Family and school influences on adolescents' adjustment: The moderating role of youth hopefulness and aspirations for the future

Journal of Adolescence 44 (2015) 1e16

Contents lists available at ScienceDirect

Journal of Adolescence journal homepage: www.elsevier.com/locate/jado

Family and school influences on adolescents' adjustment: The moderating role of youth hopefulness and aspirations for the future Jean M. Gerard*, Margaret Zoller Booth Bowling Green State University, USA

a r t i c l e i n f o

a b s t r a c t

Article history: Available online 11 July 2015

Using a school-based sample of 675 adolescents, this short-term longitudinal investigation examined the relationships among individual, family, and school influences on adolescent adjustment problems. Adolescents' perceptions of school climate and their sense of connectedness to school were negatively associated with conduct problems. A significant interaction between parental academic support and adolescents' academic aspirations was detected for the total sample, boys, and White youth, indicating that parental support serves a protective function against conduct problems for students with low academic expectations. Adolescents' hopefulness, parental academic aspirations, and school connectedness were negatively associated with depression. Adolescents' hopefulness and their academic aspirations moderated associations between both family and school influences on adolescent adjustment with youth gender and race qualifying these interaction effects. © 2015 The Foundation for Professionals in Services for Adolescents. Published by Elsevier Ltd. All rights reserved.

Keywords: Adolescence Youth hopefulness Academic expectations Conduct problems Depression

Adolescence is a period marked by rapid growth and psychosocial change. In addition to the dramatic physical alterations of puberty, adolescents develop abstract thinking skills, explore possibilities in pursuit of identity, realign relationships with parents as the need for autonomy increases, and navigate significant school transitions. These transformations present opportunities for enhanced development; however, they also present challenges that can trigger vulnerabilities. Although most teenagers traverse these changes successfully without undue stress this developmental stage is marked by increases in behavioral and emotional problems (Hawkins & Monahan, 2009; Roza, Hofstra, Ende, & Verhulst, 2003). Conduct problems are a major concern because adolescents who manifest behavioral problems are also at risk for adverse educational outcomes such as peer rejection, school failure, and dropping out (Maguin & Loeber, 1996; Roeser & Eccles, 2000). Likewise, depression involves emotional and behavioral downswings that can impede social development and scholastic performance (Keyes, 2006). Myriad factors are implicated in the onset of problem behaviors; however, some of the most critical impacts stem from the proximal settings in which adolescents reside including the family and school context (Resnick et al., 1997). Both contexts potentially promote well-being when they provide ample resources and support (e.g., Prelow, Bowman, & Weaver, 2007), yet both are potential arenas of risk that can hinder development when they are unsupportive (Call & Mortimer, 2001) or at odds

* Corresponding author. E-mail address: [email protected] (J.M. Gerard). http://dx.doi.org/10.1016/j.adolescence.2015.06.003 0140-1971/© 2015 The Foundation for Professionals in Services for Adolescents. Published by Elsevier Ltd. All rights reserved.

2

J.M. Gerard, M.Z. Booth / Journal of Adolescence 44 (2015) 1e16

with adolescents' developmental needs (Eccles et al., 1993). At the same time, consideration must be given to individual differences as children exhibit variability in response to environmental demands. Accordingly, we integrate ecological theory with a risk-protection framework to examine associations among quality of educational support systems (i.e., family and school), individual attributes, and adolescent adjustment problems. Drawing from two waves of youth-report data, the purpose of this paper is to: a) examine the unique and relative influence of the family context (i.e., parents' academic support and academic expectations for their children) and school context (i.e., climate and connectedness) on conduct problems and depression; b) determine whether adolescents' hopefulness and aspirations for the future, as individual attributes, serve as buffering agents against conduct problems and depression when youth perceive the family context and school context as unsupportive; and c) assess the equivalence of these associations across youth gender and racial groups. Theoretical framework Our conceptualization of family and school as key social arenas for developing youth is grounded in ecological theorydan approach that is useful for understanding development as it occurs in multiple intersecting ecologies. According to this perspective, the social environment can contribute to positive functioning by exposing children to supportive networks and opportunities for enhanced development; conversely, it can contribute to poor functioning through exposure to unfavorable conditions or by imposing constraints that limit psychosocial growth (Fraser, Kirby, & Smokowski, 2004). As microsystem influences, family and school constitute immediate contexts that directly impact adolescents largely through proximal processes that either restrict or support their development. These proximal processes include key people (e.g., parents, teachers, peers) in the form of interpersonal relationships and key places in the form of safety, satisfaction and opportunity (Bowen, Roderick, Powers, & Glennie, 2008). The quality of experiences in the family and school setting are a matter of great import as it is through these contexts that children not only learn norms for expected behavior but also are afforded the opportunities necessary to acquire the fundamental knowledge, skills, and values that enable them to lead productive lives, engage successfully in social relationships, and develop healthy responses to change. Reflecting a transactional process, ecological theory assumes that development is shaped by the interplay between qualities of the individual and the quality of the environment in which the individual resides (Bronfenbrenner, 1989). This assumption is compatible with a risk-protection framework, the second theoretical perspective undergirding this study, which seeks to understand why some individuals fare poorly in the face of adversity or persistent risk, whereas others seem to escape this risk with little psychological harm. For purposes of this study, risk is defined as a condition within the youth's socialization context that potentially increases the likelihood of unfavorable developmental outcomes, represented in this study as perceived lack of educational support in the familial and school context. Protective factors, on the other hand, are individual or environmental safeguards that shield youth from risk agents by fostering positive outcomes or reducing the likelihood of negative outcomes (Gerard & Buehler, 2004). Three general classes of protective factors have been identified in literature: individual attributes, family support, and external support (Masten & Garmezy, 1985). Given our primary interest in understanding how adolescents' cope with challenging educational/ecological contexts, we focus on individual attributes as possible protective factors, in this case adolescents' hopefulness and academic aspirations e personal assets or psychological strengths that youth may draw upon when they perceive their educational environments as unsupportive. Scholars have identified numerous personal characteristics that serve an ameliorative function in the context of risk. However, little attention has been given to youth hopefulness and academic aspirations e personal characteristics that warrant consideration as protective agents given their potential to enhance psychological well-being through optimism, a sense of purpose, and a goal-seeking orientation. In the present research context, high hope and academic aspirations may be particularly salient as a motivational force and deterrent to socio-emotional problems when educational support in the home and/or school environment is lacking. The emphasis on personal attributes aligns with ecological theory by recognizing adolescents' agency in managing the social contexts in which they are embedded, represented here as an operating belief system based on optimism and goal aspiration (Snyder, 2002). Family context and support for academics To the degree that parents provide sufficient resources and emotional support for their children's ongoing education, the family is a pivotal school-related context for adolescents' socio-emotional functioning (Pomerantz & Moorman, 2010). Based on this guiding premise, we focus on two academic-based aspects of the familial context: parents' academic support and educational aspirations for their teenaged childrendfacets of the familial environment that have been conceptualized in the literature as elements of social capital. Broadly defined, social capital refers to resources derived from social relationships that minimize, prevent, or solve common problems for individuals and communities (Coleman, 1988; Smylie, Medaglia, & Maticka-Tyndale, 2006). Linking this theoretical idea to the ecological framework, Crosnoe (2004) describes social capital as resources that “flow through relationship ties to enhance individual functioning”, which are evident at both the micro-level (i.e., personal relationships) and macro-level (i.e., social networks or institutions; p. 268). Such resources include information channels, support networks, and norms for behavior. According to this view, supportive parentechild networks and access to parents' aspirations facilitate the transmission of resources from parent to child in the form of encouragement, instrumental

