543005 research-article2014

HEBXXX10.1177/1090198114543005Health Education & BehaviorMann et al.

Regular Article

Improving Academic Self-Efficacy, School Connectedness, and Identity in Struggling Middle School Girls: A Preliminary Study of the REAL Girls Program

Health Education & Behavior 2015, Vol. 42(1) 117­–126 © 2014 Society for Public Health Education Reprints and permissions: sagepub.com/journalsPermissions.nav DOI: 10.1177/1090198114543005 heb.sagepub.com

Michael J. Mann1, Megan L. Smith1, and Alfgeir L. Kristjansson1

Abstract Girls struggling to be successful in middle school are often dealing with negative life experiences that affect their ability to achieve academically. Frequently, their academic failures and problem behaviors are associated with feeling overwhelmed by difficult and challenging life circumstances. In the absence of intervention, these patterns may contribute to girls chronically underperforming in school, dropping out of school, and becoming involved in delinquent and high-risk behaviors. This article describes a quasi-experimental, mixed-methods study of the REAL Girls program. REAL Girls was designed to help struggling middle school girls develop resilience—particularly academic self-efficacy, school connectedness, and identity—and achieve successful outcomes in school and life. In this study, using a crossover design, 48 girls identified as experiencing academic failure, school behavior problems, or truancy participated in one of two implementations of this 3-day intervention. Findings based on both quantitative and qualitative data suggest that REAL Girls contributed to positive increases in academic selfefficacy, school connectedness, and identity. Repeated measures analysis of variance and paired t tests suggest significant increases in each outcome variable, both immediately after program delivery and 2 weeks later, and effect size estimates suggest moderate to large program impact. Focus groups conducted 90 days after implementation of the program confirmed the quantitative findings and support the efficacy of the REAL Girls program and approach. Keywords at-risk girls, gender-responsive middle schools, resilience, youth development Most middle school girls are happy in their lives and do well in school. They typically achieve academically, make positive behavior decisions, and demonstrate leadership within their school communities (Acoca, 1999; Hawkins, Graham, Williams, & Zahn, 2009; Johnson, Roberts, & Worell, 1999). Some middle school girls, however, struggle more than others. Often, the girls who struggle the most are dealing with challenging or traumatic life experiences that affect their ability to be successful (Hawkins et al., 2009; Mullis, Cornille, Mullis, & Huber, 2004). For these girls, problem behaviors often reflect their attempts to cope with emotional pain and academic failures frequently indicate how much energy is being usurped by difficult circumstances. In the absence of intervention, these girls are more likely than boys and other girls to develop patterns of chronically underperforming in school, dropping out, and becoming involved in delinquency (Chesney-Lind, 2001). Additionally, they are more vulnerable than both groups to experiencing disproportionate, lifelong rates of victimization and abuse; high-risk sexual behavior associated with unintended pregnancy and sexually transmitted infections; alcohol and

substance abuse; and emotional health problems including depression and anxiety disorders (Aalsma & Lapsley, 2001; Crosby et al., 2004; Kofler et al., 2011; S. Miller, Malone, & Dodge, 2010; Thompson, Kingree, & Desai, 2004). The REAL Girls program was developed based on Project Challenge (Mann, 2013), which is a gender-responsive wilderness adventure program designed to promote resilience and effective coping skills in delinquent middle school girls. Although previous experimental study supported the effectiveness of the Project Challenge program, it requires specialized outdoor equipment and wilderness training not available in most schools. The REAL Girls program was designed to incorporate the 12 core strategies of the Project Challenge program into a school-based 1

West Virginia University, Morgantown, WV, USA

Corresponding Author: Michael J. Mann, Assistant Professor, Department of Social and Behavioral Sciences, West Virginia University, School of Public Health, P.O. Box 9190, Morgantown, WV 26505-9190, USA. Email: [email protected]

Downloaded from heb.sagepub.com at UNIV OF GEORGIA LIBRARIES on May 26, 2015

118

Health Education & Behavior 42(1)

Developmental Challenge

Intrapersonal Development

Abuse

Self-Confidence

Family Fragmentation

Self-Esteem

Stress Appraisal Primary Secondary

School Failure Untreated Health Problems Convergence of Risk Factors in Early Adolescence

Outcomes Associated With… Health Well-being Quality of Life

Identity

Intrapersonal Development Social Support Mattering

Coping Effort Problem-focused Emotion-focused Avoidant

Outcomes Associated With… Risk Maladaptation Loss

Figure 1.  The model of girls’ resilience.

program that can be delivered in any school by traditionally trained school personnel.

