Ethnocultural factors,resilience,and school engagement

Abstract

In this article we examine how cultural and community factors interact with individual level factors to predict school participation. Participants were 497 Atlantic Canadian youth purposefully selected because of their concurrent use of more than one government service or community program at the time they were interviewed. Results revealed that contextual factors associated with resilience (e.g. cultural adherence and involvement in one’s community) affect school engagement more than individual or relational factors among this population. Furthermore, these contextual resilience factors showed a pattern of differential impact, with the greatest influence occurring in the lives of visible minority youth. Findings suggest that improvements in school engagement are likely to result from school-based efforts to enhance children’s experience of their culture and involvement in community activities. Sampling youth outside regular classroom settings and including meso- and exo-systemic factors in studies of school engagement may help to identify protective processes not previously discussed in the literature.

Keywords

Canada Canadian Atlantic youth culture delinquency exo-systemic factors resilience school engagement service use

School engagement is commonly defined as a multidimensional construct that accounts for students’ behavior at school, emotional attachment to their school, and cognitions related to the value they place on education (Fredricks, Blumenfeld, & Paris, 2004). It has also been shown to be associated with resilience—with higher school engagement predictive of lower rates of substance use, less delinquency, and a more secure attachment to formal learning, though these associations may vary slightly depending upon the socio-economic status of students (Luthar & Ansary, 2005). Typically, factors related to school engagement are studied at the level of the student, the school, or more recently, meso-systemic processes that occur when students and teachers, students and peers, or teachers and parents interact. Ethnocultural factors are frequently referred to in these studies, though usually as individual-level confounding variables that influences patterns of school engagement. For example, when examining the influence of racial bullying by different ethnoracial groups of children in Canada, Larochette and her colleagues (Larochette, Murphy, & Craig, 2010) found that racial bullying and racial victimization that can contribute to students disengaging from school are best predicted by individual-level factors such as the student’s race, but not school-level factors like school climate. Findings such as these suggest that ethnoracial identity is a personal, immutable quality located within the child. Understood this way, it is difficult to think of ethnic, racial, or cultural factors associated with ethnoracial identity as amenable to change by educators.

A second group of researchers have demonstrated sensitivity to ethnoracial diversity by analysing differences between ethnoracial groups when reporting findings. Again, the implication is that these are fixed traits that children come to school with that the school has no control over if it wants to address problems of engagement. Even though visible minorities, notably African Americans, have been shown to have less behavioral engagement and higher emotional engagement (a sense of belonging) than their White peers (Voelkl, 1997), the factors that are studied seldom extend beyond the school itself to community factors that create the social conditions that put some ethnoracial minorities at greater risk for early school leaving, or being less invested in pursuing educational goals.

In an attempt to address this critical gap in research, we employ a social ecological perspective on resilience (Bottrell, 2009; Obrist, Pfeiffer, & Henley, 2010; Ungar, 2011) to look broadly at factors related to an ethnoracial group’s engagement at school. A social ecological perspective of resilience decenters attention from the individual, focusing rather on the proximal processes that make resilience more likely to occur. With regard to ethnicity and race, these factors include whether youth are treated fairly in their communities, feel good about their family traditions, are proud of their ethnicity and national identity, and are provided opportunities in their community to show others they are responsible (Panter-Brick & Eggerman, 2012; Theron et al., 2011). Studies of children’s mental health concerns, such as the study by Chandler, Lalonde, and Sokol (2003) of youth suicide in First Nations (Canadian Indigenous peoples) communities, have shown that ethnicity and race are not risk factors when youth are provided with a community context that promotes cultural adherence through social events, the creation of physical spaces in the community to celebrate the child’s culture, and political rights. Related work by Burack et al. (2013) has shown that school performance among Canadian First Nations youth is also influenced positively by the peer acceptance and self-reported attachment to peers, with both experiences associated with academic achievement.