J.M. Gerard, M.Z. Booth / Journal of Adolescence 44 (2015) 1e16

3

assistance, and pro-school values (Marjoribanks, 2002). Theoretically, these resources direct adolescents toward positive behavioral choices and enhance their emotional well-being through encouragement and support for academic goals. Parental academic support Considerable attention has been given to the role of parental academic support in facilitating children's academic functioning and, to a much lesser extent, their socio-emotional functioning (see review by Pomerantz & Moorman, 2010). Recognizing this disparity, Pomerantz, Wang, and Ng (2005) contend that parental support of children's scholastic activity is beneficial to children beyond the school arena through the provision of affective resources that foster emotional well-being. To date, most research has focused on parental involvement or parents' allocation of resources to the academic arena of children's lives (Grolnick & Slowiaczek, 1994). Conceived in this manner, involvement can be viewed as a behavioral property of the parentechild relationship reflecting parental investment in the child manifested primarily through instrumental assistance (e.g., help with homework) and active engagement with teachers and the broader school system. Our interest lies more specifically with parental support, which can be viewed as both a behavioral and affective property of the parentechild relationship, in this case regular expression of interest by the parent that is conveyed directly to the child through praise and encouragement for scholastic work in addition to instrumental assistance. Theoretically, parents who offer little in this regard might convey disinterest or send the message that academics are unimportant, potentially steering their children toward disengagement from school and poor behavioral choices. Lacking support from parents in the academic realm, adolescents could react to parental indifference with emotional distress or defiance to elicit parental attentiveness (Gerard, Krishnakumar, & Buehler, 2006). To our knowledge, no studies have examined the direct influence of parental academic support on both conduct problems and depression nor its interaction with youth attributes. Research on school influences has generally focused on the link between parental involvement and children's academic performance whereas research on parenting has focused on the link between general parental support (i.e., warmth and responsiveness) and various indicators of children's adjustment including mental health outcomes and academic performance. However, evidence from the school research literature backs the notion that parental academic support plays a significant role in their children's social-emotional functioning. For instance, Hill and colleagues reported that parental academic involvement in 7th grade was associated with a decrease in school behavioral problems, however, this finding held for middle-class children but not lower class children (Hill et al., 2004). In this study parental involvement was assessed via reports by teachers, mothers, and adolescents. Mothers and teacher reports reflected conventional measures of involvement (e.g., parents' attendance school events, teachers' communication with parents) More in line with our conceptualization of academic support, youth report items reflected personally supportive gestures by parents (e.g., helped children choose classes, held discussions about what children were doing in school). Similarly, Grolnick and colleagues found that personal involvement (i.e., interest in what is going on in the child's school life) but not cognitive (i.e., engaging child in intellectually stimulating activities) or school involvement (e.g., attendance at school events) predicted decreased behavioral problems (Grolnick, Kurowski, Dunlap, & Hevey, 2000; Grolnick & Slowiaczek, 1994). Parental academic aspirations A host of studies have documented a positive linear relationship between parents' educational expectancies for their children and subsequent academic achievement (e.g., Fan & Chen, 2001; Marjoribanks, 2002). In contrast, minimal research attention has been given to parents' aspirations or expectations in relation to children's mental health outcomes. One exception is a study by Resnick and colleagues. Drawing from the National Longitudinal Study on Adolescent Health (ADD Health), Resnick et al. (1997) found that high parental expectations for their teenagers' school achievement were associated negatively with concurrent violence among older adolescents and depressive symptoms among both younger and older adolescents. We extend this study by examining the influence of parents' academic expectations on conduct problems and depression longitudinally and its possible interactive effects with youth hope. School context: climate and connectedness Linkages between students' perception of the school context and indicators of academic performance are welldocumented; however, research has given much less attention to the role of school in adolescents' psychosocial development. Consideration of the school environment is important for several reasons. First, teenagers spend a considerable portion of their waking hours in the school settingda proximal context, much like the family, that provides structure through rules, routines, and potential resources in the form of social capital (i.e., supportive teachers, positive peer affiliations). Second, external support becomes increasingly salient during the teenage years as guidance from non-parental adults is sought. For instance, research has shown that supportive relationships with teachers have a salutary effect on adolescents' scholastic performance as well as their emotional well-being (e.g., Roeser, Eccles, & Sameroff, 1998). Third, the adolescent years, particularly early adolescence, are marked by motivational decline, waning interest in school, and declines in academic performance (Wigfield, Eccles, MacIver, Reuman, & Midgley, 1991) e changes that occur simultaneously with decreased parental involvement in children's schooling (Epstein, 2005; Jackson & Davis, 2000). These conditions might leave youth

4

J.M. Gerard, M.Z. Booth / Journal of Adolescence 44 (2015) 1e16

vulnerable, particularly to conduct problems. Thus, it is reasonable to expect linkages between the school context and adolescents' socio-emotional development, hence the focus on youth perceptions of school climate and connectedness. Eccles and colleagues provide a compelling framework for understanding how socio-emotional problems such as conduct problems and depression can develop in the school context (Eccles & Roeser, 2011). According to a stageeenvironment fit perspective, educational contexts in which youth are expected to learn often lack a positive goodness-of-fit as a result of ineffective school climates that do not foster positive social processes. Most difficult for adolescents, particularly middle school students, are school ecologies that include greater teacher control and a decrease in the quality of teacherestudent relationships, which can contribute to a decrease in academic motivation, a negative self-image, and an increase in behavioral problems (Eccles, Lord, & Midgley, 1991; Roeser & Eccles, 1998). Such conditions are problematic from a developmental standpoint considering the psychological and social needs of adolescents. Hallmark features of this stage of life include an increased need for autonomy, decision-making, and belongingness to peers and non-familial adults. According to Eccles and Roeser (2011), adolescents whose social environments address these changing needs are more likely to experience positive outcomes. School climate School climate is a multi-dimensional construct reflecting various aspects of the educational setting including quality of instruction, studenteteacher relations, fairness, regulation and enforcement of rules (Kuperminc, Leadbeater, Emmons, & Blatt, 1997). Several studies have documented an association between school climate and socio-emotional problems (e.g., Jia et al., 2009; Kuperminc, Leadbeater, & Blatt, 2001; Kuperminc et al., 1997; Roeser et al., 1998; Wang, Selman, Dishion, & Stormshak, 2010). Many climate studies are based on concurrent measures of school climate and adolescent adjustment, thereby focusing on the immediate impact of the school setting. Stronger support for the climateeadjustment relationship comes from longitudinal studies that document a link between school quality and change in children's socio-emotional functioning over time. For instance, in a longitudinal study of 1451 young adolescents, Way, Reddy, and Rhodes (2007) found that middle school students' perceptions of school climate declined from sixth through eighth grade and that these declines were predictive of student declines in psychological and behavioral adjustment. Similarly, in their longitudinal study of an ethnically diverse sample of 7th grade students, Kuperminc et al. (2001) reported a main effect of an unfavorable school climate, as perceived by students, on both externalizing and internalizing problems, findings that were not attributable to prior levels of adjustment. These effects were qualified by interactions with youth attributes (i.e., self-criticism and selfefficacy), supporting our focus on student characteristics that moderate the impact of school climate. School connectedness Both cross-sectional and longitudinal studies have shown that school connectedness, or “the extent to which students feel personally accepted, respected, included, and supported by others in the school social environment” (Goodenow, 1993, p. 80), is associated negatively with numerous indicators of adjustment including violence (Brookmeyer, Fanti, & Henrich, 2006; Resnick et al., 1997), conduct problems (Gerard & Buehler, 2004), depression and anxiety (Anderman, 2002; Shochet, Dadds, Ham, & Montague, 2006). Such associations can be understood when viewing the school context not simply as an educational institution that prepares students academically and vocationally but also as a developmental context where important attachments and support networks are formed e conditions that foster personal well-being (Crosnoe, 2004; Eccles & Roeser, 2011). Catalano and colleagues contend that school plays a central role as one of the primary socialization domains that can deter antisocial behavior and promote healthy development in childhood (Catalano, Haggerty, Oesterle, Fleming, & Hawkins, 2004), an assertion supported by the Seattle Social Development Project (SSDP). Utilizing a quasi-experimental design, these researchers evaluated the effects of increasing social bonding among elementary school students. Compared to the control group, the intervention group demonstrated increases in school connectedness and decreases in behavioral problems when students were in grades one through six. Gains from the intervention had long-lasting impact on students that extended into young adulthood. Other studies from the SSDP have shown that school bonding during the middle and high school years is inversely related to delinquency, violence, and gang membership (Catalano et al., 2004). It is likely that when youth lack bonds to school, they are at an increased risk of aligning with other children who are similarly detached from school and may feel pressure to engage in antisocial behavior. In terms of emotional well-being, Shochet et al. contend that school connectedness is a significant but often overlooked factor in adolescent depression. They reported a longitudinal relationship between low school connectedness and depression for boys and girls after controlling for earlier adjustment. Moreover, prior adjustment did not predict subsequent school connectedness after controlling for students' earlier reports of attachment to school. Likewise, in a longitudinal study of 2678 middle school students, Bond et al. (2007) found that children in 8th grade with low school connectedness were more likely to report depressive symptoms and engage in substance use in 10th grade. Furthermore, the researchers discovered that school connectedness interacted with social connectedness (i.e., the quality of students' friendship networks) such that students with high school connectedness and social connectedness scored lowest on measures of depression and anxiety. Thus, this study provides further support that adolescents' perceptions of the school context interact with other social/ psychological variables to predict their mental health.