Theoretical Background Theoretically, both programs are based on the model of girls’ resilience. The model of girls’ resilience (Figure 1 and Table 1) is a conceptual model designed by the first author of this study during the creation of the Project Challenge program (Mann, 2013). This model was developed when broader theoretical models did not sufficiently illuminate the specific circumstances of multiple-risk adolescent girls struggling with problems in school, the legal system, and high-consequence health decision making. At its core, this model is an adaptation of Lazarus’s transactional model of stress and coping (Lazarus & Folkman, 1984) as informed by Gilligan (1993), Josselson (1987), J. B. Miller (1986), and Belenky, Clinchy, Goldberger, and Tarule’s (1986) individual works describing women’s psychosocial development and resiliency theory (Bernard, 1991; Hawkins et al., 2009; Tomsen, 2002). In their seminal report, Guiding Principles for Promising Female Programming, the Office of Juvenile Justice and Delinquency Prevention (OJJDP; 1998) made a convincing case that higher risk adolescent boys and girls struggle for fundamentally different reasons. The model of girls’ resilience starts by identifying the developmental challenges isolated by the OJJDP as being most likely to contribute to heightened levels of risk among adolescent girls. Best summarized by Acoca (1999), girls who experience abuse, family fragmentation, a history of school failure, untreated health problems, and a convergence of risk in early adolescence are more vulnerable to a range of disappointing life outcomes than unexposed girls or similarly exposed boys. The report

concludes by contending that the interventions that recognize the unique needs of adolescent girls, and that are most responsive to those differences, produce higher rates of success. More recent empirical work by the OJJDP’s Girls Study Group continues to support these recommendations (Hawkins et al., 2009; Zahn, Hawkins, Chiancone & Whitworth, 2008). Next, the model highlights key developmental characteristics that may be adversely affected by, or whose development may be delayed by, exposure to these difficult and possibly traumatic negative life events. Perhaps not surprisingly, these key characteristics are also associated with resilience in adolescents. Meaning that young people who are able to maintain healthy development in these key areas, in spite of exposure to hardship and challenge, are more likely to experience positive life outcomes than would normally be expected given their difficult circumstances. Specifically, resiliency theory suggests that young people who “beat the odds” are able to (a) retain reasonable levels of self-esteem and self-confidence (key intrapersonal characteristics), (b) maintain relationships in which others are characterized as believing in their value and as being socially supportive (key interpersonal characteristics), and (c) develop an authentic sense of self that is both hopeful and pragmatic (Bernard, 1991; Hawkins et al., 2009; Tomsen, 2002). Conversely, resiliency theory also suggests that young people with lower levels of these critical developmental characteristics are at greater risk of disappointing life outcomes. Additionally, works by Gilligan (1993), J. B. Miller (1968), Belenky et al. (1986), and Josselson (1987) on women’s psychosocial development inform this portion of the model by helping identify the key developmental characteristics that are most salient among young women and by suggesting a tight coupling between intra- and interpersonal development. Additionally, these works contribute to the

Downloaded from heb.sagepub.com at UNIV OF GEORGIA LIBRARIES on May 26, 2015

119

Mann et al. Table 1.  Core Assumptions of the Model of Girls’ Resilience. Assumptions

1. There is great value in women’s unique ways of knowing and interacting with the events that influence their development. Interventions that acknowledge and capitalize on these strengths are more likely to produce successful outcomes for young women. 2. Negative life events exert a discernible influence on girls’ psychosocial development. These events may contribute to developmental delays, maladaptive development, or, under the right conditions, developmental growth. Girls who are struggling with problem behavior, academic failure, and high-risk health behaviors are often coping with negative life events that divert their attention and energy from more adaptive behaviors. The most vulnerable girls often experience developmental delays or establish maladaptive patterns of coping that, in the absence of intervention, influence their short- and long-term outcomes. 3. Girls are fundamentally capable and valuable. They can successfully overcome difficulty, actively engage challenge, and achieve life success. Experiencing developmental challenge or trauma does not reduce their worth or diminish their capacity to live a quality life. Girls who have been adversely affected by experiencing negative life events are capable of recovering delayed developmental growth and establishing behaviors conducive to healthy decision making and life success. 4. Intra- and interpersonal development are intimately related to one another and equally essential as girls develop the ability to achieve successful outcomes. Both intra- and interpersonal development are most likely to be fostered in rich interpersonal environments. Vulnerable young women often discover their value and acquire the courage to take growth-oriented risks in the context of respectful, reflective, and encouraging interpersonal relationships. 5. The intrapersonal characteristic of self-esteem and interpersonal characteristic of mattering are tightly coupled constructs. Selfesteem consists of a girl’s estimation of her value, whereas mattering consists of her estimation of the value others place on her. Together, these constructs influence whether or not she will attempt to cope effectively, that is, “Am I worth the effort?” 6. The intrapersonal characteristic of self-confidence and the interpersonal characteristic of social support are also tightly coupled constructs. Self-confidence includes a girl’s belief in her ability to act on her own behalf, whereas social support includes her ability to develop relationships in which others are willing and able to assist her when necessary. Together, these constructs influence a girl’s level of persistence and the quality of her coping effort, that is, “How certain am I that my efforts will eventually be successful?” 7. Girls compose their identities as both individuals and as they relate to significant others. Factors associated with girls’ resilience become more stable as they are more deeply ingrained into girls’ intra- and interpersonal identities. For instance, a young woman who identifies herself as being “the kind of person who does (behavior) when I’m with (person or type of relationship)” is more likely to choose behaviors congruent with that identity than a young woman who does not identify herself in that manner. 8. Intrapersonal, interpersonal, and identity-related factors contribute to a young woman’s stress appraisals and the selection of her coping efforts. Girls who feel personally confident, have histories that include successfully overcoming challenging circumstances, and believe in the effectiveness of their communities of support are most likely to select the problem-focused coping strategies associated with positive health and life outcomes.