The social ecology of resilience

Recent advances to the theory of resilience, most notably as a social ecological construct (Ungar, 2012), are contributing to a definition of resilience that directs attention to the processes whereby individuals who face significant challenges interact with their environments to optimize personal success (Ungar & Liebenberg, 2011). As defined by Ungar (2011), resilience occurs when there is significant exposure to adversity, such that protective processes interact with the stressors a child experiences (like racism). In these contexts of stress, resilience is the capacity of children to navigate to the psychological, social, cultural, and physical resources that help them nurture and sustain well-being, and their capacity on their own and with others to negotiate for what they need to be provided in culturally meaningful ways.

There is a growing appreciation for the construct of resilience among school researchers who distinguish between ubiquitous strengths, assets (Edwards, Mumford, & Serra-Roldan, 2007) or buoyancy (Martin & Marsh, 2008) that characterize the capacity of all students to do well at school when facing normative amounts of stress. Appreciation is also rendered for the mechanisms associated with resilience when stress levels are above normal but children cope in ways that are better than expected. Not only, then, do studies of resilience distinguish promotive processes that are good for all children from the protective processes that are relevant to children under the greatest stress, but they also recognize that the influence these processes exert over children will differ by the level of exposure to risk the child experiences (Luthar, Cicchetti, & Becker, 2000). For example, after Hurricane Katrina in the USA, the re-establishment of schools for displaced children fostered a great many mental health benefits that are not typical of children’s everyday experience of attending school when their environment is stable (Heath, Nickerson, Annandale, Kemple, & Dean, 2009). This promotive health enhancing function of school attendance exerts a differential impact on children’s well-being the more children are at-risk.

Informed by the social ecological theory of resilience, we will review research on school engagement that has included examination of ethnoracial factors beyond the school itself that protect children from disengaging from educational institutions.

Contextual factors and school engagement

Among both privileged and non-privileged populations, characteristics like self-esteem, locus of control, and level of participation in school activities are predictive of higher engagement (Finn & Rock, 1997). Self-reports of disengagement, however, are reportedly greatest among students from homes with higher socioeconomic status (SES) and those who are visibly the dominant cultural group (typically those of Anglo-European ethnoracial backgrounds). Lower SES and ethnoracial minority students may report higher rates of school engagement but lower grade point averages (GPA). This pattern is explained by Shernoff and Schmidt (2008) who conducted a five-year longitudinal study of 586 youth from 13 ethnically diverse middle schools and found that while student engagement is a significant predictor of GPA among White students after controlling for SES, GPA and engagement are inversely related for Black students. This paradox can be explained if engagement is understood as a protective factor for students who are under stress (90% of the Black students in the Shernoff and Schmidt study were from lower SES homes). For students who do not experience economic hardship and racial marginalization, school engagement appears to be lower, although these results are inconsistent.

Regardless of which pattern is supported by the research, studies such as these indicate that SES, community safety, racism, and other contextually relevant barriers to adolescents getting an education do exert an influence on school engagement, though not necessarily GPA, which is a measure of an individual trait rather than relational process (Annunziata, Hogue, Faw, & Liddle, 2006). Relatively few studies, however, have examined the relationship between social and economic factors and school engagement (American Psychological Association Task Force on Resilience and Strength in Black Children and Adolescents, 2008; Dotterer, McHale, & Crouter, 2009). Furthermore, most of the studies that have been published examine school engagement among youth who are present in the school system, avoiding the issue of whether the youth who have already disengaged from formal education did so because of unnamed factors.

The most common way exo-systemic factors like racism and its impact on cultural identification are investigated in studies of school engagement is to preselect the student sample by ethnoracial population, most commonly African American and Latino/Latina youth in the United States who are assumed to be at greater risk than their White or Asian American peers (this assumption has been challenged by those who question whether class is a more important factor than race in determining risk) (Shernoff & Schmidt, 2008). Aspects of racial discrimination, poverty, and family dysfunction are then assessed in order to understand their impact on engagement. The tacit assumptions of this methodology (which can reinforce stereotypes of urban racial minorities as disengaged) are: (a) that factors like racial identity, prejudice, and urban poverty are less relevant to students who are of Anglo-European backgrounds (typically defined as White); and (b) within-population differences can be ignored as homogeneity is assumed among young people from very diverse ethnoracial heritages who share vague phenotypical features.