J.M. Gerard, M.Z. Booth / Journal of Adolescence 44 (2015) 1e16

5

To summarize this section, theory and research support links between the school context and various adolescent outcomes; however, most research to date has focused on academic achievement. We extend a limited research base by testing main effects of family influences (parents' academic support and academic aspirations) and school influences (climate and connectedness) simultaneously across two indicators of mental health (i.e., conduct problems and depression) and by examining whether these forms of social capital interact with adolescents' outlook for the future to predict later adjustment problems. Youth hopefulness and aspirations as moderators of the family and school context A primary objective of this study is to determine whether youth hopefulness and aspirations for the future serve as protective factors against poor developmental outcomes when parents provide minimal academic support or when their aspirations for children's education are low. Likewise, does hopefulness serve students well when the school climate is perceived negatively or when connections to teachers and students at school are lacking? An assumption of the riskprotection framework is that personal attributes buffer children from the adverse effects of external risk factors and stressful life events, in this case low academic support and educational expectations by parents, low school connectedness, and negative perceptions of the school climate. Hopefulness A sense of hope, purpose, and meaning in life are vital to emotional well-being, evasion of risk behaviors, and goal-directed behaviors and cognitions (Duke, Borowsky, Pettingell, & McMorris, 2011). Hope has received considerable attention by researchers interested in identifying psychological strengths that promote healthy development (Valle, Huebner, & Suldo, 2006), a focus that is warranted given robust associations between hopefulness and a variety of outcomes including academic performance, physical health, psychological adjustment, and self-esteem (Snyder, 2002). However, much of what is known about the salutary effects of hope is based on samples of adults. The extent to which optimism and positive expectations for the future are useful in warding off conduct problems and depressive symptoms in teenagers has not been examined at great length. Yet, there is evidence that children who hold a positive outlook for their future manage stress better than those with less optimism, an indication that hopefulness, as a personal attribute, promotes healthy adaptation and resilience in the face of risk (Wyman, Cowen, Work, & Kerley, 1993). Drawing from Snyder's (2002) hope theory, hopefulness is viewed as an individual characteristic that is cognitive and motivational in nature or, more specifically, as “the perceived capability to derive pathways to desired goals, and motivate oneself via agency thinking to use those pathways” (p. 249). Defined as such this idea ties in with the assumption of ecological theory that individual development is determined in part by the characteristics one brings into his or her social world. Hopeful individuals demonstrate the ability to formulate goals, develop strategies for attaining goals, and possess the motivation necessary for initiating and sustaining effort directed toward goal achievement (Snyder, Lopez, Shorey, Rand, & Feldman, 2003). According to Snyder et al. (2003) children who encounter repeated obstructions to goals and failed attempts to surmount these obstacles are vulnerable to adjustment problems. Research supports this theoretical assertion. Hopefulness among adolescents has been shown to be inversely related to concurrent measures of behavioral and emotional problems (Ruchkin, Eisemann, & Hagglof, 1999; Snyder et al., 1997). Using a large sample of 6th, 9th, and 12-grade students who participated in the 2007 Minnesota Student Survey, Duke and colleagues found a positive contemporaneous association between hopelessness and delinquency, a reliable association evident across several racial groups (Duke et al., 2011). Longitudinal research on the main effects and moderating effects of adolescent hopefulness is sparse. Valle et al. (2006) reported links between hope and decreases in internalizing but not externalizing problems over time. These authors also reported that adolescents' hope buffered the impact of early stressful life events on later internalizing problems (assessed approximately one year later). However, results are by no means conclusive given the paucity of research in this area. Based on extant research, we expect that youth hope will show a stronger direct association with depression than conduct problems. We also expect greater support for the moderating effect of hope when depression is considered as the outcome variable. Youth academic aspirations In addition to a general sense of hopefulness, we also consider adolescents' academic aspirations, a more specific indicator of students' outlook for the future that is in line with the study focus on the linkage between aspects of the academic context and adolescent mental health outcomes. Using ADD Health data, Harris, Duncan, and Boisjoly (2002) found that adolescent students who do not expect to graduate from college are more likely to participate in risk-taking behavior including early sexual activity, selling drugs, and weapon use, providing support for their “nothing to lose” hypothesisdthe idea that youth with low expectations for their future are more likely to engage in risky conduct because they have less to lose than do youth with high expectations. To our knowledge no studies have examined students' academic aspirations as a predictor of adolescent depression or as a moderator of family and school influences. We address this gap by examining possible linkages between students' academic aspirations and both conduct problems and depression and the possibility that students' academic aspirations interact with available social capital in the family (i.e., parents' academic aspirations, parental academic support) and school (climate and connectedness). Research has shown that parents' and children's academic aspirations

6

J.M. Gerard, M.Z. Booth / Journal of Adolescence 44 (2015) 1e16

mutually influence each other when considering their impact on students' academic achievement (Zhang, Haddad, Torres, & Chauansheng, 2011). There is also some evidence that parental expectations interact with the expectations of others in the prediction of academic achievement. For instance, Benner and Mistry (2007) found that high parental expectations mitigated the influence of low teacher expectations on students' academic outcomes. Given the close linkage between negative educational outcomes and conduct problems (Maguin & Loeber, 1996), we expect students' academic aspirations to play a more salient role as a moderator of conduct problems than for depressive symptoms. However, to the extent that there is a mismatch between parents' aspirations for their children and students' own aspirations, adolescents may exhibit emotional distress in the form of depressive symptoms. Macro-level influences: the role of gender and race As a final study objective we examine the role of gender and race as possible moderating factors in the proposed relationships. An ecological framework assumes that developmental trajectories potentially differ by gender and race given variation in behavioral norms, cultural experiences, and socialization processes (Bronfenbrenner, 1989). In terms of gender, research has shown that conduct problems are more prevalent among adolescent boys (Lahey et al., 2000), whereas depressive disorders are more prevalent among girls, a pattern that emerges in middle adolescence (Hankin et al., 1998; Twenge & Nolen-Hoeksema, 2002). Furthermore, gender differences in response to environmental stressors are evident. For instance, Gerard and Buehler (2004) and Call and Mortimer (2001) found that girls who perceived their social environment (i.e., family, peers, school) as unsupportive were more vulnerable to depressive symptoms than were boys. In terms of race, research has shown that African American and Hispanic students tend to perceive the school environment more negatively and less supportive than do White students (KewalRamani, Gilbertson, Fox, & Provasnik, 2007; Ruck & Wortley, 2002)dperceptions that could have bearing on their connectedness or attachment to the school context and, ultimately, the mental health outcomes under consideration. In light of these differences, gender and race are included in this analysis as possible moderating variables in the proposed relationships among family/school variables, youth hope and aspirations, and mental health outcomes. Given how little is known about the role of race and gender in this context, we treat this inquiry as exploratory rather than stating specific hypotheses. Methods The current study is part of a larger four-year longitudinal project that focuses on adolescents' subjective experiences of their academic contexts during middle and high school. The study is situated within a mid-sized city in Northwest Ohio, with racial diversity and lower socio-economic characteristics more closely aligned with national trends than most towns in the region. In 2009e2010 (the first year of data collection), the average daily enrollment of 4213 students in the entire school district (for all grades) was 15.1% Hispanic, 7.5% African-American, 12.8% multi-racial, and 64.0% White. Furthermore, 57.8% of the students were categorized as economically disadvantaged, 6.9% as English language learners, and 1.0% as migrant youth (Ohio Department of Education, 2010). The long-term history of low-income minority migration to the town looking for manual labor has influenced the average level of adult educational attainment. Only 10.5% of adults have attained a bachelor's degree or higher, compared to a national average of 24.7% (U.S. Department of Commerce, 2010). Sampling procedures During fall and spring of the 2009e2010 academic year all students in the middle school (580 7th and 8th grade students) and all 9th and 10th grade students (710) in the high school were invited to participate in the study. Consent letters were sent home with all students, utilizing passive consent procedures, which resulted in an 89% response rate for the middle school and 74% response rate for the high school. This response rate included 13 middle and 64 high school students with negative consent forms and 49 middle and 120 high school students who were either absent or decided for themselves on that day not to complete the survey. As a result, in fall 2009 (wave 1) survey data were collected from 518 middle and 526 high school students (N ¼ 1044); in spring 2010 (wave 2), data were collected from 482 middle and 509 high school students (N ¼ 991), reflecting a retention rate of 85% for the middle school and 86% for the high school for wave 2. These surveys were completed during the first hour of classes and included demographic questions, various measures related to the family and school context, self-appraisals (e.g., academic self-efficacy, self-esteem) as well as other measures not used in the present study. The present study is based on data collected during the second year of the study. Measures of conduct problems and depression were introduced to the youth survey during the second year of the project in an effort to include additional developmental outcomes that have bearing on students' academic performance. Survey data were collected from 986 students in fall 2010 (wave 3) and from 887 students in spring 2011 (wave 4). The sample for this study includes 675 students who participated across the first four waves of data collection (65% of the original sample) and who reported being AfricanAmerican, Hispanic, Multi-racial, or White. Demographic makeup of the sample is as follows: 51% female (n ¼ 343), 49% male (n ¼ 332); 24% 8th graders (n ¼ 164), 26% 9th graders (n ¼ 176), 29% 10th graders (n ¼ 196), 15% 11th graders (n ¼ 103), 5.3% missing data on grade level (n ¼ 36); 6.5% African-American (n ¼ 44), 10.8% Hispanic (n ¼ 73), 19.4% Multi-racial (n ¼ 131), and 63.3% White (n ¼ 427). The demographic makeup of this final sample did not differ markedly from the original fall 2009 sample on gender, race, and school lunch status, an indicator of economic disadvantage.