model by emphasizing the role of challenge (Gilligan, 1993; J. B. Miller, 1968) and relationships (Belenky et al., 1986; Gilligan, 1993; Josselson, 1987; J. B. Miller, 1986) as contexts in which young women explore their capacities and develop an authentic sense of identity rooted in both an intraand interpersonal sense of self (J. B. Miller, 1986). Perhaps most important, their works firmly establish two core assumptions of the model: (a) the value of women’s unique ways of knowing and interacting with the events that influence their development (Belenky et al., 1986; Josselson, 1987) and (b) the potential of young women, especially communities of young women, to grow amid challenge and adversity (Gilligan, 1993). Finally, the model suggests that these key developmental characteristics influence young women’s stress appraisal and coping efforts. Generally speaking, young women with higher levels of these key developmental characteristics will perceive new challenges as less stressful (primary appraisal) and themselves as more capable of coping (secondary appraisal) than young women with lower levels of these characteristics. This is critically important as Lazarus and Folkman’s (1984) work suggests that these appraisals directly affect the quality of a person’s coping efforts, which over time influences their

corresponding life outcomes. Specifically, young women whose appraisals are characterized by confidence in themselves and their communities of support are more likely to choose problem-focused, instead of emotion-focused or avoidant coping strategies. This choice matters, because although problem-focused coping strategies require more energy, effort, and ego strength, they are also more likely to produce outcomes associated with health, wellbeing, and quality of life (Wilson, Pritchard, & Revalee, 2005), especially when they are compared to emotion-focused coping strategies that produce less consistent results or to avoidant coping strategies that are associated with a variety of poor health and life outcomes (Wilson et al., 2005). Fundamentally, the model of girls’ resilience is an adaptation and integration of well-tested theories that have individually withstood the scrutiny of years of empirical study. So far, it has been used to guide two intervention programs, and three empirical studies evaluating those programs suggest the utility of the model. Although preliminary in nature, these findings indicate that the model may help practitioners develop interventions designed to promote the health, school, and life success of struggling young women. Additional empirical testing of this model is currently in progress.

Downloaded from heb.sagepub.com at UNIV OF GEORGIA LIBRARIES on May 26, 2015

120

Health Education & Behavior 42(1)

Purpose of the Study This quasi-experimental, mixed-methods study examines the influence of the REAL Girls 3-day intervention program on the intrapersonal characteristic of academic self-efficacy, the interpersonal characteristic of school connectedness, and identity. The findings presented in the current study represent the culmination of the second stage of formative program evaluation. The first stage included a mixed-methods, pre–posttest study of the REAL Girls program previously described by Mann (2012). Together, the findings from both of these preliminary studies will be used to further improve the REAL Girls program, shape future experimental study, and guide other efforts to promote the well-being of struggling middle school girls.

Method This quasi-experimental study used a mixed-methods QUAN + qual approach (Morse, 2003). The quantitative portion used a quasi-experimental crossover design to compare the effects of the program on participant levels of academic selfefficacy, school connectedness, and identity. This design compares the effects of a treatment on two matched groups and allows each to be assigned to both experimental and control conditions at different times during the study (Streiner & Norman, 1998). There are two main advantages to using a crossover design that are particularly relevant to this study. First, it is ethically questionable to exclude a group from treatment when we have prior evidence supporting the effectiveness of that treatment. Since two previous studies strongly suggest the efficacy of the core strategies used in the REAL Girls program, the crossover design allowed both groups to receive the treatment and helped us avoid questionable ethical practice. Second, crossover designs include repeated measures, which not only provide evidence about the short-term sustainability of the intended outcomes but also increase the statistical power of the study (Jones & Kenward, 2003; Stufken, 1996; Vonesh & Chinchilli, 1997). In a preliminary study, especially one that uses a relatively low number of program participants, it is particularly important to maximize analytical power when possible. An outcome survey was administered at three separate time points conducted at approximately 2-week intervals. At Time 1 (baseline), the survey was administered to all participants prior to either group receiving the treatment. At Time 2, all participants completed the survey a second time, after only Group 1 had received the treatment and while Group 2 served as a comparison group. At Time 3, all participants completed the survey, after Group 2 had received the treatment and while Group 1 served as a crossover comparison group. Additionally, participants completed a program satisfaction survey immediately following program participation. Qualitative methods were used in a supportive role. Two focus groups were used to elaborate on participant perceptions

Table 2.  Demographic Composition of Participants in the REAL Girls Program (N = 48). Characteristic

n (%)

Age, years  12  13  14 Race/ethnicity  White  Hispanic  Black

  7 (14.6) 37 (77.1) 4 (9.3)   26 (54.6) 14 (29.4) 8 (16.8)

of the intervention and the outcomes associated with the program participation (Glesne, 2006; Hatch, 2002). Both focus groups lasted 1 hour, consisted of 10 participants, and were conducted approximately 90 days after participation in the REAL Girls program by university students trained in focus group facilitation. This study was reviewed and approved by both the university institutional review board and the participating school district.

Participants This study included 48 girls aged 12 to 14 years (Table 2). Seven participants (14.6%) were 12 years of age, 37 (77.1%) were 13 years, and 4 (9.3%) were 14 years. Twenty-six participants (54.6%) were White, 14 (29.4%) were Hispanic, and 8 (16.8%) were Black. All participants were referred by two public middle schools in California. Each school selected 32 girls with the highest school-related rates of problem behavior, poor academic performance, and truancy. Teachers nominated participants, and the schools’ assistant principals and counselors confirmed that referred students met the criteria for program participation. One hundred percent of participants had a confirmed history of developmental challenge including one of the following: family fragmentation or extended parent absence, abuse or neglect, significant medical or mental health diagnoses, or a history of academic failure originating in elementary school. Informed consent was obtained in writing from each participant’s parent or guardian. Participants were assigned to groups based on their school of origin, with all participants from the same school attending the program together. No statistically significant differences were found between the two groups based on participant ethnicity, socioeconomic status, or parent education levels. All participants completed all informed consent requirements.