Dotterer et al. (2009) tried to avoid these problems in their investigation of the effect of racial discrimination, racial socialization, and ethnic identity on student relationships, school bonding, and school self-esteem (pride in one’s school accomplishments). Unexpectedly, 12th grade African-American students living in two-parent families reported relatively low levels of racial discrimination at school and relatively good school bonding, ethnic identification, and school self-esteem. Furthermore, racial socialization and ethnic identity positively influenced school engagement and mitigated the impact of racial discrimination when it did occur. Again, while none of these factors affected GPA, higher level of engagement at school was a protective factor for experiences of discrimination related to racial marginalization and lower SES. It is this complex association between ethnoracial background, community factors that support resilience, and school engagement, which is the focus for the present analysis of data drawn from a larger study of risk, resilience, and service use patterns (see Ungar, Liebenberg, Armstrong, Dudding, & van de Vijver, in press).

Purpose

Given that students from ethnoracial minorities show different engagement patterns than majority culture students, we conducted an analysis to investigate the relationship between community and cultural factors related to resilience and their influence on school engagement for at-risk young people. We hypothesized that contextual factors beyond the school that promoted cultural adherence (a positive sense of one’s ethnoracial background) would be more influential on school engagement than individual or relational factors involving caregivers, peers, or teachers. We also hypothesized that the effect of cultural adherence and civic responsibility would be strongest (i.e. show a differential impact) for ethnoracial minorities who face significant structural adversity because of their minority status.

Method

Participants and sampling

Participants were sampled from urban and rural communities in Atlantic Canada between January 2008 and December 2009. Uniquely, this study sampled vulnerable youth engaged with multiple services, rather than mainstream school-based youth. The research team worked with government departments responsible for child welfare services, mental health services, youth criminal justice, community services for homeless youth and educational support services (resource teachers, guidance counsellors, school psychologists, alternative education settings, etc.). Frontline staff identified youth for referral to the study, who were active on their caseloads and were known to have received services from at least two different service providers in the last six months. Permission to share contact details with the research team was obtained by staff from the youth and a legal guardian (when required) prior to nomination. Given the confidential nature of the nomination process, we have no way of knowing what proportion of youth agreed to participate, though agency staff reported a very high rate of interest in the study. To our knowledge, all youth who agreed to have their contact details shared went on to complete the questionnaire.

The diversity of the services and communities in which the study took place meant consent requirements differed substantially across service using populations. At least 15 separate ethics applications were required to comply with each service’s consent process, including those that protect ethnoracial minorities (e.g. ethics boards in Aboriginal communities) and clients under provincial mandates (such as those using juvenile justice and child welfare services).

Results from 497 youth are reported here; of these youth, 281 (57%) were boys with a mean age of 17 years (SD = 1.87). Only 198 participants (40%) lived with both their parents; 80 (16%) lived with a single parent and 219 (44%) had an alternative living arrangement. Three hundred and sixty-eight (75%) were attending school at least part-time when they participated in the study though they may have been sampled from any of the services which nominated youth, not just educational settings. Fifty-five (12%) had graduated. Two hundred and twenty participants (45%) self-identified as visible minorities (i.e. non-White).

Measures

Each of the study variables were assessed using established measures or measures adapted for the purposes of this study and piloted with 40 youth prior to the full study being initiated. Predictor variables included three risk variables, three resilience variables, and a service use variable. School engagement was the outcome variable.

