary due to the relatively small sample size. Item parceling was used to create the “observed” model variables as opposed to employing confirmatory factor analysis (CFA) to test and refine latent variables created from multiple scale items (Kishton & Widaman, 1994). Although the use of CFA and latent variables to create a structural model is the preferred analysis strategy, the number of parameters needed to estimate this study’s models would have led to an unacceptably low parameter to sample size ratio. Kline (2005) suggested that researchers should strive for at least a 5:1 sample size to parameter ratio. To meet this benchmark for our study, item parceling was necessary.

Qualitative Methodology

Data collection. Working with GEx administrators, one of the participating groups was invited to serve as a case study for the qualitative portion of the evaluation. This group was selected due to the number of student participants (N = 46) and teacher support for the project, and because the group was traveling to Peru, GEx’s most popular travel destination. Qualitative data collection involved focus groups and dyadic interviews (Table 3) as well as responses from all participant group members to a variety of open ended items on the T2, T3, and T4 questionnaires.

Focus groups and dyadic interviews were conducted with youth participants and their parents during three site visits conducted by the principal investigator (PI). The first two visits occurred during the middle and towards the end of the preparatory portion of the program. These site visit lasted approximately three days and consisted of student focus groups (i.e., 4–6 participants), one large parent focus group with 8–12 parents, and prepara- tory program observations. The third site visit occurred during fall 2008. This visit allowed the PI to interview the same groups of individuals regarding their overall assessment of the program as well as their perceptions of the long-term impact of their experiences.

TABLE 3 Number of Case Study Interviews/Focus Groups

Preparatory International Follow- Program Workshop up Total

Participants 10 23 11 44 Parents 2 5 1 8 Group sponsors 3 1 1 5 GEx staff — 2 — 2

306 M. D. Duerden and P. A. Witt

The PI also traveled with and observed the group during their two-week international workshop in Peru. The first week was spent at several guest lodges in the Peruvian Amazon basin, and the second week took place in southern Peru hiking the Inca Trail to Machu Picchu. The entire group participated in the Amazon portion of the trip, and approximately half of the group stayed for the Inca Trail portion. Interviews and focus groups were conducted with all participants, including teachers and GEx staff members regarding issues pertaining to this study. The PI also conducted participant observations each day of the workshop and took field notes.

Analysis procedures. Interviews and the PI’s field notes were recorded and transcribed. The analysis process was guided by grounded theory methodology as outlined by Strauss and Corbin (1998a) and the study’s research questions. Open coding process enabled the development of themes that were grounded in the data themselves (Strauss & Corbin, 1998b). Axial coding allowed for the development of abstract categories and grouped subcategories (Strauss & Corbin, 1998b). Once categories emerged, the researchers moved to selective coding, whereby a core category was identified and the focus of the analysis shifted to connecting other categories to this core category (Strauss & Corbin, 1998b).

The final step of the analysis process involved the integration of themes and relation- ships between these themes into a coherent response to the research questions. Throughout the analysis process, the codes, analyses, and the emerging theory were reviewed by co-PIs as well as the participants themselves to ensure that all analyses remained true to the raw data and lived experience of the respondents (Strauss & Corbin, 1998a). This study em- ployed four validation strategies (Creswell, 2007): extensive time spent in the field with the subjects, the use of multiple forms of data (e.g., interviews with parents, teachers, GEx staff and youth, field notes, open ended survey questions), member checking, and peer review.

Researcher’s relationship to the data. As noted, the PI spent time with members of the case study group as a passive observer of the program. The focus was on building rapport with participants to develop relationships that would foster the open sharing of information. The PI had previous experience as a director of youth programs but taking on the role of observer represented a new experience that required a conscious effort to avoid taking a participatory position in the program. However, the PI’s presence in the field invariably influenced the youths’ experience. For example, without the interviews and focus groups, many of the youth would not have had a comparable opportunity to discuss and debrief their experiences.

Findings

Quantitative Findings

Descriptive statistics. Descriptive statistics for the relevant variables are provided in Table 4. One-way ANOVA’s and chi-square tests were conducted to investigate the pos- sibility of group age, gender, and ethnicity differences as well as baseline equivalence on program measures. Results indicated that participants had slightly higher composite pro- gram outcome scores at baseline (F (1, 150) = 4.08; p = .05) and had a higher mean age (F (1, 150) = 11.7; p = .001) than the comparisons. Comparison girls reporting higher lev- els of pro-environmental behavior at T2 (F(1, 47) = 5.76, p = .02) was the only significant gender difference.