Instruments/Measures Three types of data collection were used in this study: an outcome survey, a program satisfaction survey, and a semistructured focus group question guide.

Downloaded from heb.sagepub.com at UNIV OF GEORGIA LIBRARIES on May 26, 2015

121

Mann et al. Outcome Survey. The following concepts and scales were used for main outcome measurement: Self-efficacy: The Academic Self-Efficacy subscale of the Patterns of Adaptive Learning Scales (Midgley et al., 2000) uses five items to measure student confidence in their ability to meet academic challenges and to be successful in school. Sample items include “I am certain I can figure out how to do the most difficult class work” and “I can do even the hardest work in this class if I try” (Cronbach’s α = .78). School connectedness: The School Connectedness Scale (McNeely, Nonnemaker, & Blum, 2002) uses five items to measure the extent to which students feel they are a cared for part of the school community. Sample items include “I feel close to people at this school” and “I am a part of this school” (Cronbach’s α = .82). Identity: Identity was measured using eight items from the Identity subscale of the Adolescent Personality Style Inventory and four items from the Hope Scale. The Identity subscale (Lounsbury, Huffstetler, Leong, & Gibson, 2005) measures an individual’s sense of his or her level of identity formation. Sample items include “I have a firm sense of who I am,” and “I have a clear set of personal values or moral standards” (Cronbach’s α = .84). The Pathways Subscale of the Hope Scale (Snyder et al., 1991) uses four items to measure the perception that goals can be met, and the individual’s self-identification as the type of person who can positively influence his or her future. Sample items include “My past has prepared me for future success,” “I energetically pursue my goals” (Cronbach’s α = .80). Program Satisfaction Survey. This survey provided participants the opportunity to express their opinions about the overall quality of the program, their perceptions about its value, and their willingness to recommend the program to other middle school girls. Focus Group Question Guide. A brief semistructured focus group question guide was developed for this study. It was designed to encourage participants to speak candidly about their experience with the REAL Girls program, especially regarding the program elements that most contributed to the success or failure of the program.

Treatment The REAL Girls 3-day intervention was conducted over two full school days with a 2-hour booster treatment 10 days later. During that time, the program was implemented using the 12 Program Strategies for Promoting Resilience in Middle School Girls (Mann, 2012) as a framework (Table 3). These twelve strategies were developed using the model of girls’ resilience as a guide and in an effort to translate current

research related to promoting the success of vulnerable girls into effective prevention practice.

Analysis Quantitative Data Analysis. Repeated measures analysis of variance was used to examine differences between groups of participants based on the timing of each group’s exposure to the REAL Girls program. Results from these analyses indicated whether or not there was a significant change in the standardized means associated with the studies’ three dependent variables over time. Effect size estimates were calculated using standardized partial eta squared, which described the strength of the treatment’s effect on the targeted outcome variables. Paired samples t tests were used to examine within groups pre- and posttest differences between Times 1 and 3. Descriptive statistics were used to describe participant satisfaction with the REAL Girls program. Qualitative Data Analysis. Qualitative data were analyzed using Hatch’s (2002) interpretative analysis model. Hatch described interpretative analysis as a process in which the researcher gives meaning to qualitative data by actively “making inferences, developing insights, . . . drawing conclusions, and extrapolating lessons” (p. 180). Hatch (2002) encourages using interpretative methods to illuminate an otherwise established phenomenon. Steps used to conduct the qualitative analysis included the following: (a) reading transcripts from both focus groups while attempting to gain a big picture sense of participant comments and recording broad impressions in researcher’s notes; (b) hand coding transcripts from both focus groups and developing corresponding theme statements; (c) studying theme statements and notes for salient interpretations; (d) reviewing researcher notes, transcripts, and theme statements while looking for places where interpretations were supported or challenged; (e) writing a draft summary of the participant themes including quotes; (f) reviewing the draft summary with a group of six focus group/program participants; and (g) writing a revised summary as informed by participant feedback.

Results Quantitative Results Table 4 displays the change in trends between groups from Time 1 through Time 3 on the three subcategories: academic self-efficacy, connectedness, and identity. Effect size estimates ranged from .27 for identity to .46 for connectedness with academic self-efficacy falling in between the two. A moderately larger increase in score was also observed for Group 2 across all categories. Table 5 shows the trends in standardized mean scores for all three subcategories across both groups. In line with the

Downloaded from heb.sagepub.com at UNIV OF GEORGIA LIBRARIES on May 26, 2015

122

Health Education & Behavior 42(1)

Table 3.  Twelve Strategies for Promoting Resilience in Struggling Middle School Girls (Mann, 2012). Strategy Choosing a private or special location Intentionally choosing intensity and energy