J.M. Gerard, M.Z. Booth / Journal of Adolescence 44 (2015) 1e16

7

Measurement Outcome measures Conduct problems were assessed using youth reports on 15 items measuring the extent to which they engaged in aggressive or delinquent behavior during the past year. Items were taken from the National Longitudinal Study of Adolescent Health. Students were asked to report how often they engaged in behaviors such as group fighting, shoplifting, and selling drugs. Items are measured on a four-point scale ranging from 0 (never) to 3 (5 or more times). Cronbach's alpha for fall and spring assessments was .89 and .90, respectively. Depressive symptoms were assessed using the Center for Epidemiology Studies Depression Scale for Children (CES-DC), a commonly used index of depression that has evidence of reliability and validity (Radloff, 1977) including cross-cultural equivalency (Tally et al., 2000). The 19-item measure assesses somatic disturbances, interpersonal problems, and depressed affect. Youth reported the degree to which they experienced depressive symptoms during the past week. Sample items include “bothered by things,” “felt lonely,” and “poor appetite.” The four-point scale ranges from 0 (never or rarely) to 3 (most or all of the time). Items expressing positive content (e.g., “felt happy”) were reverse coded so that high values reflect greater depressive affect. Alphas for fall and spring assessments were .86 and .85. Fall measures were used as control variables to partition variance in spring assessments. Youth attributes Hope was assessed with the 10-item hopefulness subscale of the Child/Adolescent Measurement System (Doucette & Bickman, 2002). Sample items include “I feel good about what's going on in my life right now”, “there are people I can count on to help me out if I need it”, and “I am able to accomplish the things I want to do in life”. Reliability of the scale was good (fall a ¼ .90). The four-point response format ranges from 1 (strongly disagree) to 4 (strongly agree). Academic aspirations were assessed with one item asking students to report the highest level of education they would like to receive based on the following categories: “leave school as soon as possible”, “finish high school”, “high school plus some additional professional training”, “some college”, “graduate from college with a 2-year or 4-year degree”, and “postgraduate degree or further education for a special job such as doctor or lawyer”. High scores reflect greater hope and aspirations for the future. Family variables Parental academic support was assessed using the Perceived Family Capital Scale (Marjoribanks, 2002). Sample items for the seven-item scale include “my mother (father) is very interested in my schoolwork”, “my mother (father) often helps me with my homework”, and “my mother (father) often tells me that a good education is important”. The 4-point scale for these items ranges from 1 (strongly disagree) to 4 (strongly agree), thus high scores on this measure reflect greater perceived support. Reliability of these measures was good (fall assessment: .90 mothers, .94 fathers). Youth perceptions of maternal and paternal support were moderately correlated (r ¼ .46), so the two scales were averaged to create a measure of overall parental support. Parents' academic aspirations for their children were assessed by youth with an item that was similar to their self-reported academic aspirations. In addition to the six responses noted above, the one-item measure included an additional categoryd“my parents do not care.” Given low frequencies on this first category, it was collapsed with the second category (“leave school as soon as possible”). Youth perceptions of maternal and paternal aspirations were moderately correlated (r ¼ .57), supporting a combined measure. School variables School climate was assessed with a modified version of Marjoribanks Perceived School Capital Scale (PSCS; Marjoribanks, 2002). The 40-item PSCS assesses student perceptions of the school environment across four dimensions: imaginative (e.g., “Most of my teachers make this school a very exciting place in which to learn”), interpersonal (e.g., “Most of the teachers in this school are very interested in the personal problems of students”), regulative (e.g., “There are many rules and regulations in this school”) and instructional (e.g., “The teachers often seem like they are not very interested in what they are teaching”). In the interest of keeping the youth survey at a manageable level in terms of length and completion time and to eliminate redundancy in questions, a subset of the original 40 items was used (i.e., 29 items; contact first author for more information about the modified measure). Eliminated items were those considered to be redundant or possibly difficult for children to understand. In keeping with original usage, items with a negative valence were reverse-coded, and a total score was created by averaging scores across items. The modified measure has good predictive validity as evidenced by significant correlations with outcomes measures in the expected direction that are consistent with other studies (Kuperminc et al., 2001). Reliability of the PSCS was good (fall a ¼ .93) and is in line with the internal consistency estimate reported in the original study (.87, Marjoribanks, 2002). School connectedness was measured using the mean value of five items that assessed the degree to which youth have friends, a supportive teacher, and trouble with peers (fall a ¼ .63). The 4-point scale for both school variables ranges from 1 (strongly disagree) to 4 (strongly agree), thus high scores reflect more positive perceptions of school climate and a greater sense of connectedness. Youth background characteristics Gender and race were included in analytic models to control for their influence on outcome measures. Boys and girls were coded respectively as zero (reference group) and one. Racial groups included African-American, White, Hispanic, and multiracial students. Three dummy codes were created using White youth as the reference group (coded as 0 and all other groups

8

J.M. Gerard, M.Z. Booth / Journal of Adolescence 44 (2015) 1e16

as 1). As a matter of specificity, separate analyses were conducted by gender and race in addition to analysis of the total sample to determine whether relationships among youth attributes, family/school variables, and outcome measures are similar across boys and girls and across racial groups, which permits identification of common and unique sources of variation in conduct problems and depression across youth subgroups. Analytic procedures All scales and indices were created in SPSS. The degree of missing data varied across variables of interest, but generally constituted fewer than 10% of cases. Missing data were imputed at the scale level using the expectation maximization method in SPSS (EM). EM is a full information method of imputing missing values that uses an iterative procedure to fit the best values. It is preferable to other procedures for handling missing data because it produces less biased results (Acock, 1997). Data were examined using hierarchical multiple regression, a procedure that is useful for identifying variance in a dependent variable attributable to individual variables or sets of variables entered in blocks and for testing statistical interactions between predictor variables. Entry of variables proceeded in the following order: Block 1 e early outcome (i.e., conduct problems or depression); Block 2 e control variables (i.e., gender and race); Block 3 e individual attributes (youth hope and academic expectations), family variables (parental academic support and aspirations), school variables (climate and connectedness); and Block 4 e interactions between family/school variables and individual attributes. Following guidelines by Aiken and West (1991), interaction terms were created by centering individual, family, and school variables on their mean to reduce potential problems with multi-collinearity. Significant interaction terms were probed using simple slope analysis. High and low values on the youth hopefulness scale were represented, respectively, by one standard deviation above and below the centered mean of zero. Given a skewed distribution on the one-item measure of student academic expectations (75% of students expected to graduate from a 2- or 4-year college or pursue postgraduate education), a dichotomous split was used to represent high and low values on student academic expectations (i.e., high ¼ graduate from college with a two year degree or more; low ¼ some college or less). Reported findings are significant at a p-value of .05 or less. Results Descriptive statistics for individual attributes, family and school variables, and indicators of youth maladjustment are presented in Table 1, as well as zero-order correlations between all variables. Each of the fall predictor variables was significantly related in the expected direction to spring outcome measures, with the exception of an insignificant association between youth academic aspirations and depression. Although associations between youth attributes and family/ school context variables are not the central inquiry of this study, it is noted that students' hopefulness and academic aspirations were significantly related to their reports of parental support and parental academic aspirations in addition to their reports of school climate and school connectednessdevidence of predictive validity for the measures utilized in this study. Regression results for total sample Table 2 summarizes results from analyses regressing outcome measures on youth attributes, family/school variables, and specified two-way interaction terms for the entire sample. Reported estimates reflect standardized beta coefficients. Unstandardized coefficients are used to interpret significant interaction terms and their corresponding slopes.

Table 1 Inter-correlations among individual, family, school influences and indicators of youth maladjustment. Variable

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

(9)

(10)

(1) T1 Youth hope (2) T1 Youth academic aspirations (3) T1 Parental academic support (4) T1 Parental academic aspirations (5) T1 School climate (6) T1 School connectedness (7) T1 Conduct problems (8) T1 Depression (9) T2 Conduct Problems (10) T2 Depression M SD

.18** .42** .25** .35** .23** .21** .64** .15** .50** 3.17 .53

.18** .49** .21** .13** .16** .09* .17** .08 5.03 1.00

.32** .36** .23** .19** .34** .14** .22** 3.00 .63

.16** .14** .24** .23** .16** .22** 4.77 1.12

.34** .27** .26** .23** .21** 2.70 .43

.09* .14** .17** .22** 2.72 .97

.23** .54** .23** 3.93 6.13

.13** .67** 15.34 9.87

.23** 4.25 6.60

15.28 8.99

Notes: *p  .05; **p  .01. T1 ¼ fall semester assessment; T2 ¼ spring semester assessment.

J.M. Gerard, M.Z. Booth / Journal of Adolescence 44 (2015) 1e16

9

Table 2 Conduct problems and depression regressed on individual, family, and school variables. Variables

Conduct problems Block 1

Early problem behavior Gender African American Hispanic Multiracial Youth hope (HOPE) Youth academic aspirations (YA) Parent academic support (PS) Parent academic aspirations (PA) School climate School connectedness PS  HOPE PA  HOPE School climate  HOPE School connectedness  HOPE PS  YA PA  YA School climate  YA School connectedness  YA R2 R2D F FD

.45***

.21 .21*** 174.9 174.9***

Block 2 .45*** .01 .00 .01 .02

.21 .00 34.8 .09

Depression Block 3 .40*** .01 .02 .01 .02 .02 .05 .02 .01 .12** .10*

.23 .02** 18.1 3.47**

Block 4a

Block 4b

.41*** .02 .01 .01 .02 .03 .07 .03 .04 .13** .10* .06 .05 .05 .05

.40*** .02 .02 .01 .03 .03 .01 .01 .01 .13* .10*

.24 .01 13.7 1.56

.11* .01 .09 .01 .26 .03*** 14.6 3.29***

Block 1 .61***

.37 .37*** 389.4 389.4***

Block 2 .59*** .08** .03 .04 .04

.37 .01 80.0 2.08

Block 3 .52*** .08** .03 .03 .04 .11** .01 .06 .07* .03 .13***

.40 .03*** 40.2 4.83***

Block 4a

Block 4b

.51*** .08** .03 .04 .04 .12** .02 .06 .09* .04 .12*** .01 .05 .01 .06

.52*** .08** .03 .04 .04 .11** .01 .07* .08* .03 .12***

.41 .00 29.8 1.16

.04 .06 .11 .01 .41 .01** 30.7 3.11**

Note: *p  .05, **p  .01, ***p  .001.

Conduct problems Controlling for early conduct problems, gender, and race (Block 3), school climate and connectedness were negatively related to conduct problems, albeit weakly, supporting our expectation that positive experiences in the school environment, as perceived by youth, would be linked to lower involvement in conduct problems. No other main effects were detected. However, a statistically significant interaction between parental academic support and youth academic aspirations was detected (Block 4b), providing partial support for our expectations of interactive effects. As shown in Fig. 1a, the slope for youth with low academic aspirations was negative such that a one point increment in parental academic support (more support) was associated with a 3.80 decrease in conduct problems (p  .01), whereas the slope for youth with high aspirations was positive (b ¼ .93) but not significant. This finding supports a risk-protection framework; however, the protective influence in this instance is parental academic support, not students' academic aspirations. Depression Turning to the next set of models presented in Table 2, gender was a significant covariate of depression (Block 2) after baseline levels were controlled statistically. The positive coefficient reflects a higher level of depressive symptoms for girls compared to boys. After controlling for earlier depression, gender, and race (Block 3), youth hopefulness, parental academic aspirations, and school connectedness were negatively related to depression, reflecting a weak compensatory effect of these variables on the total sample, and, unexpectedly, parental academic support was positively, but weakly, associated with depression. None of the interaction terms accounted for unique variance in depression.