Hypothesis 1. Results from a repeated measures ANOVA comparing participant and comparison KAB scores across T1, T2, and T4 (T3 was not used in the analysis due to

The Impact of Socialization 307

TABLE 4 Participant and Comparison Descriptive Statistics

T1 T2 T3 T4

Measure Group M SD M SD M SD M SD

KAB Participant 2.79 .66 3.66 .57 3.88 .55 4.00 .53 Comparison 2.57 .52 3.03 .65 — — 3.26 .60

Opportunities Participant — — 4.00 0.62 4.28 0.56 — — Rewards Participant — — 4.00 0.71 3.23 0.47 — — Bonding Participant — — 4.47 0.54 4.65 0.51 — — Skills Participant — — 3.92 0.98 3.55 0.41 — — Beliefs Participant — — 4.60 0.40 4.65 0.37 — —

the lack of comparison data from this collection period) supported the hypothesis that the participant group experienced significant program outcome growth in relation to the com- parison group. Although Levene’s tests indicated nonhomogenous error variance between the participant and comparison groups for KAB at T1, assumptions for sphericity were met, which represents the most critical assumption for repeated measures ANOVA’s (Field, 2005). Consequently, no transformations were made to the data. Results revealed a signifi- cant main effect for time of testing (F(2, 208) = 130.43, p < .001, partial eta squared = .56), a significant interaction effect for time of testing x group (i.e., participant or comparison; F(2, 208) = 11.20, p < .001, partial eta squared = .10), and a significant group effect (F(1, 104) = 27.49, p < .001, partial eta squared = .21). These findings indicated that the participant group reported higher overall KAB scores and experienced a significantly greater growth pattern (i.e., time x group interaction; see Figure 3) than the comparisons.

Hypothesis 2a. A covariance structure analysis was used to test the hypothesized mediating role of the SDM in terms of KAB development during the preparatory portion of the program. Before testing the conceptual model at T2, a correlation matrix of all of the model’s variables (Table 5) was reviewed. Due to a nonsignificant correlation between T1 KAB and T2 opportunities, this path was removed and a path from T1 KAB to T2 beliefs was added (Figure 4). The authors considered this adaptation to be both empirically and theoretically justified. In this revised model, belief mediated relationships between T1 KAB and T2 KAB and was influenced by the SDM variables.

Perceived Rewards (T2)

Skills for Involvement (T2)

Bonding (T2)

Beliefs (T2) KAB (T2)

Perceived Opportunities (T2)

KAB (T1)

FIGURE 3 Participant vs. comparison KAB scores.

308 M. D. Duerden and P. A. Witt

TABLE 5 Intercorrelations between T2 Model Variables (n = 103)

1 2 3 4 5 6 7

1. T1 KAB — 0.10 0.08 0.02 0.10 0.33∗∗ 0.60∗∗

2. T2 Opportunities — 0.52∗∗ 0.43∗∗ 0.16 0.41∗∗ 0.29∗∗

3. T2 Rewards — 0.41∗∗ 0.30∗∗ 0.25∗∗ 0.21∗

4. T2 Bonding — 0.28∗∗ 0.44∗∗ 0.25∗∗

5. T2 Skills — 0.24∗ 0.13 6. T2 Beliefs — 0.57∗∗

7. T2 KAB —

∗ p < .05; ∗∗ p < .01.

The model was run as specified in Figure 4. All exogenous variables (i.e., opportunities, skills, and KAB) were allowed to correlate. Rewards and bonding were also allowed to correlate as this addition significantly improved model fit. Results indicated that all paths were significant and that the data fit the model well (χ 2 = 11.41, 10 df, N = 103; NNFI = 0.99, CFI = 0.99, and RMSEA = 0.03). The model explained 52% of the variance in T2 KAB. The full results from this model are presented in Figure 5.

The indirect effect of T1 KAB to T2 KAB was significant (t = 3.18). Thus, T2 beliefs partially mediated the development of KAB from T1 to T2. Table 6 contains a complete presentation of all direct and indirect model effects.