Developing supportive and cohesive teams

Providing attentive adult women as role models

Using fun to create a climate of trust and rapport

Establishing and reestablishing relevance

Practicing actively evaluating situations and making decisions

Using preassessment to tailor a portion of the program

Provide time to reflect individually and collaboratively

Using culminating activities and ceremonies to debrief and consolidate learning Providing take-home gifts that communicate care and reinforce community Reconnecting through follow-up events and booster treatments

Description and Brief Rationale The program was conducted off the school grounds in a private location. Choosing an off-site location allowed girls to participate in an environment free from distractions and the perceived expectations of nonparticipating peers. Early adolescent brains are predisposed to direct attention toward novelty, emotional intensity, and social connection. REAL Girls intentionally capitalized on this tendency by using intensity and energy to heighten girls’ attention. The program was conducted during a relatively brief but concentrated time period—two 7-hour days and one 2-hour lunch. Decorations, music, teams, cheers, dancing, and games were all used to create a high-energy atmosphere that felt very different from an ordinary school day and demanded a high level of attention and engagement. After a brief introduction to the program, all participants were randomly assigned to four teams of approximately 9 to 10 girls. Each team developed a team name, banner, and cheers. Teams rotated through each of the learning activities together, and in these teams, participants were able to establish supportive, positive, and empathetic peer and adult connections that were used as the context for processing and applying the program curriculum. A “Big Sister” was assigned to each participant. In this study, Big Sisters were university health promotion students who administered the program. One of the guiding principles of REAL Girls was that “attention is treatment.” As a result, Big Sisters were encouraged to intentionally and purposefully invest time and attention in their Little Sister throughout the program. They did so by trying to sit with, participate in activities with, and be available to listen and talk with their Little Sisters as frequently as possible. The primary goals for Big Sisters included making sure that every girl received enough attention to feel noticed and valued, providing support and encouragement, and identifying and referring girls who might require extra assistance. Many of the risk factors that heighten struggling girls’ vulnerability to diminished school and life outcomes are personal in nature and not easily shared with strangers. As a result, trying to discuss these issues can be counterproductive if initiated before a reasonable amount of positive rapport and trust has been established. To build rapport, the first half of the first day of the program was devoted to team competition focused on playing novel and fun games. Having fun together gave program leaders an opportunity to connect with participants in a nonthreatening environment and helped build a foundation of trust prior to engaging sensitive topics. After a safe environment and positive rapport were established, a large group session was conducted to help girls make a successful transition to more serious topics. During this session, Big Sisters led a discussion about the challenges girls face and the skills required to successfully overcome those challenges. Session leaders emphasized the purpose of the REAL Girls program. Willingness to address relevant issues directly and frankly, while remaining compassionate and encouraging, was a hallmark of the overall program philosophy. Throughout the program, Big Sisters shared their own stories of struggle and trauma and emphasized how they overcame the difficulties they faced as middle school students. The REAL Girls approach focuses on encouraging girls to make and discuss decisions while providing constructive feedback from peers and adults that positively reinforce healthy decision making. Many techniques were used to elicit these types of decision-making opportunities. Some examples include debriefing skits, responding to scenarios, and participating in games where participants categorized items based on risk while identifying choices that promoted school and life success. Although most of the program was delivered using a curriculum focused on specific program goals, approximately 20% of the program was devoted to frank conversation addressing specific concerns identified by the participants themselves. Student input was collected in three primary ways. First, prior to the program, participants completed an interest survey that was used to tailor a portion of the program to address issues of interest not covered in the standard curriculum. Second, each team was given an anonymous question box. Throughout each day, participants were encouraged to put questions in the box that they may have been unwilling to ask in public. Answers to these questions were provided to the entire group before the program ended. Finally, REAL Girls team members intentionally developed a program culture that encouraged participants to ask honest questions and participate in authentic discussions throughout the program. Girls were given time to reflect individually and collaboratively throughout the program. Individual reflections were based on structured questions and recorded in a personal journal. Additionally, girls spent time each day reflecting on their program experiences with their Big Sister. During this Big Sister–Little Sister reflection time, participants discussed what they had learned, what they hoped to remember from their REAL Girls experience, and strategies for overcoming obstacles. At the end of the program, each participant was encouraged to reflect on the lessons she wanted to remember, and to set goals related to school and life success. They reviewed and discussed these lessons and goals with their Big Sisters, created artistic reminders, and presented at least one of these lessons or goals to the entire REAL Girls group. To make this strategy meaningful, a generous amount of time was scheduled to help participants synthesize their “learnings” and prepare to present to the group. The tone of the closing ceremonies was serious, supportive, and encouraging. Each REAL Girls participant left the program with (a) a REAL Girls t-shirt, (b) a framed picture of the whole REAL Girls group, (c) a framed picture of her with her Big Sister, and (d) a handwritten note from her Big Sister describing what they most admired about her Little Sister. These gifts were intentionally chosen to help girls feel connected to a community that accepted them, supported their goals, and cared about their outcomes. A REAL Girls reunion lunch was held approximately 10 days after the program. This provided participants the opportunity to reconnect with their Big Sisters and the broader REAL Girls community. No formal program was presented. Informal activities focused on providing participants with encouragement and support regarding the lessons they learned in REAL Girls and reviewing the goals they set during the program.

Downloaded from heb.sagepub.com at UNIV OF GEORGIA LIBRARIES on May 26, 2015

123

Mann et al. Table 4. Between-Groups F Tests for Changes Across the Three Time Points in the REAL Girls Program. Variable Academic efficacy Connectedness Identity

F

df

η2p

24.38** 41.60** 16.89**

2 2 2

.35 .46 .27

Note. df = degrees of freedom. **p < .01.