Gender and racial considerations A primary objective was to determine whether significant variables identified in the total sample reflect common or unique sources of variation across gender and race. To address this objective, regression analyses were run separately for boys and girls (Table 3) and for racial groups (Table 4). In light of small sample sizes and concerns about statistical power, ethnic minority groups were combined for the racial analyses. While we understand that different cultural contexts of youth from various racial/ethnic groups should be treated separately under ideal conditions, sample demographics do not permit reliable statistical methods to do so. Thus, this set of regression results are based on a comparison between White students and students of all other races (African American, Hispanic, and multi-racial). Gender Replicating results from the full sample, positive appraisals of school climate and school connectedness were negatively associated with conduct problems among boys. The interaction between parental academic support and youth academic aspirations was also a significant predictor of conduct problems for boys, suggesting that gender is partially driving the same finding in the total sample. The plotted interaction is shown in Fig. 1b. Boys with low academic aspirations engaged in fewer

10

J.M. Gerard, M.Z. Booth / Journal of Adolescence 44 (2015) 1e16

Fig. 1. Interactions for youth academic aspirations.

Table 3 Conduct problems and depression regressed on individual, family, and school variables by gender: final block (B) with interaction terms. Variables

Conduct problems Boys

Early problem behavior African-American Hispanic Multiracial Youth hope (HOPE) Youth aspirations (YA) Parental support (PS) Parental aspirations (PA) School climate School connectedness (SC) PS  HOPE PA  HOPE School climate  HOPE School connectedness  HOPE PS  YA PA  YA School climate  YA School connectedness  YA R2 F

B4a

B4b

.37*** .03 .02 .02 .04 .09 .09 .05 .14* .13* .01 .06 .06 .04

.35*** .04 .02 .02 .07 .03 .08 .01 .15* .13*

.22 6.40

Depression Girls

.17** .01 .05 .02 .24 7.21

B4a .48*** .09* .02 .06 .04 .05 .04 .00 .08 .00 .08 .03 .04 .05

.31 10.42

Boys B4b .49*** .11* .03 .06 .02 .03 .07 .03 .09 .01

.09 .02 .15 .11 .34 11.85

Girls

B4a

B4b

.41*** .04 .01 .03 .15* .06 .15** .01 .04 .11 .03 .01 .05 .00

.41*** .04 .01 .04 .13* .01 .18** .01 .03 .12**

.26 7.73

.02 .08 .20*** .09 .28 8.66

B4a .56*** .02 .07 .01 .17** .07 .01 .13* .05 .15** .02 .07 .09* .11*

.51 24.61

B4b .59*** .01 .06 .02 .13* .03 .02 .17** .04 .16***

.02 .04 .05 .05 .50 23.30

Note: *p  .05, **p  .01, ***p  .001. Sample size for boys ¼ 330; sample size for girls ¼ 341. F value reflects final block of the model with interaction terms.

conduct problems with increasing levels of parental academic support, further support for the buffering influence of parental support among low-aspiring students (b ¼ 2.65, p  .05). In contrast, boys with high academic aspirations showed a tendency, albeit weak, to engage in more conduct problems with increasing levels of parental academic support (b ¼ 1.76, p  .01). Significant predictors of conduct problems for girls were limited to race. African American girls demonstrated a slightly higher risk for engagement in conduct problems than did White girls. The gender breakdown for depression shows a more complicated picture. For boys, hope and school connectedness were negatively associated with depression. Contrary to expectations, parental academic support was positively associated with depression, suggesting that the provision of support by parents, however manifested, might not always be beneficial to boys or perceived by them as helpful. The interaction between school climate and youth academic aspirations was also statistically significant. As depicted in Fig. 1c, the plotted interaction shows that boys with low academic aspirations benefited significantly from a positive school climate through lower depressive symptoms (b ¼ 4.59, p  .05) in contrast to boys with high academic aspirations who showed insignificant variation in depressive symptoms across levels of school climate (b ¼ .77). For girls, hopefulness, parental academic aspirations, and school connectedness were negatively associated with depression, suggesting that goal-directed cognitions by students, high expectations by parents, and positive school connections exert a compensatory influence in warding off depressive symptoms. In addition to these main effects, significant

J.M. Gerard, M.Z. Booth / Journal of Adolescence 44 (2015) 1e16

11

Table 4 Conduct problems and depression regressed on individual, family, and school variables by race: final block (B) with interaction terms. Variables

Conduct problems White B4a

Early problem behavior Gender Youth hope (HOPE) Youth aspirations (YA) Parental support (PS) Parental aspirations (PA) School climate School connectedness PS  HOPE PA  HOPE School climate  HOPE School connectedness  HOPE PS  YA PA  YA School climate  YA School connectedness  YA R2 F

.33*** .01 .01 .18** .01 .08 .13* .09 .10 .16 .04 .09

.24 11.12

B4b .33*** .00 .01 .11 .02 .06 .13* .08

.21*** .05 .04 .06 .27 12.46

Depression All other races

White

B4a

B4b

B4a

.52*** .07 .05 .04 .02 .02 .07 .01 .01 .10 .02 .04

.55*** .08 .07 .03 .02 .05 .10 .03

.30 8.25

.06 .08 .08 .11 .31 8.89

.53*** .07 .09 .05 .05 .05 .03 .11* .01 .05 .08 .03

.37 20.59

All other races B4b .55*** .07 .09 .07 .06 .06 .01 .11**

.08 .08 .17 .01 .39 21.88

B4a .44*** .10* .19** .12* .08 .31*** .10 .14** .01 .09 .01 .19**

.49 18.71

B4b .44*** .11* .17** .10* .10 .31*** .07 .16**

.03 .03 .01 .02 .48 18.05

Note: *p  .05, **p  .01, ***p  .001. Sample size for racial groups is as follows: White ¼ 425; all other races ¼ 242 (African-American ¼ 42, Hispanic ¼ 71, Multi-racial ¼ 129). F value reflects final block of the model with interaction terms.

interactions between youth hope and both school variables were detected. The interaction between school climate and youth hope (Fig. 2a) reflects a significant decrease in depressive symptoms with increments in school climate among girls in the low-hope group (.3.26, p  .05), indicating a protective effect of school climate. In contrast, a significant increase in depressive symptoms with increments in school climate was found for girls with high hope (3.86, p  .01). The interaction between school connectedness and youth hope (Fig. 2b) shows that positive school connections are beneficial to both girls with high hope and those with low hope; however, this relationship was significantly stronger for girls with low hope (b ¼ for low-hope girls ¼ 2.76, p  .01; b for high-hope girls ¼ .86, p  .05). Overall, though, girls with high hope had fewer depressive symptoms than did girls with low hope across all levels of school connectedness. Race Consistent with results for the total sample and boys, school climate was inversely associated with conduct problems for White students and youth aspirations moderated the association between parental support and conduct problems for this racial group. As shown in Fig. 1d, the interaction reveals a negative slope for low-aspiring White students (b ¼ 5.31, p  .01), or decreases in conduct problems with increments in parental support, additional support for the mitigating influence of parental support among students with low academic aspirations. In contrast, the positive slope for high-aspiring White

Fig. 2. Interactions for youth hopefulness.