Because the social development model makes claims regarding the predictive sequence of its variables, two nested models were run to test this assumption. The first model involved all T2 model variables except for T1 KAB. The second model contained the same variables but with all paths running in the opposite direction (e.g., T2 KAB predicting T2 beliefs). The difference in fit between the two models, forward nested (χ 2 = 11.56, 8 df, N = 103; NNFI = 0.96, CFI = 0.98 and RMSEA = 0.06) and backward nested (χ 2 = 11.13, 7 df, N = 103; NNFI = 0.5, CFI = 0.98 and RMSEA = 0.07), was nonsignificant regarding change in chi-square (.43).

2.00

2.50

3.00

3.50

4.00

4.50

T4T2T1

Comparison Participant

FIGURE 4 Preparatory program social development model.

The Impact of Socialization 309

Perceived Rewards (T2)

Skills for Involvement (T2)

Bonding (T2)

Beliefs (T2) KAB (T2)

Perceived Opportunities (T2)

KAB (T1)

.48

.24

.32

.46

.85 .43 .42

FIGURE 5 Preparatory program social development model results.

Hypothesis 2b. The same model was retested at T3 to determine the mediating effects of the SDM during the international workshop portion of the program. A correlation matrix including all of the model’s variables (Table 7) was developed for all applicable variables. All proposed model relationships were supported by the appropriate significant correlations.

All exogenous variables (i.e., KAB, opportunities and skills) were allowed to freely correlate. Although all paths, except for skills → rewards (β = .14, p > .05), were significant, the fit of this model to the data was weak (χ 2 = 38.39, 11 df, N = 102; NNFI = 0.85, CFI = 0.92 and RMSEA = 0.14). A second model was run with an added path based on the modification indices from KAB to rewards. In this revised model, which represented a significant chi-square change (χ 2 = 19.17) from the initial model, all paths were significant and fit was adequate (χ 2 = 19.22, 10 df, N = 102; NNFI = 0.95, CFI = 0.97 and RMSEA = 0.09). The model explained 61% of the variance in T3 KAB. The full results from this model are presented in Figure 6. The indirect effect of T2 KAB to T3 KAB was significant (t = 1.45). Thus a portion of the SDM (i.e., rewards, bonding, and beliefs) partially mediated the development of KAB over the course of the international workshop. See Table 8 for a complete presentation of all direct and indirect model effects.

To test the model’s linear sequence the same procedure was followed as employed with T2 nested models. Neither model fit the data particularly well, forward nested

TABLE 6 Summary of Preparatory Program Effects

Path Indirect Total SE t

T1 KAB → T2 KAB — .59∗∗ .07 7.69 T1 KAB → T2 KAB .13∗∗ — .04 3.18 T2 Opportunities → T2 KAB .07∗∗ — .02 3.01 T2 Skills → T2 KAB .04∗ — .01 2.39 T2 Rewards → T2 KAB .15∗∗ — .04 2.91 T2 Bonding → T2 KAB .18∗∗ — .05 3.82 ∗Significant at the p < .05 level. ∗∗Significant at the p < .01 level.

310 M. D. Duerden and P. A. Witt

TABLE 7 Intercorrelations between T3 Model Variables (n = 103)

1 2 3 4 5 6 7

1. T2 KAB — 0.29∗∗ 0.48∗∗ 0.32∗∗ 0.02 0.52∗∗ 0.76∗∗

2. T3 Opportunities — 0.54∗∗ 0.38∗∗ 0.31∗∗ 0.26∗∗ 0.31∗∗

3. T3 Rewards — 0.57∗∗ 0.29∗∗ 0.39∗∗ 0.59∗∗

4. T3 Bonding — 0.21∗ 0.55∗∗ 0.43∗∗

5. T3 Skills — 0.11 0.17 6. T3 Beliefs — 0.56∗∗

7. T3 KAB —

∗ p < .05; ∗∗ p < .01.

(χ 2 = 40.63, 9 df, N = 102; NNFI = 0.77, CFI = 0.87 and RMSEA = 0.18) and backward nested (χ 2 = 37.92, 9 df, N = 102; NNFI = 0.79, CFI = 0.87 and RMSEA = 0.16), but the backward nested model had a significantly smaller (i.e., better) chi-square statistic.