Predictor variables

Risk refers to both community dangers and personal characteristics of youth that reflect exposure to acute or chronic adversity. Measures included the Delinquency sub-scale of the 4HSQ (α = 0.73, rated on a five-point scale from 1 = Never to 5 = 5 or more times), from the 4-H study of Positive Youth Development (Phelps et al., 2007; Theokas & Lerner, 2006). Five items assess youth engagement in theft, property damage, violence, and interaction with the police during the past year. The alpha coefficient in the present study was 0.84. The 12-item version of the Centre for Epidemiological Studies Depression Scale (CES-D-12-NLSCY) (Poulin, Hand, & Boudreau, 2005), a measure of depression risk, is rated on a four-point scale from 0 = Rarely or none of the time to 3 = All of the time (α = 0.85). Items ask how often during the past week the youth has experienced symptoms indicative of depression. The scale in this study had an alpha coefficient of 0.83. Finally, a composite score assessing sense of community danger was established using items from the Boston Youth Survey (BYS; Boston Centers for Youth and Families, 2001). Six items are measured on a four-point scale, assessing levels of neighborhood trust, interaction, monitoring and safety. The alpha coefficient was 0.69.

Resilience was measured using the three sub-scales of the 28-item Child and Youth Resilience Measure (CYRM-28; Liebenberg, Ungar, & Van de Vijver, 2012; Ungar & Liebenberg, 2011). Items are rated on a five-point scale from 1 = does not describe me at all to 5 = describes me a lot and assess. The three subscales are: Individual resources (11 items; α = 0.79) such as personal skills, social skills, and peer support; physical and psychological caregiving by primary caregivers (7 items; α = 0.83); and contextual resources that facilitate sense of belonging (10 items; α = 0.81) including spiritual, cultural and educational resources. For the purposes of this analysis, the items related to school engagement were omitted to avoid redundancy. The alpha coefficient with these items removed was 0.78.

Service use comprised a composite self-report score assessing service use history (i.e. has the youth ever used a service) and frequency of contact with mental health, child welfare, youth corrections (or contact with the police), and educational supports beyond regular classroom programming. Higher scores indicate greater involvement with service providers.

Outcome variable

Degree of school engagement was assessed using the two education items of the CYRM-28 and three items from the Canadian National Longitudinal Survey of Children and Youth (NLSCY), a survey regarding factors influencing children’s social, emotional and behavioral development. Collectively, the five items measured behaviors related to school engagement, attitudes towards education and emotional attachment to school. The alpha coefficient for these items was 0.67.

Procedures

Following signed consent, questionnaires were administered individually and in private with youth. All questions were read out loud regardless of the young person’s ability to read. Researchers met with the youth at school, home, or a community setting such as a recreation center or library. Administration of the questionnaire took approximately 45 minutes. Youth were compensated for their time and any expenses incurred participating in the study.

Data analysis

Analysis of Variance (ANOVA) was used to examine differences in the dependent variable, school engagement, by the ten independent variables. Interactions between the independent variables (resilience, service use, and risk) and their subsequent impact on school engagement were then examined using hierarchical multiple regression analysis. As we had no assumptions regarding the relative importance of the predictor variables in relation to the outcome variable, forced entry analysis was used (Studenmund & Cassidy, 1987). Resources such as resilience and service supports were entered into the model before risk as the focus of the study was on factors that contribute to positive growth and development (resilience).

Results

With regard to our first hypothesis, we were particularly concerned with the relationship between contextual factors that would reasonably be associated with experiences of marginalization among ethnoracial minorities and their relationship to students’ behaviors, emotions, and cognitions related to school engagement.

ANOVA results demonstrated differences in levels of school engagement for all independent variables except the use of school-based services, F(34, 459) = 1.381, p = 0.078, and child welfare, F(34, 458) = 1.327, p = 0.107. Both measures were removed from the regression analysis. Results from the hierarchical regression analysis explained 33% of the variability in school engagement. Interestingly, in this full analysis, the CYRM Context sub-scale (measuring aspects of a child’s cultural adherence and community connections) showed a significant relationship with school engagement, t(490) = 4.840, p = 0.000, while individual characteristics and relationships with caregivers were statistically insignificant. With regards to service use, only involvement with youth criminal justice was significant, t(490) = −2.139, p = 0.033. In terms of risk, both engagement in delinquent behaviour, t(490) = −6.039, p = 0.000, and risk of depression, t(490) = −2.620 p = 0.009, were significant, while sense of community danger was not.