12

J.M. Gerard, M.Z. Booth / Journal of Adolescence 44 (2015) 1e16

students (b ¼ 1.12, p  .05) reflects slight increases in conduct problems with increments in parental support. With the exception of early conduct problems, none of the predictor variables accounted for significant variation in conduct problems for minority youth. Turning to depression, school connectedness was the only significant main effect for White youth. Among minority youth, main effects were detected for hopefulness, youth academic aspirations, parental academic aspirations, and school connectedness after controlling for earlier depression, gender, and race. All of these associations were in the expected direction with the exception of youth academic aspirations, suggesting that this variable operates as a risk factor for depression, albeit weakly, for minority students when compared to White students. Youth hope moderated the association between school connectedness and depression. As shown in Fig. 2c, the association between school connectedness and depression was not significant for minority students with high hope (b ¼ .38); however the association for those with low hope was statistically significant (b ¼ 2.66, p  .01). Although the plotted slopes reveal a protective influence of school connectedness for the low hope group, minority youth with high hope had lower depressive symptoms than did those with low hope at all levels of school connectedness. Taking into account early problem behaviors and gender, the set of predictor variables accounted for more variation in depression among minority youth. The set of predictor variables accounted for 37e39% of variance in depression among White youth and 48e49% of variance in minority youth. Discussion A focus on both characteristics of the adolescent and characteristics of the adolescent's social environment is in line with an ecological approach to understanding the relationship between contextual risk and youth maladjustment. According to this theoretical perspective, behavioral and emotional problems arise from the interplay between attributes of the individual and the environment (Bronfenbrenner & Crouter, 1983). Thus, assessing youth problem behaviors with a singular focus on environmental circumstances disregards active contributions by adolescents to adapt to and regulate their experience. Likewise, a sole focus on individual attributes neglects environmental characteristics that shape the quality of experiences available to developing youth. Three key findings emerged from this study, which we use to frame our discussion of results. Unique and relative influence of family and school context variables First, the degree to which adolescents perceive the family and school context as academically supportive has implications for their mental health as evidenced by direct associations between ecological variables and measures of adjustment (main effects). Findings from this study suggest that supportive educational environments e assessed here with student reports of parental academic support, their parents' expectations for educational attainment, school climate, and school connectedness e have more far-reaching impact on youth that extends beyond scholastic performance to their behavioral and emotional well-being. Each of these variables is uniquely related to at least one indicator of adolescent adjustment. Many of the resulting associations are qualified by gender and race, a point we turn to in the next section, so our discussion of main effects is limited. However, the most pervasive influences are school climate and school connectedness, variables that cut across gender and racial groups and two indicators of mental health. In contrast to the widespread influence of school variables, specialized effects emerged for variables representing the family educational climate. Parental academic support serves as a deterrent to conduct problems, particularly for students with low academic aspirations, whereas parents' academic aspirations appear to benefit teenage students most by reducing their risk for depressive symptoms. To date, most of the empirical literature that deals with educational support in the family and school context has focused on children's academic achievement. Thus, programmatic efforts to engage parents in their children's education, improve school climate, and strengthen students' attachments to teachers and peers may benefit youth across multiple dimensions of functioning. Adolescents' hopefulness and academic aspirations as buffering agents Reflecting the second key finding, associations between family/school variables and adolescent mental health outcomes are complex and intertwined with youth attributes as demonstrated by significant statistical interactions with youth hope and academic aspirations. Gender and race qualified most of these interactive relationships e the third key finding; therefore, we merge our discussion of the second and third key finding. We hypothesized that youth hopefulness and high educational aspirations would serve a mitigating role when social capital in the family and school is lacking. In line with our expectations, hopefulness showed protective properties, particularly as a deterrent to depression. A sense of hope and purpose in life are central to healthy development, goal formulation, and avoidance of problematic behaviors. Although the majority of U.S. adolescents are optimistic in their beliefs about the future, a significant percentage experience psychological distress and feelings of hopelessness. Data from the 2009 Youth Risk Survey reveal that 26% of 9the12th grade students reported feeling sad or hopeless almost every day for two or more weeks in a row in the past year, enough to interfere with usual activities; prevalence of these negative emotions was higher for girls compared to boys and for Hispanic and Black youth compared to white youth (Eaton et al., 2010). Mirroring these gender and racial results to some extent, our findings reflect the special meaning of hope in the lives of girls and minority youth. A significant direct longitudinal association between hope and depression, albeit weak, was detected for boys, girls and minority youth. But, on an encouraging note, it appears that the link between low hope and

J.M. Gerard, M.Z. Booth / Journal of Adolescence 44 (2015) 1e16

13

depression for girls and minority youth is offset by positive connections at school. It is important to note that all three of the interactions with hope involved school variables, with two of these being school connectedness. Although students with high hope had lower scores on our index of depression across all levels of school connectednessda testament to the shielding properties of an optimistic belief system e our data also show that social connections at school can offset depressive symptoms when hope is low or lacking. A sense of belonging to others outside of the familial sphere is especially salient for adolescents who, as part of the individuation process, seek extra-familial relationships such as those found in the school setting through attachments to peers and teachers (Shochet et al., 2006). These affirmative social networks appear to be more critical for female students than for boys, perhaps because of gender socialization processes that promote centrality of social relationships among girls (Gilligan, 1982), and for minority youth who, as a group, tend to experience the school setting as biased and unfair (Rowley, Kurtz-Costes, & Cooper, 2010). A more complicated picture emerges when considering the moderating role of students' academic aspirations. All four of the significant interactions indicate that positive educational environments, particularly in the form of parental academic support and school climate, are most beneficial to students with low academic aspirations. In the U.S. one of the major indicators of success in young adulthood is attainment of a college degree. Research indicates that the vast majority of high school students expect to attend college. For instance, data from the National Center for Education Statistics (U.S. Department of Education, 2004) has shown that the percentage of U.S. 10th graders who hope to complete a bachelor's degree or higher doubled from 40% in 1980 to 80% in 2002, with significant increases across racial and ethnic groups and among students from low income families. However, local rates of college attendance will also influence attitudes pertaining to college attainability. The low rate of college attendance in this town (10.5% of adults as compared to a national average of 24.7%) (United States Census, 2010) may be influencing measures of academic aspirations. Spera, Wentzel, and Matto (2008) suggest that parents are often influenced by their own experiences of school when they were a student. Those who had strong positive experiences in school (and are therefore more likely to go on to college) are more likely to demonstrate high academic expectations for their students. Thus, in this study, it is not surprising that overall, students' academic aspirations were significantly associated with their perceptions of parental support and parental academic aspirations (see Table 1). However, parental support appears to have its greatest influence when students have low academic aspirations. As a result, it may be that when students come from family backgrounds that are less likely to produce college graduates, strong parental support serves a protective function, much more than for youth who already feel that college is naturally within reach. Gender and race qualified the interaction between parental support and youth academic aspirations. Given boys' vulnerability to externalizing problems, it is not surprising that educationally supportive measures by parents are more beneficial to low-aspiring boys than to low-aspiring girls by steering them away from antisocial behavior. In terms of race, research indicates that parental academic involvement/ support is a better predictor of academic achievement for White students than for African American and Hispanic students (see review by Ceballo, Huerta, & Epstein-Ngo, 2010). It might be that parental academic support is operating the same way in our sample and that academically supportive efforts by parents have greater impact as a protective agent against the development of conduct problems among low-aspiring White youth. These findings are consistent with a risk-protection framework, demonstrating the buffering influence of the family environment. However, a question that arises from this study is why high academic aspirations among students do not serve as a protective factor or deterrent to the outcomes under consideration. Academic aspirations may be viewed as a desire or hope whereas academic expectations may be thought of as a more realistic belief (Trusty, 2002). Given the distribution of scores on the measure of youth academic aspirations (45% aspire to a 2-year or 4-year college degree, 30% aspire to a postgraduate degree), a significant percentage of youth in the high aspirations group appear to be reporting idealistic goals rather than realistic beliefs about what they expect to accomplish. Research has shown that average expectations of graduating from college decrease from middle school to high school as students develop more realistic expectations (Harris et al., 2002)da pattern found in our own quantitative data. As illustrative of this possibility, a representative sample of 53 students from the current study was interviewed about their future goals as a portion of annual conversations about school. Within those conversations, younger adolescents often lacked understanding of the realistic connection between college and specific professions. For instance, of the high school youth, ninth grade students often did not understand the different levels of academic preparation that would be needed between being a cosmetologist, a forensics expert, or psychologist. Thus, it would behoove researchers who are interested in assessing early and middle adolescents' academic aspirations to include a measure of what adolescents realistically expect as well as a measure of what their ideal educational goals entail. Other gender and racial considerations In addition to the aforementioned qualifying effects of gender and race, a few others deserve attention as they point to different sources of vulnerability and protection for male and female youth and for White and minority youth. As noted previously, high parental academic support appears to be a significant mechanism for warding off conduct problems particularly among boys who aspire to low levels of education; however, in contrast to its inhibitory effect on conduct problems, parental support functioned as a risk factor for depression among boys. Though seemingly counterintuitive and contradictory to a large body of literature that documents the salutary effect of supportive parenting on children's psychosocial development, this finding aligns with research and theory on gender socialization processes that orient boys toward agency and mastery of the environment (Block, 1983; Harter, 1990). Although speculative, supportive measures by parents in