Qualitative Findings

The focus of the qualitative research questions was to understand the role of socialization processes across the components of the program. The core category of shared experiential experiences emerged as the main driving force behind social bonding. The nature and impact of these shared experiences differed between the preparatory and international workshop portions of the program. Accordingly, the characteristics and role of shared experiences are discussed separately for the preparatory and international workshop portions of the program. Attention is also given to the occurrence and impact of negative social processes. Finally, a theoretical schema is proposed regarding the interrelationship of these two categories of shared experiences and their impact on bonding as an outcome of participation.

Preparatory program shared experiences. When discussing their experience during the preparatory portion of the program, youth frequently highlighted the positive role of team building and group activities. In addition to the preparatory curriculum provided by GEx, the case study group participated in a three-day retreat at a local camp. This experience represented a key shared experience for teachers, parents, and youth. The principal of the case study school, who also participated with the GEx group, noted the role of the retreat in helping the group come together. He also noted he would not forget the interactions and that “ . . . those kids [who attended the retreat] and I have a bond in the hallway that I do not have with any of the other kids.”

Perceived Rewards

(T3)

Skills for Involvement

Bonding (T3)

Beliefs (T3)

KAB (T3)

Perceived Opportunities (T3)

KAB (T2)

.39

.16

.38

.56 .43 .23

.36

.64

FIGURE 6 International workshop social development model results.

The Impact of Socialization 311

TABLE 8 Summary of International Workshop Model Effects

Path Indirect Total SE t

T2 KAB → T3 KAB — .75∗∗ .06 11.63 T2 KAB → T3 KAB .11∗∗ — .04 2.79 T3 Opportunities → T3 KAB .02∗ — .01 2.23 T3 Skills → T3 KAB .01 — .004 1.60 T3 Rewards → T3 KAB .06∗ — .02 2.55 T3 Bonding → T3 KAB .10∗∗ — .04 2.74 ∗Significant at the p < .05 level. ∗∗Significant at the p < .01 level.

Thus, even at an early stage in the program, shared experiences facilitated bonding within the group. One of the teachers indicated that “even the parents liked it [the retreat] because they got to know the kids better and you could see the personalities of the kids and what you might need to do on the trip to help them be more successful.” Regarding the SDM, the retreat afforded participants the opportunities for involvement specifically related to getting to know each other which laid the foundation for the development of stronger bonds during the international workshop.

International workshop shared experiences. During the international workshop, all participants were placed into 24/7 contact with each other over a 10–16 day period. Based upon the qualitative findings, the subcategories linked to shared experiences during the international workshop were leaving the comfort zone, challenge, social support, and social equality. While approximately 50% of the youth participants had traveled internationally before, only two had previously participated in a GEx program. The culture, climate, language, food, and ecology starkly contrasted their home environment. Most participants at one point or another felt they had left their comfort zone, which in turn appeared to positively affect many of the youth. For example, one participant reported that “ . . . being away from home and my family and put in this totally different environment has definitely made me stronger and kind of showed me that I can do this. I have really gained a lot of confidence.”

The physical and emotional challenges faced by many of the participants added to their out of the comfort zone perceptions. In describing their experience along the trail, some participants noted that the experience was both physically and mentally challenging but they felt confidence in themselves after completing it. Others experienced challenge in dealing with heights while traversing the rain forest canopy walkways (e.g., “when I got up there and I actually started walking and it was wobbly and that’s when I got really scared”), a new climate (e.g., “The climate was very different from home and was very uncomfortable”), or homesickness (e.g., “I did not really know how much I was going to miss my mom and my brother until now”).

These challenges as well as others created opportunities for participants to both give and receive social support. They were put into foreign situations where they had to rely on each other. The PI observed that on a particularly difficult section of the trail, each time the group stopped for a break group members would continue to encourage those who were still hiking. One male participant shared the following social support experience he shared with his mom while on the trail:

312 M. D. Duerden and P. A. Witt

I feel like the most important thing I learned today was that helping somebody else’s experience could end up helping yours in the end . . . .I stayed back with my mom today and I know that she appreciated it a lot because [we] all kept supporting her and telling her to keep going.

The support needed and given was not always due to physical challenges. At times youth who were far away from home simply needed someone to talk: “I think just having some- body to talk to is really important and I think everybody stepped up . . . you were able to talk to them if you needed them and I think we just kind of grew closer that way.”