Given the focus of interest, we then ran separate analyses for visible minority and visible majority youth (Tables 1 and 2). The full model accounted for more of the variability in visible minority youth (R² = 0.392) than it did for visible majority youth (R² = 0.320).
Table 1.
Results of hierarchical regression to predict school engagement by resilience, risk and supports for Visible Minority Youth (n = 220).

Variable Model 1 Model 2 Model 3

B SE β β B SE E β B SE E Β

Constant 11.402 2.274 14.391 2.160 15.176 2.884

Resilience

Individual −0.063 0.061 −0.077 −0.008 0.057 −0.010 −0.004 0.055 −0.005

Primary relationships 0.114 0.061 0.131 0.081 0.056 0.094 0.077 0.055 0.089

Context 0.368 0.070 0.416 0.276 0.067 0.312 0.248 0.065 0.281

Service Use

Mental health 0.019 0.133 0.009 0.098 0.135 0.047

Corrections −0.669 0.117 −0.370 −0.348 0.134 −0.193*

Risk

4HSQ delinquency −0.353 0.075 −0.337

CES-D-12-NLSCY 0.014 0.045 0.019

Sense of community danger 0.030 0.107 0.016

R ² 0.205 0.328 0.392

F for change in R² 18.457 19.569 7.387

*
p ≤ 0.05; **p ≤ 0.001.

Table 2.
Results of hierarchical regression to predict school engagement by resilience, risk and supports for Visible Majority Youth (n = 273).

Variable Model 1 Model 2 Model 3

B SE β β B SE E β B SE E β

Constant 5.449 2.058 8.770 2.189 13.290 2.946

Resilience

Individual 0.209 0.057 0.250** 0.175 0.057 .209* 0.126 0.056 0.150*

Primary relationships 0.072 0.058 0.081 0.055 0.057 .062 0.026 0.057 0.029

Context 0.169 0.060 0.196* 0.184 0.059 .214* 0.164 0.057 0.191*

Service use

Mental health −0.211 0.114 −.103 −0.047 0.119 −0.023

Corrections −0.345 0.106 −.179 −0.096 0.127 −0.050

Risk

4HSQ delinquency −0.307 0.071 −0.288

CES-D-12-NLSCY −0.139 0.046 −0.187*

Sense of community danger 0.099 0.085 0.068

R ² 0.195 0.239 0.320

F for change in R² 21.347 7.569 10.190**

*
p ≤ 0.05; **p ≤ 0.001.

With regard to our second hypothesis, that contextual factors would show a differential impact on resilience depending on level of risk exposure, our analysis showed important differences between visible minority and visible majority youth. While context remained statistically significant for the two groups (visible majority, t(267) = 2.873, p = 0.004; visible minority, t(218) = 3.837, p = 0.000), individual characteristics were also significant for visible majority youth, t(267) = 2.231, p = 0.027. Furthermore, engagement with youth criminal justice remained significant only for visible minority youth, t(218) = −2.602, p = 0.010. Finally, while delinquency was significant for both groups (visible majority, t(267) = −4.331, p = 0.000; visible minority, t(218) = −4.702, p = 0.000), depression was significant only for visible majority youth, t(267) = −3.026, p = 0.003.

Discussion

Findings from this study demonstrate that first, as hypothesized, contextual factors associated with resilience affect school engagement more than individual or relational factors among this population of multiple service-using youth. Second, these contextual resilience factors are most influential in the lives of visible minority youth, demonstrating the concept of differential impact. School engagement, therefore, may be influenced most when service providers (including educators) focus attention on the needs of young people for cultural adherence and ethnoracial identification. Individual and relational aspects of marginalized children’s lives may appear to be more amenable to direct intervention; however, interventions that target changing the child’s experience of cultural forces associated with resilience may in fact be more effective at increasing school engagement.