Table 5.  Standardized Within-Group Mean Scores at Each Time Point in the REAL Girls Program. Variable

Time 1

Academic self-efficacy   Group 1 −0.78   Group 2 −0.55 School connectedness   Group 1 −1.05   Group 2 −0.72 Identity   Group 1 −0.92   Group 2 −0.28

Time 2

Time 3

0.73 −0.60

0.38 0.81

1.13 −0.60

0.11 1.12

0.59 −0.44

0.03 1.02

Table 6.  Paired Samples t Tests for Groups 1 and 2 Between Time 1 and Time 3 in the REAL Girls Program. Variable Academic efficacy   Group 1   Group 2 Connectedness   Group 1   Group 2 Identity   Group 1   Group 2

t

df

−5.30** −6.45**

24 22

−5.01** −6.88**

24 24

−3.38** −5.37**

23 22

Note. df = degrees of freedom. **p < .01.

crossover design, the findings reveal a consistent pattern of change from Time 1 to Time 2 for Group 1 and from Time 2 to Time 3 for Group 2 with a moderate reduction in score between Time 2 and Time 3 for Group 1 but stable scores for Group 2 between Times 1 and 2. In terms of overall program effects, Table 6 displays the paired samples t tests for within-group change between Times 1 and 3 for all three subcategories. A significant change was observed for both groups on all three subcategories, with increases being marginally larger for Group 2. To further corroborate our findings we also employed paired sample t tests for change across Times 2 and 3 For group 1.

The tests for academic efficacy, t(24) = 2.41, p < .05; and identity, t(23) = 2.73, p = .05, were significant at the 95% level, and the test for school connectedness was significant at the 99% level, t(24) = 5.83, p< .01. The aggregate results from the program satisfaction survey are provided in Table 7. These results indicate that all or most participants found the program helpful, engaging, and an experience they would recommend for other girls. No girls indicated a negative experience with the program.

Qualitative Results Focus group results confirmed the quantitative findings. Girls reported that participating in REAL Girls helped them feel more connected to school, confident academically, goal oriented, and self-assured when facing challenges in general. Furthermore, participants stated that REAL Girls helped them reflect on their experiences, refine their sense of identity, develop new goals, and deepen their commitment to previously established goals. Four themes stood out as particularly influential. Theme 1: Attentive Role Models.  The role model component was described as particularly powerful. Most participants describe their “Big Sister” as “the most important part of the program.” Participants described the value of having an advocate in the program who paid attention to them, listened to them, and was invested in their success. Theme 2 : The REAL Girls Community as a Point of Entry.  REAL Girls provided a community within the school that provided a point of entry from which girls were able to enter the broader school community. One participant stated, “I never got along with other girls until REAL Girls.” Another suggested, “REAL Girls made me want to come to school because people I thought were my enemies became my friends.” Perhaps most tellingly, several participants suggested that in REAL Girls they formed positive relationships with “girls and teachers that understand me.” Theme 3: The Self-Esteem to Self-Efficacy Pathway. Participants described the REAL Girls program as helping them develop the self-esteem necessary to build self-confidence. Many girls described needing to believe they were “worth trying for” before they participated in efficacy-related behaviors. Both the REAL Girls role models and community effectively communicated the “you are valuable” message to participants that allowed them to take the risks required to build self-confidence. Theme 4: Frank Discussion About Relevant Topics. Participants described the importance of REAL Girls staff “talking about what I wanted to know about” and “things that were really happening [in my life].” Additionally, participants described appreciating the opportunity to get real answers about issues they

Downloaded from heb.sagepub.com at UNIV OF GEORGIA LIBRARIES on May 26, 2015

124

Health Education & Behavior 42(1)

Table 7.  Aggregate Program Satisfaction Frequencies in Percentages. Survey Item The REAL Girls program helped me . . .   1. Feel confident and strong   2. Get along better with other girls   3. Feel closer to people who care about me   4. Think about being successful in school   5. Set positive goals for my life   6. Learn how to deal with problems   7. Learn how to deal with stress   8. Help me feel closer to the adults at my school   9. Answer some important questions about life Overall, I believe REAL Girls was . . . 10. Helpful 11. Good for me 12. Good for girls in general 13. Fun

Strongly Agree

Agree

Neutral

Disagree

Strongly Disagree

71.3 88.5 82.8 97.7 93.1 88.5 73.6 97.7 86.2

28.7 6.9 17.2 2.3 4.6 11.5 25.4 0.0 6.9

0.0 4.6 0.0 0.0 2.3 0.0 0.0 2.3 2.3

0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

82.8 82.8 97.7 100.0

17.2 17.2 2.3 0.0

0.0 0.0 0.0 0.0

0.0 0.0 0.0 0.0

0.0 0.0 0.0 0.0

cared about and the REAL Girls staff accepting them “how I really am, instead of trying to make me how they wish I was.”