While going through the international workshop, the nature of many of the relationships changed. Participants appeared to see each other from more equitable perspectives. The salience of roles such as teacher, adult, student, or cool kid was superseded by being a member of the “rainforest posse,” as one youth dubbed the group. Relationships became more egalitarian as adults and youth participated in many of the same activities together.

In addition, social cliques dissolved as new relationships formed across old social boundaries. One of the teachers indicated that the program took the students out of their school peer groups and put them with kids they would “not normally run with.” The teacher explained the impact on one particularly popular student “ . . . [who] let down some guards that she usually has [and] realize[d] that hey, wow, I do kind of like you. I never paid any attention to you before because I thought I was too cool for you but I like you.”

Youth and adults also began to see each other differently. In talking about the teachers on the trip one youth commented: “It is like the adults are kids too because they are here to learn and have fun just like we are, and that way they are like us.” After completing a difficult portion of the Inca trail a parent noted that: “I kind of see these kids as not kids anymore . . . I mean they are interacting with adults like adults . . . it was almost like they are starting to cross over from being a kid to being an adult and interacting with us on an adult level.”

In summary, during the international workshop the shared experiences built upon the foundation of the preparatory program and provided a context for group bonding, a process with parallels to the SDM. Participants were afforded opportunities for involvement (e.g., getting to know each other, activity workshops, service projects, physically strenuous activ- ities) within contexts that also promoted the development of requisite skills and rewarded participants for successful participation. Thereby, strong bonds were produced. One mom, who joined the group halfway through the trip, noted that relationships had deepened and changed in ways that were immediately apparent even to an outside observer: “It was like you guys were this bonded group that had this deep relation . . . . I was completely blown away walking into that.”

Negative program experiences. In contrast to experiences marked by social equality and social support, instances of negative interactions occurred. These interactions seemed to be most closely associated with situations where youth felt a lack of autonomy or involvement in the program. For example, some youth expressed a desire to play a more active role in the preparatory program. In addition, during the international workshop logistical and safety concerns sometimes led the adults to take on more controlling roles. On the trip the PI noted that “youth travelers have to make very few decisions and accordingly, appear to get distracted easily . . . I wonder if it would be easier to manage the groups if the kids were given more leadership opportunities.” The adults’ responsibilities also appeared to create stress that at times spilled over into negative interactions with the youth. For example, one youth noted that some adults got really grouch during the trip. A few

The Impact of Socialization 313

Shared Experiences

International Workshop

Preparatory Program Bonding

• Group Activities • Getting to know

each other

• Leaning the comfort zone

• Challenge • Social support • Social equality

FIGURE 7 Shared experience and bonding framework.

parents also noted that while the trip was a positive experience, in some instances adults reprimanded youth unnecessarily.

While the qualitative data and the PI’s observations suggest that the majority of the social interactions within the program were positive, noting and acknowledging negative occurrences is important. One of the main differences between the observed positive and negative social interactions was the degree to which individuals adhered to their traditional teacher, parent, and youth roles. When individuals adopted these roles fewer opportunities for youth involvement were provided, which seemed to impede the bonding process as suggested by the SDM. Accordingly, it appeared that during the international workshop negative shared experiences promoted an unequal social hierarchy while positive interac- tions equalized youth/adult relationships and created situations where they all just became part of the “rain forest posse.”

Proposed shared experience theoretical framework. Although the nature, quality, and impacts of the shared experiences differed somewhat between the preparatory program and the international workshop, they both had roots in experiential activities. Whether it was a team building activity during a preparatory meeting or learning about leaf cutter ants while on a hike through the rain forest, these shared experiences combined to produce a sense of bonding among participants. Given the categories and subcategories that emerged from the study, we propose the following theoretical framework to explain the interrelationship between these constructs (Figure 7).

The framework suggests that bonding is an end result of the accumulation of shared experiences across the program. The foundational importance of the preparatory program shared experiences, which allowed the group to get to know each other and thereby prepare themselves to experience the more intense and rich shared experiences associated with the international workshop, is also highlighted. Without the socializing during the preparatory program, the international workshop bonding likely would not have been as strong. Even before traveling, the youth seemed aware of this connection and the importance of getting to know each other; these experiences set the groundwork for the development of strong social bonds during the international workshop. This finding echoes Brofenbrenner’s (1979) assertion that the developmental potential of a new context is enhanced when individuals are accompanied by others with whom they share a pre-existing attachment.