The group of young people in this study, however, are not necessarily the same as those in other studies of youth engagement. The sample selected here are likely more representative of the youth who do not appear regularly in school-based studies. It could, therefore, be speculated that exo-systemic factors that protect against threats to racial identity and marginalization are most influential in the lives of youth who face the most adversity, a situation that is more likely to occur among visible minority youth and other young people who are systemically marginalized, use of multiple services, and experience the least supports to ensure they attend school. The identification of factors related to context that are associated with less school engagement among minority youth in particular suggests that status as a minority is not in and of itself a risk factor, but exposure to specific social conditions that cause problems for minority youth creates the conditions that forecast school disengagement. As case studies of Latino/Latina and Black youth have shown (Brown & Rodriguez, 2009) students of color are not inherently at risk for school dropout, but are at risk because of the social and intellectual alienation that they experience. As Brown and Rodriguez (2009) explain, ‘disengagement from school is a socially mediated phenomenon’ (p. 221). This study’s results suggest that for these students, contextual factors predict whether they will have in their lives the protective factors associated with resilience that are required to engage at school. It should also be noted that while this is an older sample of youth, and that 12% of the youth had already graduated from high school, the retrospective nature of the questions asked ensure that the experiences of these youth as they relate to school engagement are of value to the study. The experiences of those youth who have successfully completed their schooling are particularly important in this regard.

Implications

These patterns suggest the need for educators and policy makers to address the problem of school engagement at multiple levels, placing much more emphasis on contextual factors that influence educational choices than is currently done. There is growing appreciation among professionals concerned with marginalized youth that addressing their social and cultural needs creates noticeable improvements in their functional outcomes such as risky sexual practices, delinquency, and school drop out (Shernoff, & Schmidt, 2008; Shin, Daly, & Vera, 2007).

While most of our efforts to increase school engagement, especially among at-risk groups such as visible minority youth, have targeted school-level variables like teacher-student relationships, peer group selection, bullying, and school climate, our findings suggest that broader contextual factors may also be associated with whether children come to school or disengage. Factors related to pride in one’s cultural traditions, a sense of nationalism, experiences of being treated fairly in one’s community, rites of passage, spirituality and religious affiliation, and other similar factors, may help young people participate in processes that increase their resilience, which in turn affect their school engagement (Bottrell & Armstrong, 2012). Schools themselves, however, may not be able to influence these broader social processes without working in collaboration with other service providers who likely have more contact with the highest need youth, many of whom do not attend school regularly. School psychologists, social workers, and other school-based mental health professionals may be ideally suited for this role, given their training in collaboration and consultation. In addition, educators and school-based clinicians alike can create the conditions that reinforce the processes that nurture resilience and contribute to higher rates of school engagement. For example, community members may be asked to volunteer as hall monitors and mentors to youth who may not feel a connection with cultural outsiders but be willing to engage with adults who share the student’s background. Curriculum may also be used to engage young people by ensuring that course content contributes to a positive cultural identification. In each case, our results suggest the need for schools to see themselves as important to creating cultural continuity for higher risk youth and to creating bridges for youth to participate in activities that bring them recognition from their communities.

Variable	Model 1	Model 2	Model 3
Constant	11.402	2.274	14.391	2.160	15.176	2.884
Resilience
Individual	−0.063	0.061	−0.077	−0.008	0.057	−0.010	−0.004	0.055	−0.005
Primary relationships	0.114	0.061	0.131	0.081	0.056	0.094	0.077	0.055	0.089
Context	0.368	0.070	0.416**	0.276	0.067	0.312**	0.248	0.065	0.281**
Service Use
Mental health	0.019	0.133	0.009	0.098	0.135	0.047
Corrections	−0.669	0.117	−0.370**	−0.348	0.134	−0.193*
Risk
4HSQ delinquency	−0.353	0.075	−0.337**
CES-D-12-NLSCY	0.014	0.045	0.019
Sense of community danger	0.030	0.107	0.016
R ²	0.205	0.328	0.392
F for change in R²	18.457**	19.569**	7.387**