Discussion and Conclusions This quasi-experimental mixed-methods follow-up study provides additional evidence (Mann, 2012) suggesting that the REAL Girls program may be successfully promoting academic self-efficacy, school connectedness, and identity in struggling middle school girls. In keeping with the nature of the crossover design, each group demonstrated increases in each outcome variable while assigned to the treatment condition and either remained stable or decreased moderately while assigned to the comparison condition. Short-term postprogram comparisons suggest that although there is a reduction in some of the standardized means 2 weeks after program completion, some positive differences may be sustained, at least in the short term. Although the results associated with Group 1 suggest some reductions in academic self-efficacy, school connectedness, and identity 2 weeks after program completion, qualitative results suggest that the benefits of the program persist at least 90 days after program completion. Program satisfaction survey findings indicate that girls appreciate the REAL Girls experience, believe it helped them, and think other girls will benefit from participating in it. Qualitative results confirm these findings and highlight the importance of strategies related to providing attentive role models, building communities of support, promoting self-esteem as a precursor to developing self-efficacy, and engaging possibly sensitive topics that are relevant to participating girls. Additionally, this is the third study that has supported the use of the 12 Strategies for Promoting Resilience in Middle School Girls (Mann, 2012, 2013). In each study, these 12 strategies appear to successfully contribute to building resilience in vulnerable young women. Preliminary

evidence suggests this collection of strategies effectively translates theories related to resilience and women’s development into practice in a manner that meets the unique needs of vulnerable early adolescent girls. This study also offers several broader implications for practitioners. First, the study supports the efficacy of genderresponsive programming designed to build girls’ confidence and sense of community. Although adolescent boys and girls share many developmental needs and are well served by a number of common intervention strategies, this study supports the growing body of evidence suggesting that girls who are at higher risk appear to be well served by interventions that also account for gender differences and use genderresponsive strategies (Hawkins et al., 2009; Zahn et al., 2008). Second, a number of resiliency studies suggest the importance of considering context when designing and implementing interventions (Bernard, 1991; Henderson & Millstein, 1996; Tomsen, 2002, Wolin & Wolin, 1993). Qualitative evidence from this study suggests the value of creating self-perpetuating communities that support health and well-being. In this case, because the intervention helped participants create authentic communities of support, girls reported continued group membership in these communities 90 days after program completion. In fact, many girls reported not only that they continued to participate in these communities but also that participation continued to exert a positive influence on their school and academic success. Finally, this study suggests that struggling middle school girls are capable of engaging in and benefiting from frank discussion about the difficult and sensitive issues in their lives. Program satisfaction surveys and focus groups indicate that, even considering the challenging content of the program, girls overwhelmingly rated the experience as valuable and as an experience they would recommend to other girls. In particular, girls described appreciating being able to have adults help them be successful

Downloaded from heb.sagepub.com at UNIV OF GEORGIA LIBRARIES on May 26, 2015

125

Mann et al. in the context of their actual experience, even when that experience included sensitive or challenging topics. Study limitations include using a quasi-experimental design and the relatively short interval between the administration of Group 1’s postprogram outcome survey (Time 2) and follow-up measurement 2 weeks later at Time 3. As part of the crossover design, this strategy effectively provided Group 2 with a comparison group; however, it also provided a short time frame for collecting quantitative evidence related to program effects over time. This strategic decision limited the ability to collect evidence regarding the longer term impact of the program on the study’s quantitative variables. This is especially important as quantitative evidence suggests some reduction in the Group 1 means 2 weeks after program completion and limits our ability to interpret the full impact of the program quantitatively. Finally, using a crossover design may have contributed to underestimating effect size estimates. In conclusion, this second round of formative evaluation supports the efficacy of the REAL Girls program and suggests the value of further study that includes using a true experimental design and including long-term measurement of the program’s impact on associated psychosocial variables and academic outcomes. Additionally, it would be helpful if other researchers replicated this study, particularly in different geographical regions, using the REAL Girls program in full or having developed similar programs using the 12 Strategies for Promoting Resilience in Middle School Girls and/or the model of girls’ resilience as a framework. Acknowledgment Thank you to the many health promotion students that have been leaders in the REAL Girls program and to Patty Haley who gave so generously to this project. You made a truly important difference.

Declaration of Conflicting Interests The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding The authors received no financial support for the research, authorship, and/or publication of this article.

References Aalsma, M. C., & Lapsley, D. K. (2001). A typology of adolescent delinquency: Sex differences and implications for treatment. Criminal Behaviour and Mental Health, 11, 173-191. Acoca, L. (1999). Investing in girls: A 21st century strategy. Juvenile Justice, 6, 3-13. Belenky, M. F., Clinchy, B. M., Goldberger, N. R., & Tarule, J. M. (1986). Women’s ways of knowing: The development of self, voice, and mind. New York, NY: Basic Books. Bernard, B. (1991). Fostering resiliency in kids: Protective factors in the family, school, and community. Portland, OR: Western Center for Drug-free Schools and Communities.