314 M. D. Duerden and P. A. Witt

Discussion

The results of this study support the efficacy of the program under evaluation. The findings also contribute to the leisure literature by presenting a significant application of a SDM to a youth recreation setting. While social processes have long been identified as a key positive component of youth recreation programs (Bocarro & Witt, 2005), this study connected these processes, both theoretically and empirically, to the program’s outcomes. Also, the successful application of the SDM in a structured leisure context identified at least some of the socialization processes that may account for previous findings related to the influence of peers on structured recreation program participation and outcomes (Barber et al., 2005). These findings strengthen the efficacy and scope of the SDM and validate practitioner efforts to target positive socialization as a core component of their programs.

Portions of the SDM played a partial mediating role in both the preparatory and international travel components of the program regarding the development of program outcomes related to environmental knowledge, attitude, and behaviors. Youth who felt more involved, rewarded, and bonded to the program also reported higher levels of program outcomes across both stages of their experiences. The SDM appeared to function more robustly during the preparatory program. Indirect effects and path coefficients were stronger in the preparatory model, and this model delivered a better overall fit than the international workshop model. This finding is not surprising because during the travel workshop much of the adult (i.e., coordinators, teachers, and parents) attention was occupied with logistics and programming concerns due to the group size. Accordingly, socializing processes may not have been given a chance to operate fully. Conversely, perhaps the processes were operating but the socialization was more a peer-to-peer rather than a youth-to-adult process during this portion of the program, which was not quantitatively assessed in this study.

The quantitative findings provided empirical evidence of the relationship between bonding and program outcomes and the qualitative findings offer insights into the an- tecedents of bonding. At the heart of the processes that led to bonding, the role of shared experiences was identified as a major component. Intentionally designed recreation-based youth programs can serve as prime contexts for youth and adults to share positive ex- periences. In addition, the proposed shared experience and bonding framework suggests the ability of shared experiences across the duration of a program may have a cumulative impact on bonding.

The data also supported the role that adults can play in youth program contexts. In- volving adults and especially parents in youth programs is a noted best practice for youth program providers and educators (Eccles & Harold, 1993; Trotman, 2002) and research find- ings support the efficacy of this practice (Catalano, Berglund, Ryan, Lonczak, & Hawkins, 2002). However, involving adults without providing them adequate direction and training may actually have a negative impact on youth participants’ program experiences.

The data further supported the claim that socialization processes play an important role in the development of program outcomes. Implications for the SDM suggest that data should be collected not only on the degree to which opportunities within a context exist, but also the nature of these experiences in terms of perceived equality. The perceived level of social equality associated with a context’s opportunities for involvement may moderate the impact of this variable within the model. Other variables may have significant predictive links to bonding that the SDM does not take into account such as perceived challenge and opportunities to provide and receive social support. Future research should be conducted to ascertain whether or not adapting the involvement variable and adding additional variables (e.g., perceived challenge and social support) to the SDM would prove efficacious. Recreation program contexts could serve as a prime setting for such research efforts.

The Impact of Socialization 315

The study had several limitations. The lack of randomization of treatment and con- trol assignment adversely affected generalizability. The sample also consisted mainly (i.e., >80%) of Caucasian youth from a middle to upper-middle class suburb of a major midwest- ern urban area. The homogeneity of this group and the socioeconomic status also affects the generalizability of the findings. The program implementation variability across the groups also may have influenced our ability to identify within-person variability. In addition, a larger sample would have allowed the testing of measurement models and more complex and perhaps, more meaningful structural models. Although most variables exhibited strong psychometric properties, some such as the GEx beliefs scale suffered from weak internal consistencies. Improved measurement of the beliefs variable may lead to more accurate assessments of the SDM in youth program contexts.