14

J.M. Gerard, M.Z. Booth / Journal of Adolescence 44 (2015) 1e16

the academic realm, however well-intentioned, may be viewed by boys as interference or as a reflection of their academic competence, particularly if these gestures come across as controlling versus autonomy-supportive or performance-oriented versus process-oriented (i.e., emphasis on effort and learning; Pomerantz & Moorman, 2010). From a practical standpoint, this finding suggests that efforts to engage parents in their teenagers' academic lives might be effective in preventing conduct problems in boys, particularly those with low aspirations; however, these gains might be offset by heightened susceptibility to depression among boys, in general, if supportive measures by parents are perceived as psychologically intrusive or limited with respect to promoting independence in schoolwork. Our measure of parental academic support was designed to assess social capital in the family via items that assess the degree to which parents offer praise, support, and verbal encouragement to stay in school. However, other items in this measure, such as the degree to which parents are interested in children's schoolwork and help with homework, could be tapping ‘helicopter parenting’dbehavior indicative of parents who are both highly active and interested in their teenager's schooling but also highly controlling (Pomerantz & Moorman, 2010). Parental involvement in children's homework typically declines during the middle school and high school years due, in part, to adolescents' increasing need for autonomy. Teenagers of parents who maintain a high degree of assistance in this domain of schooling may experience difficulty being autonomous in their schoolwork. In terms of racial differences one notable finding is the positive association between youth academic aspirations and depression among minority youth, an indication, perhaps, of stereotype threat at work. According to Garcia-Coll et al. (1996), consideration of developmental processes for minority youth must take into account aspects of their social experience that are not reflected in the lives of White youth. Mindful of racial and socioeconomic inequalities in educational and occupational attainment, high academic aspirations among minority youth might compete with the reality of membership in a stigmatized group, causing cognitive dissonance, emotional distress and, possibly, fatalism about their educational prospects (Ceballo et al., 2010). In contrast to their own academic aspirations, a fairly strong negative association between parental academic aspirations and depression was found among minority youth. These findings are in line with other studies. For example, Spera et al. (2008) found that minority youth were more influenced than their White peers by their parents' academic aspirations and degree of school connectedness. Other research has found Hispanic youth to be strongly influenced by perceptions of parental academic expectations (Carranza, You, Chhuon, & Hudley, 2009) with Hispanic girls in particular to place great importance on family support and obligations to the family. Therefore, it is not surprising that for the students in this study, there is a stronger association between parents' educational aspirations for their children and depression for minority youth when compared to White youth. Future research should explore the interaction between adolescents' educational aspirations/expectations and those of their parents in relation to adolescents' psychological adjustment. Limitations This investigation is not without limitations. First, all measures are based on youth reports, an approach that potentially introduces bias in the form of shared method variance. Yet, these are worthwhile in the present context. Adolescents' appraisals of their family and school environment might be more powerful predictors of their psychological well-being than the actual quality of these social contexts (Call & Mortimer, 2001; Eccles et al., 1993). Moreover, adolescents are more cognizant than are outside reporters of the conduct problems they engage in across settings and the negative mood states they experience. Nevertheless, outcome measures based on multiple informants are preferable to eliminate method variance as an explanation for findings. Second, small subsample sizes precluded a more detailed analysis of the racial groups represented in this study. To its credit, however, this study provides evidence of racial differences, pointing to the error of assuming that one model of development fits all. Often times, researchers treat gender and race as background variables by controlling statistically for their influence rather than treating them as sources of variation with practical significance. Third, parental aspirations and support for their children's education are likely influenced by parents' level of education and socioeconomic status, measures that were unavailable in our data. However, preliminary analysis using lunch status as a proxy for family SES revealed that this measure was unrelated to outcome measures utilized in this study. It could be that family SES is a more salient predictor of high-school completion or academic performance than problem behaviors. With these limitations in mind, this study contributes to the small but growing body of literature that has demonstrated links between educational support systems in the family and school environment and adolescent mental health. The focus on the interplay between students' hopes/aspirations for their future and their perceptions of the school and family academic climate provides a nuanced understanding of the relationship between adolescents' academic environments and their emotional and social functioning. Findings serve as a reference point for designing supportive educational contexts and direct interventions for youth who are struggling in school. School-based efforts to improve school climate, strengthen students' bonds to teachers and peers, and increase parents' academic support may have broad impact on students' academic achievement; however, results of this study suggest that these efforts may pay off in additional ways by helping vulnerable students evade antisocial behavior and depressive symptoms, particularly those with low educational aspirations and those who believe they have poor prospects for the future. Directed social-psychological interventions that target students' beliefs and feelings about school are also a worthwhile effort. Yeager and Walton (2011) have noted that sometimes only brief, simple interventions that target students' cognitions are necessary to produce significant and lasting gains in children's functioning, particularly in the academic realm. These researchers demonstrated significant gains in academic achievement by “precisely targeting students' experience in school from the student's perspective” (p. 283). In a likewise fashion, Snyder et al. (2003) offer recommendations for school-based hope-enhancing techniques that can be geared to all students or to targeted groups

J.M. Gerard, M.Z. Booth / Journal of Adolescence 44 (2015) 1e16

15

(i.e., students identified as having low levels of hopefulness), which include helping students set goals in whatever form they take (e.g., feeling happier, making new friends, deciding whether or not to go to college), teaching students how to set clear markers for identified goals, helping students identify various routes to a desired goal, and helping students to breakdown broad goals into more manageable subgoals. Finding from the present study and recommended practical applications serve as guiding posts to facilitate mental health adjustment of adolescents as they wrestle with the academic contexts in which they are embedded and the personal beliefs they bring to these social arenas.

References Acock, A. C. (1997). Working with missing values. Family Science Review, 10, 76e102. Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and interpreting interactions. Newbury Park, CA: Sage. Anderman, E. M. (2002). School effects on psychological outcomes during adolescence. Journal of Educational Psychology, 94, 795e809. Benner, A. D., & Mistry, R. S. (2007). Congruence of mother and teacher educational expectations and low-income youth's academic competence. Journal of Educational Psychology, 1, 140e153. Block, J. H. (1983). Differential premises arising from differential socialization of the sexes: some conjectures. Child Development, 54, 1335e1354. Bond, L., Butler, H., Thomas, T., Carlin, J., Slover, S., Bowes, G., et al. (2007). Social and school connectedness in early secondary school as predictors of late teenage substance use, mental health, and academic outcomes. Journal of Adolescent Health, 40, 357.e9e357.e18. Bowen, G. L., Roderick, A. R., Powers, J. D., & Glennie, E. J. (2008). The joint effects of neighborhoods, schools, peers, and families on changes in the school success of middle school students. Family Relations, 57, 504e516. Bronfenbrenner, U. (1989). Ecological systems theory. Annals of Child Development, 6, 187e249. Bronfenbrenner, U., & Crouter, A. C. (1983). The evolution of environmental models in development research. In P. Mussen (Series Ed.) & W. Kassen (Vol. Ed.) (4th ed.,, Handbook of child psychology: Vol. 1. History, theories and methods (pp. 357e414). New York: Wiley. Brookmeyer, K. A., Fanti, K. A., & Henrich, C. C. (2006). Schools, parents, and youth violence: a multilevel, ecological analysis. Journal of Clinical Child and Adolescent Psychology, 35(4), 504e514. Call, K. T., & Mortimer, J. T. (2001). Arenas of comfort in adolescence: A study of adjustment in context. Mahwah, NJ: Erlbaum. Carranza, F. D., You, S., Chhuon, V., & Hudley, C. (2009). Mexican American adolescents' academic achievement and aspirations: the role of perceived parental educational involvement, acculturation, and self-esteem. Adolescence, 44, 313e333. Catalano, R. F., Haggerty, K. P., Oesterle, S., Fleming, C. B., & Hawkins, J. D. (2004). The importance of bonding to school for healthy development. Journal of School Health, 74, 252e261. Ceballo, R., Huerta, M., & Epstein-Ngo, Q. (2010). Parental and school influences promoting academic success among Latino students. In J. L. Meece, & J. S. Eccles (Eds.), Handbook of research on schools, schooling, and human development (pp. 293e307). NY: Routledge. Coleman, J. S. (1988). Social capital and the creation of human capital. American Journal of Sociology, 94, S95eS120. Crosnoe, R. (2004). Social capital and the interplay of families and schools. Journal of Marriage and Family, 66, 267e280. Doucette, A., & Bickman, L. (2002). Child/adolescent measurement system (CAMS). http://www.vanderbilt.edu/VIPPS/CMHP/measurementsys.html. Duke, N. N., Borowsky, I. W., Pettingell, S. L., & McMorris, B. J. (2011). Examining youth hopelessness as an independent risk correlate for adolescent delinquency and violence. Maternal and Child Health Journal, 15(1), 87e97. Eaton, D. K., Kann, L., Kinchen, S., Shanklin, S., Ross, J., Hawkins, J., et al. (2010). Youth risk behavior surveillance e United States, 2009. Center for Disease Control and Prevention Surveillance Reports Data. Eccles, J. S., Lord, S., & Midgley, C. (1991). What are we doing to early adolescents? The impact of educational contexts on early adolescents. American Journal of Education, August, 521e542. Eccles, J. S., Midgley, C., Wigfield, A., Buchanan, C. M., Reuman, D., Flanagan, C., et al. (1993). Development during adolescence: the impact of stageeenvironment fit on young adolescents' experiences in schools and in families. American Psychologist, 48(2), 90e101. Eccles, J. S., & Roeser, R. W. (2011). Schools as developmental contexts during adolescence. Journal of Research on Adolescence, 21, 225e241. Epstein, J. L. (2005). School-initiated family and community partnerships. In T. Erb (Ed.), This we believe in action: Implementing successful middle level schools (pp. 77e96). Westerville, OH: National Middle School Association. Fan, X., & Chen, M. (2001). Parental involvement and students' academic achievement: a meta-analysis. Educational Psychology Review, 13, 1e13. Fraser, M. W., Kirby, L. D., & Smokowski, P. R. (2004). Risk and resilience in childhood. In M. W. Fraser (Ed.), Risk and resilience in childhood: An ecological perspective (pp. 13e66). Washington, DC: National Association of Social Workers Press. Garcia-Coll, C., Lamberty, G., Jenkins, R., McAdoo, H. P., Crnic, K., Wasik, B. H., et al. (1996). An integrative model for the study of developmental competencies in minority children. Child Development, 67, 1891e1914. Gerard, J. M., & Buehler, C. (2004). Cumulative environmental risk and youth problem behavior. Journal of Marriage and Family, 66, 702e720. Gerard, J. M., Krishnakumar, A., & Buehler, C. (2006). Marital conflict, parentechild relations, and youth maladjustment: a longitudinal investigation of spillover effects. Journal of Family Issues, 27, 951e975. Gilligan, C. (1982). In a different voice. Cambridge, MA: Harvard University Press. Goodenow, C. (1993). Classroom belonging among early adolescent students: relationship to motivation and achievement. Journal of Early Adolescence, 13, 21e43. Grolnick, W. S., Kurowski, C. O., Dunlap, K. G., & Hevey, C. (2000). Parental resources and the transition to junior high. Journal of Research on Adolescence, 10, 465e488. Grolnick, & Slowiaczek. (1994). Parents' involvement in children's schooling: a multidimensional conceptualization and motivational mode. Child Development, 65, 237e252. Hankin, B. L., Abramson, L. Y., Moffitt, E. E., Silva, P. A., McGee, R., & Agnell, K. E. (1998). Development of depression from preadolescence to young adulthood: emerging gender differences in a 10-year longitudinal study. Journal of Abnormal Psychology, 107, 128e140. Harris, K. M., Duncan, G. J., & Boisjoly, J. (2002). Evaluating the role of “nothing to lose” attitudes on risky behavior in adolescence. Social Forces, 80(3), 1005e1039. Harter, S. (1990). Adolescent self and identity development. In S. S. Feldman, & G. R. Elliott (Eds.), At the threshold: The developing adolescent (pp. 352e387). Cambridge, MA: Harvard University Press. Hawkins, D., & Monahan, K. (2009). Covariance of problem behaviors in adolescence. Paper presented at IOM Committee on the Science of Adolescence Workshop, Washington, DC. Hill, N. E., Castellino, D. R., Landsford, J. E., Nowlin, P., Dodge, K. A., Bates, J. E., et al. (2004). Parent-academic involvement as related to school behavior, achievement, and aspirations: demographic variations across adolescence. Child Development, 75, 1491e1509. Jackson, A., & Davis, P. G. (2000). Turning points 2000: Educating adolescents in the 21st century. New York: Teachers College Press. Jia, Y., Way, N., Ling, G., Yoshikawa, H., Chen, X., Hughes, D., et al. (2009). The influence of student perceptions of school climate on socioemotional and academic adjustment: a comparison of Chinese and American adolescents. Child Development, 80, 1514e1530. KewalRamani, A., Gilbertson, L., Fox, M., & Provasnik, S. (2007). Status and trends in the education of racial and ethnic minorities (NCES 2007-039). Washington, D.C.: National Center for Education Statistics, Institute of Education Sciences, U.S. Department of Education. Keyes, C. L. M. (2006). Mental health in adolescence: is America's youth flourishing? American Journal of Orthopsychiatry, 76(3), 395e402.