Variable	Model 1	Model 2	Model 3
Constant	5.449	2.058	8.770	2.189	13.290	2.946
Resilience
Individual	0.209	0.057	0.250**	0.175	0.057	.209*	0.126	0.056	0.150*
Primary relationships	0.072	0.058	0.081	0.055	0.057	.062	0.026	0.057	0.029
Context	0.169	0.060	0.196*	0.184	0.059	.214*	0.164	0.057	0.191*
Service use
Mental health	−0.211	0.114	−.103	−0.047	0.119	−0.023
Corrections	−0.345	0.106	−.179**	−0.096	0.127	−0.050
Risk
4HSQ delinquency	−0.307	0.071	−0.288**
CES-D-12-NLSCY	−0.139	0.046	−0.187*
Sense of community danger	0.099	0.085	0.068
R ²	0.195	0.239	0.320
F for change in R²	21.347**	7.569**	10.190**

Footnotes

Author biographies

Michael Ungar, PhD, is a Professor of Social Work at Dalhousie University and Co-Director of the Resilience Research Centre. He has published over 100 peer-reviewed articles and chapters and 11 books on the topic of resilience and its application to clinical and community work with children and families with complex needs (the social ecological approach to counseling). His latest work includes a clinical textbook Counseling in Challenging Contexts, an edited volume of international papers, The Social Ecology of Resilience: A Handbook of Theory and Practice, and a novel The Social Worker.

Linda Liebenberg, PhD, is Co-Director of the Resilience Research Centre, and Adjunct Professor at the School of Social Work, Dalhousie University. She is a methodologist with an interest in image-based methods and mixed-methods designs. Linda’s research examines the use of these methods in understanding the lives of youth living in challenging contexts. She has published and presented internationally on resilience themes relevant to youth across cultures and contexts.

References

American Psychological Association, Task Force on Resilience and Strength in Black Children and Adolescents. (2008). Resilience in African American children and adolescents: A vision for optimal development. Washington, DC: Author. Retrieved from http://www.apa.org/pi/families/resources/task-force/resilience-af-am.aspx.

Annunziata

Hogue

Faw

Liddle

H. A.

(2006) Family functioning and school success in at-risk, inner-city adolescents. Journal of Youth and Adolescence 35(1): 105–113. doi: 10.1007/s10964-005-9016-3.

Boston Centers for Youth and Families (2001) Report of the 2001 Boston Youth Survey, Boston, MA: AuthorAvailable online at: www.cityofboston.gov/BCYF/reports.asp.

Bottrell

(2009) Understanding ‘marginal’ perspectives: Towards a social theory of resilience. Qualitative Social Work 8: 321–340. doi: 10.1177/1473325009337840.

Bottrell

Armstrong

(2012) Local resources and distal decisions: The political ecology of resilience. In: Ungar

(eds) The social ecology of resilience: A handbook of theory and practice, New York, NY: Springer, pp. 247–264.

Brown

T. M.

Rodriguez

L. F.

(2009) School and the co-construction of dropout. International Journal of Qualitative Studies in Education 22: 221–242. doi: 10.1080/09518390802005570.

Burack, J.A., D’Arrisso, A., Ponizovsky, V., Troop-Gordon, W., Mandoor, T., Tootoosis, C., et al. (2013). ‘Friends and grades’: Peer preference and attachment predict academic success among Naskapi youth. School Psychology International, 34(4), 371–386.

Chandler

M. J.

Lalonde

C. E.

Sokol

B. W.

(2003) Personal persistence, identity development, and suicide: A study of Native and non-native North American adolescents. Monographs of the Society for Research in Child Development 68(2): vii–130. doi: 10.1111/1540-5834.00246.

Dotterer

A. M.

McHale

S. M.

Crouter

A. C.

(2009) Sociocultural factors and school engagement among African American youth: The roles of racial discrimination, racial socialization, and ethnic identity. Applied Developmental Science 13: 61–73. doi:10.1080/10888690902801442.

10.

Edwards

O. W.

Mumford

V. E.

Serra-Roldan

(2007) A positive youth development model for students considered at-risk. School Psychology International 28(1): 29–45. doi: 10.1177/0143034307075673.