Conclusion

This study represents a contribution to the youth development, recreation, and SDM lit- eratures. First, the study supported the efficacy of the program under evaluation. Program participants experienced positive growth across selected outcome measures. The growth can be attributed to the program due the study’s quasi-experimental design. Findings also supported the predictive efficacy of employing the SDM to understand the relationship between program processes and targeted outcomes within the context of youth programs. The parsimonious nature of the SDM allowed for a straightforward transition from findings to application. Thus, the model illustrates action-oriented research (Small & Uttal, 2005) by promoting theoretically sound research that is conducive to practitioner applicability. The SDM also deserves further consideration in the youth development and recreation literature for both its predictive power and its practitioner accessibility.

References

Akers, R. L. (1977). Deviant behavior: A social learning approach. Belmont, CA: Wadsworth Publishing.

Babbie, E. (2005). The practice of social research (10th ed.). Belmont, CA: Wadsworth/Thomson Learning.

Bandura, A. (1977). Social learning theory. Englewood Cliffs, NJ: Prentice-Hall, Inc. Barber, B. L., Stone, M., Hunt, J., & Eccles, J. (2005). Benefits of activity participation: The roles

of identity affirmation and peer group norm sharing. In J. L. Mahoney, R. Larson, & J. Eccles (Eds.), Organizes leisure activities as contexts of development. Mahweh, NJ: Lawrence Erlbaum Associates.

Baumrind, D. (1991). Effective parenting during the early adolescent transition. In P. A. Cowan & M. Hetherington (Eds.), Family transitions (pp. 111–163). Hillsdale, NJ: Lawrence Erlbaum.

Beier, S. R., Rosenfeld, W. D., Spitalny, K. C., Zansky, S. M., & Bontempo, A. N. (2000). The potential role of an adult mentor in influencing high-risk behaviors in adolescents. Archives of Pediatrics & Adolescent Medicine, 154(4), 327–331.

Bocarro, J. N., & Witt, P. A. (2005). The power of people: Relationship-based programming. In P. A. Witt & L. L. Caldwell (Eds.), Recreation and adolescent development (pp. 266–280). State College, PA: Venture.

Brofenbrenner, U. (1979). The ecology of human development. Cambridge, MA: Harvard University Press.

Catalano, R. F., Berglund, M. L., Ryan, J. A. M., Lonczak, H. S., & Hawkins, J. D. (2002). Positive youth development in the United States: Research findings on evaluations of positive youth development programs. Prevention & Treatment, 5(1), 1–117.

316 M. D. Duerden and P. A. Witt

Catalano, R. F., Haggerty, K. P., Oesterle, S., Fleming, C. B., & Hawkins, J. D. (2004). The importance of bonding to school for healthy development: Findings from the Social Development Research Group. Journal of School Health, 74(7), 252–261.

Catalano, R. F., & Hawkins, J. D. (1996). The social development model: A theory of antisocial behavior. In J. D. Hawkins (Ed.), Delinquency and crime: Current theories (pp. 149–197). New York: Cambridge University Press.

Catalano, R. F., Oxford, M. L., Harachi, T. W., Abbott, R. D., & Haggerty, K. P. (1999). A test of the social development model to predict problem behavior during the elementary school period. Criminal Behavior and Mental Health, 9, 39–56.

Creswell, J. W. (2007). Qualitative inquiry & research design: Choosing among five approaches. Thousand Oaks, CA: Sage.

Eccles, J. S., & Harold, R. D. (1993). Parent-school involvement during the early adolescent years. Teachers College Record, 94(3), 568–587.

Field, A. (2005). Discovering statistics through SPSS. Thousand Oaks, CA: Sage. Fleming, C. B., Brewer, D. D., Gainey, R. R., Haggerty, K. P., & Catalano, R. F. (1997). Parent

drug use and bonding to parents as predictors of substance use in children of substance abusers. Journal of Child and Adolescent Substance Abuse, 6(4), 75–86.

Global Explorers. (2008). Core disciplines. Retrieved March 24, 2000, from http://www. globalexplorers.org/site/main/about/core disciplines/

Grossman, J. B., & Bulle, M. J. (2006). Review of what youth programs do to increase connectedness of youth with adults. Journal of Adolescent Health, 39, 788–799.

Hanson, W. E., Creswell, J. W., Clark, V. L., Petska, K., & Creswell, J. D. (2005). Mixed methods research designs in counseling psychology. Journal of Counseling Psychology, 52(2), 224–235.

Hartup, W. W. (1996). The company they keep: Friendships and their developmental significance. Child Development, 67, 1–13.