16

J.M. Gerard, M.Z. Booth / Journal of Adolescence 44 (2015) 1e16

Kuperminc, G. P., Leadbeater, B. J., & Blatt, S. J. (2001). School climate and individual differences in vulnerability to psychopathology among middle school students. Journal of School Psychology, 39, 141e159. Kuperminc, G. P., Leadbeater, B. J., Emmons, C., & Blatt, S. J. (1997). Perceived school climate and difficulties in the social adjustment of middle school students. Applied Developmental Science, 1, 76e88. Lahey, B. B., Schwab-Stone, M., Goodman, S. H., Waldman, I. D., Canino, G., Rathouz, P. J., et al. (2000). Age and gender differences in oppositional behavior and conduct problems: a cross-sectional household study of middle childhood and adolescence. Journal of Abnormal Psychology, 109, 488e503. Maguin, E., & Loeber, R. (1996). Academic performance and delinquency. In M. Tonry (Ed.), Crime and justice: A review of research (Vol. 20, pp. 145e264). Chicago: University of Chicago Press. Marjoribanks, K. (2002). Family and school capital: Towards a context theory of students' school outcomes. Norwell, MA: Kluwer Academic Publishers. Masten, A. S., & Garmezy, N. (1985). Risk, vulnerability, and protective factors in developmental psychopathology. In B. B. Lahey, & A. E. Kazdin (Eds.), Advances in clinical child psychology (Vol. 8, pp. 1e52). New York: Plenum Press. Ohio Department of Education. (2010). 2009e2010 School year report card. www.ode.state.oh.us/reportcardfiles/2009-2010/DIST/044016.pdf. Pomerantz, E. M., & Moorman, E. A. (2010). Parents' involvement in children's schooling: a context for children's development. In J. L. Meece, & J. S. Eccles (Eds.), Handbook of research on schools, schooling, and human development (pp. 398e416). NY: Routledge. Pomerantz, E. M., Wang, Q., & Ng, F. F. (2005). Mothers' affect in the homework context: the importance of staying positive. Developmental Psychology, 41, 414e427. Prelow, H. M., Bowman, M. A., & Weaver, S. R. (2007). Predictors of psychological well-being in urban African American and European American youth: the role of ecological factors. Journal of Youth and Adolescence, 36(4), 543e553. Radloff, L. S. (1977). The CES-D scale: a self-report depression scale for research in the general population. Applied Psychological Measurement, 1, 385e401. Resnick, M. D., Bearman, P. S., Blum, R. W., Bauman, K. E., Harris, K. M., Jones, J., et al. (1997). Protecting adolescents from harm: findings from the national longitudinal study on adolescent health. Journal of the American Medical Association, 278, 823e832. Roeser, R. W., & Eccles, J. S. (1998). Adolescents' perceptions of middle school: relation to longitudinal changes in academic and psychological adjustment. Journal of Research on Adolescence, 8(1), 123e158. Roeser, R. W., & Eccles, J. S. (2000). Schooling and mental health. In A. J. Sameroff, & M. Lewis (Eds.), Handbook of developmental psychopathology (Vol. 2, pp. 135e156). New York: Kluwer Academic/Plenum Press. Roeser, R. W., Eccles, J. S., & Sameroff, A. J. (1998). Academic and emotional functioning in early adolescence: longitudinal relations, patterns, and prediction by experience in middle school. Development and Psychopathology, 10, 321e352. Rowley, S. J., Kurtz-Costes, B., & Cooper, S. M. (2010). The schooling of African American children. In J. L. Meece, & J. S. Eccles (Eds.), Handbook of research on schools, schooling, and human development (pp. 275e292). NY: Routledge. Roza, S. J., Hofstra, M. B., Ende, J., & Verhulst, F. C. (2003). Stable prediction of mood and anxiety disorders based on behavioral and emotional problems in childhood: a 14-year follow-up during childhood, adolescence, and young adulthood. American Journal of Psychiatry, 160, 2116e2121. Ruchkin, V. V., Eisemann, M., & Hagglof, B. (1999). Hopelessness, loneliness, self-esteem, and personality in Russian male delinquent adolescents versus controls. Journal of Adolescent Research, 14, 466e477. Ruck, M. D., & Wortley, S. (2002). Racial and ethnic minority high school students' perceptions of school disciplinary practices: a look at some Canadian findings. Journal of Youth and Adolescence, 31, 185e195. Shochet, I. M., Dadds, M. R., Ham, D., & Montague, R. (2006). School connectedness is an underemphasized parameter in adolescent mental health: results of a community prediction study. Journal of Clinical Child and Adolescent Psychology, 35, 170e179. http://dx.doi.org/10.1207/s15374424jccp3502_1. Smylie, L., Medaglia, S., & Maticka-Tyndale, E. (2006). The effect of social capital and socio-demographics on adolescent risk and sexual health behaviours. Snyder, C. R. (2002). Hope theory: rainbows in the mind. Psychological Inquiry, 13(4), 249e275. Snyder, C. R., Hoza, B., Pelham, W. E., Rapoff, M., Ware, L., Danovsky, M., et al. (1997). The development and validation of the children's hope scale. Journal of Pediatric Psychology, 22, 399e421. Snyder, C. R., Lopez, S. J., Shorey, H. S., Rand, K. L., & Feldman, D. B. (2003). Hope theory, measurements, and applications to school psychology. School Psychology Quarterly, 18, 122e139. Spera, C., Wentzel, K. R., & Matto, H. C. (2008). Parental aspirations for their children's educational attainment: relations to ethnicity, parental education, children's academic performance, and parental perceptions of school climate. Journal of Youth and Adolescence, 38, 1140e1152. Tally, S., Beam, M., Farruggia, S., Whang, S. M., Dong, Q., & Chen, C. (2000, April). Cross-cultural measurement equivalency of the CES-D in five adolescent populations. Poster session presented at the biennial meeting of the Society for Research on Adolescence. Trusty, J. (2002). African Americans' educational expectations: longitudinal causal models for women and men. Journal of Counseling and Development, 80, 332e345. Twenge, J. M., & Nolen-Hoeksema, S. (2002). Age, gender, race, socioeconomic status, and birth cohort differences on the children's depression inventory: a meta-analysis. Journal of Abnormal Psychology, 111, 578e588. U.S. Census. (2010). State and city quick facts. http://quickfacts.census.gov/qfd/states/39/3928826.html. U.S. Department of Commerce. (2010). U.S. Census, 2010. http://quickfacts.census.gov/qfd/states/39/3928826.html. U.S. Department of Education, National Center for Education Statistics. (2004). The condition of education 2004 (NCES 2004-077). Washington, DC: U.S. Government Printing Office. Valle, M. F., Huebner, E. S., & Suldo, S. M. (2006). An analysis of hope as a psychological strength. Journal of School Psychology, 44(5), 393e406. Wang, M., Selman, R. L., Dishion, T. J., & Stormshak, E. A. (2010). A Tobit regression analysis of the covariation between middle school students' perceived school climate and behavioral problems. Journal of Research on Adolescence, 20(2), 274e286. Way, N., Reddy, R., & Rhodes, J. (2007). Students' perceptions of school climate during the middle school years: associations with trajectories of psychological and behavioral adjustment. American Journal of Community Psychology, 40, 194e213. Wigfield, A., Eccles, J. S., MacIver, D., Reuman, D. A., & Midgley, C. (1991). Transitions during early adolescence: changes in children's domain-specific selfperceptions and general self-esteem across the transition to junior high school. Developmental Psychology, 27(4), 552e565. Wyman, P. A., Cowen, E. L., Work, W. C., & Kerley, J. H. (1993). The role of children's future expectations in self-system functioning and adjustment to life stress: a prospective study of urban at-risk children. Development and Psychopathology, 5, 649e661. Yeager, D. S., & Walton, G. M. (2011). Socialepsychological interventions in education: they're not magic. Review of Educational Research, 81(2), 267e301. Zhang, Y., Haddad, E., Torres, B., & Chauansheng, C. (2011). Reciprocal relationships among parents' expectations, adolescents' expectations, and adolescents' achievement: a two-wave longitudinal analysis of the NELS data. Journal of Youth and Adolescence, 40, 479e489.