Chesney-Lind, M. (2001). What about the girls? Delinquency programming as if gender mattered. Corrections Today, 63, 38-45. Crosby, R., Salazar, L. F., DiClemente, R. J., Yarber, W. L., Caliendo, A. M., & Staples-Horne, M. (2004). Health risk factors among detained adolescent females. American Journal of Preventive Medicine, 27, 404-410. Gilligan, C. (1993). In a different voice: Psychological theory and women’s development (Rev. ed.). Cambridge, MA: Harvard University Press. Glesne, C. (2006). Becoming qualitative researchers: An introduction (3rd ed.). Boston, MA: Pearson Education. Hatch, J. A. (2002). Doing qualitative research in education settings. New York: State University of New York Press. Hawkins, S. R., Graham, P. W., Williams, J., & Zahn, M. A. (2009). Resilient girls: Factors that protect against delinquency. Washington, DC: U.S. Department of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention. Henderson, N., & Millstein, M. M. (1996). Resiliency in schools: Making it happen for students and educators. Thousand Oaks, CA: Corwin. Johnson, N. G., Roberts, M. C., & Worell, J. (1999). Beyond appearance: A new look at adolescent girls. Washington, DC: American Psychological Association. Jones, B., & Kenward, M. G. (2003). Design and analysis of crossover trials (2nd ed.). London, England: Chapman & Hall. Josselson, R. (1987). Finding herself: Pathways to identity development in women. San Francisco, CA: Jossey-Bass. Kofler, M. J., McCart, M. R., Zajac, K., Ruggiero, K. J., Saunders, B. E., & Kilpatrick, D. G. (2011). Depression and delinquency covariation in an accelerated longitudinal sample of adolescents. Journal of Consulting and Clinical Psychology, 79, 458-469. Lazarus, R. S., & Folkman, S. (1984). Stress, coping, and adaptation. New York, NY: Springer. Lounsbury, J. W., Huffstetler, B. C., Leong, F. T., & Gibson, L. W. (2005). Sense of identity and collegiate academic achievement. Journal of College Student Development, 46, 501-514. Mann, M. J. (2012). Improving struggling middle school girls’ levels of academic self-efficacy, school connectedness, and identity: A pilot study of the REAL Girls program. Chronicle of Middle Level Research, 12, 8-22. Mann, M. J. (2013). Helping middle school girls at risk of school failure recover their self-confidence and achieve school success. Research in Middle Level Education, 36(9), 1-14. McNeely, C. A., Nonnemaker, J. M., & Blum, R. W. (2002). Promoting school connectedness: Evidence from the National Longitudinal Study of Adolescent Health. Journal of School Health, 72, 138-146. Midgely, C., Maehr, M. L., Hruda, L. Z., Anderman, E., Anderman, L., Freeman, K. E., . . . Urdan, T. (2000). The manual for the patterns of adaptive learning scales. Ann Arbor, MI: The University of Michigan. Miller, J. B. (1986). Toward a new psychology of women (2nd ed.). Boston, MA: Beacon Press. Miller, S., Malone, P. S., & Dodge, K. A. (2010). Developmental trajectories of boys’ and girls’ delinquency: Sex differences and links to later adolescent outcomes. Journal of Abnormal Child Psychology, 38, 1021-1032.

Downloaded from heb.sagepub.com at UNIV OF GEORGIA LIBRARIES on May 26, 2015

126

Health Education & Behavior 42(1)

Morse, J. M. (2003). Principles of mixed methods and multimethod research design. In A. Tashakkori & C. Teddle (Eds.), Handbook of mixed methods in social and behavioral research (pp. 189-208). Thousand Oaks, CA: Sage. Mullis, R. L., Cornille, T. A., Mullis, A. K., & Huber, J. (2004). Female juvenile offending: A review of characteristics and contexts. Journal of Child and Family Studies, 13, 205-218. Office of Juvenile Justice and Delinquency Prevention. (1998). Guiding principles for promising female programming. Retrieved from http://www.ojjdp.gov/pubs/principles/contents.html Snyder, C. R., Harris, C., Anderson, J. R., Holleran, S. A., Irving, L. M., Sigmon, S. T., . . . Harney, P. (1991). The will and the ways: Development and validation of an individualdifferences measure of hope. Journal of Personality and Social Psychology, 60, 570-585. Streiner, D. L., & Norman, G. R. (1998). PDQ epidemiology (2nd ed.). Hamilton, Ontario, Canada: B. C. Decker. Stufken, J. (1996). Optimal crossover designs. In S. Ghosh & C. R. Rao (Eds.), Design and analysis of experiments (pp. 63-90). Amsterdam, Netherlands: North-Holland.

Tomsen, K. (2002). Building resilient students: Integrating resiliency into what you already know and do. Thousand Oaks, CA: Corwin Press. Thompson, M. P., Kingree, J. B., & Desai, S. (2004). Gender differences in long-term health consequences of physical abuse of children: Data from a nationally representative survey. American Journal of Public Health, 94, 599-604. Vonesh, E. F., & Chinchilli, V. G. (1997). Linear and nonlinear models for the analysis of repeated measurements. London, England: Chapman & Hall. Wilson, G. S., Pritchard, M. E., & Revalee, M. (2005). Individual differences in adolescent health symptoms: The effects of gender and coping. Journal of Adolescence, 28, 369-379. Wolin, S., & Wolin, S. J. (1996). Beating the odds. Learning, 25, 66-69. Zahn, M. A., Hawkins, S. R., Chiancone, J., & Whitworth, A. (2008). The girls study group: Charting the way to delinquency prevention for girls. Washington, DC: U.S. Department of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention.

Downloaded from heb.sagepub.com at UNIV OF GEORGIA LIBRARIES on May 26, 2015

Improving academic self-efficacy, school connectedness, and identity in struggling middle school girls: a preliminary study of the REAL girls program.

Girls struggling to be successful in middle school are often dealing with negative life experiences that affect their ability to achieve academically...
379KB Sizes 0 Downloads 6 Views