Hawkins, J. D., Catalano, R. F., & Arthur, M. W. (2002). Promoting science-based prevention in communities. Addictive Behaviors, 27(6), 951–976.

Hawkins, J. D., Smith, B. H., Hill, K. G., Kosterman, R., Catalano, R. F., & Abbott, R. (2003). Understanding and preventing crime and violence: Findings from the Seattle Social Devel- opment Project. In T. P. Thornberry & M. D. Krohn (Eds.), Taking stock of delinquency: An overview of findings from contemporary longitudinal studies (pp. 255–312). New York: Kluwer Academic/Plenum Publishers.

Hawkins, J. D., & Weis, J. G. (1985). The social development model: An integrated approach to delinquency prevention. The Journal of Primary Prevention, 6(2), 73–97.

Hill, E., & Sibthorp, J. (2006). Autonomy support at diabetes camp: A self-determination theory approach to therapeutic recreation. Therapeutic Recreation Journal, 40(2), 107–125.

Hirschi, T. (1969). Causes of delinquency. Edison, NJ: Transaction Publishers. Hughes, J. N., Cavell, T. A., & Willson, V. (2001). Further support for the developmental significance

of the quality of the teacher-student relationship. Journal of School Psychology, 39(4), 289–301. Hughes, J. N., & Kwok, O. M. (2007). Influence of student-teacher and parent-teacher relationships

on lower achieving readers’ engagement and achievement in the primary grades. Journal of Educational Psychology, 99(1), 39–51.

Kishton, J. M., & Widaman, K. F. (1994). Unidimensional versus domain representative parceling of questionnaire items: An empirical example. Educational and Psychological Measurement, 54(3), 757–765.

Kleiber, D., Larson, R. W., & Csikszentmihalyi, M. (1986). The experience of leisure in adolescence. Journal of Leisure Research, 18(3), 169–176.

Kline, R. (2005). Principles and practice of structural equation modeling (2nd ed.). New York: Guilford Press.

Leeming, F., & Dwyer, W. (1995). Children’s environmental attitude and knowledge scale: Construc- tion and validation. Journal of Environmental Education, 26, 22–31.

Lonczak, H. S., Huang, B., Catalano, R. F., Hawkins, J. D., Hill, K. G., Abbott, R. D., et al. (2001). The social predictors of adolescent alcohol misuse: A test of the social development model. Journal of Studies on Alcohol, 62(2), 179–189.

The Impact of Socialization 317

Matsueda, R. L. (1982). Testing control theory and differential association: A causal modeling approach. American Sociological Review, 47(4), 489–504.

Persson, A., Kerr, M., & Stattin, H. (2007). Staying in or moving away from structured activities: Explanations involving parents and peers. Developmental Psychology, 43(1), 197–207.

Pratt, C. C., McGuigan, W. M., & Katzev, A. R. (2000). Measuring program outcomes: Using retrospective pretest methodology. American Journal of Evaluation, 21(3), 341–350.

Shaw, S. M., Kleiber, D. A., & Caldwell, L. L. (1995). Leisure and identity formation in male and female adolescents: A preliminary examination. Journal of Leisure Research, 27(3), 245–263.

Sibthorp, J., Paisley, K., Gookin, J., & Ward, P. (2007). Addressing response-shift bias: Retrospective pretests in recreation research and evaluation. Journal of Leisure Research, 39(2), 295–315.

Small, S. A., & Uttal, L. (2005). Action-oriented research: Strategies for engaged scholarship. Journal of Marriage and Family, 67(4), 936–948.

Strauss, A., & Corbin, J. (1998a). Basics of qualitative research (2nd ed.). Thousand Oaks, CA: Sage Publications.

Strauss, A., & Corbin, J. (1998b). Grounded theory methodology: An overview. In N. K. Denzin & Y. S. Lincoln (Eds.), Strategies of qualitative inquiry (Vol. II, pp. 158–183). Thousand Oaks, CA: Sage.

Trotman, M. F. (2002). Involving the African American parent: Recommendations to increase the level of parent involvement within African American families. Journal of Negro Education, 70(4), 275–285.

Copyright of Leisure Sciences is the property of Routledge and its content may not be copied or emailed to

multiple sites or posted to a listserv without the copyright holder's express written permission. However, users

may print, download, or email articles for individual use.