11.

Finn

J. D.

Rock

D. A.

(1997) Academic success among students at risk for school failure. Journal of Applied Psychology 82: 221–234. doi: 10.1037/0021-9010.82.2.221.

12.

Fredricks

J. A.

Blumenfeld

P. C.

Paris

A. H.

(2004) School engagement: Potential of the concept, state of the evidence. Review of Educational Research 74(1): 59–109. doi: 10.3102/00346543074001059.

13.

Heath

M. A.

Nickerson

A. B.

Annandale

Kemple

Dean

(2009) Strengthening cultural sensitivity in children’s disaster mental health services. School Psychology International 30(4): 347–373. doi: 10.1177/0143034309106944.

14.

Larochette

Murphy

A. N.

Craig

W. M.

(2010) Racial bullying and victimization in Canadian school-aged children: Individual and school level effects. School Psychology International 31(4): 389–408. doi: 10.1177/0143034310377150.

15.

Liebenberg

Ungar

Van de Vijver

F. R. R.

(2012) Validation of the Child and Youth Resilience Measure-28 (CYRM-28) among Canadian youth with complex needs. Research on Social Work Practice 22(2): 219–226. doi: 10.1177/1049731511428619.

16.

Luthar

S. S.

Ansary

N. S.

(2005) Dimensions of adolescent rebellion: Risks for academic failure among high- and low-income youth. Development and Psychopathology 17: 231–250. doi: 10.1017/S0954579405050121.

17.

Luthar

S. S.

Cicchetti

Becker

(2000) The construct of resilience: A critical evaluation and guidelines for future work. Child Development 71(3): 543–562. doi:10.1111/1467-8624.00164.

18.

Martin

A. J.

Marsh

H. W.

(2008) Academic buoyancy: Towards an understanding of students’ everyday academic resilience. Journal of School Psychology 46: 53–83. doi: 10.1016/j.jsp.2007.01.002.

19.

Obrist

Pfeiffer

Henley

(2010) Multi-layered social resilience: A new approach in mitigation research. Progress in Development Studies 10: 283–293.

20.

Panter-Brick

Eggerman

(2012) Understanding culture, resilience, and mental health: The production of hope. In: Ungar

(eds) The social ecology of resilience: A handbook of theory and practice, New York, NY: Springer, pp. 369–386.

21.

Phelps

Balsano

A. B.

Fay

Peltz

J. S.

Zimmerman

S. M.

Lerner

R. M.

Lerner

J. V.

(2007) Nuances in early adolescent developmental trajectories of positive and problematic/risk behaviors: Findings from the 4-H study of positive youth development. Child and Adolescent Psychiatric Clinics of North America 16(2): 473–496.

22.

Poulin

Hand

Boudreau

(2005) Validity of a 12-item version of the CES-D used in the National Longitudinal Study of Children and Youth. Chronic Diseases in Canada 26(2/3): 65–72.

23.

Shernoff

D. J.

Schmidt

J. A.

(2008) Further evidence of an engagement-achievement paradox among US high school students. Journal of Youth and Adolescence 37: 564–580.

24.

Shin

Daly

Vera

(2007) The relationships of peer norms, ethnic identity, and peer support to school engagement in urban youth. Professional School Counseling 10(4): 379–388.

25.

Studenmund

A. H.

Cassidy

H. J.

(1987) Using econometrics: A practical guide, Boston, MA: Little Brown.

26.

Theokas

Lerner

R. M.

(2006) Observed ecological assets in families, schools, and neighborhoods: Conceptualization, measurement and relations with positive and negative developmental outcomes. Applied Developmental Science 10(2): 61–74.

27.

Theron

Cameron

Lau

Didkowsky

Ungar

Liebenberg

(2011) A ‘day in the lives’ of four resilient youths: Cultural roots of resilience. Youth and Society, New York, NY: Springer. doi: 10.1177/0044118X11402853, pp. 369–386.

28.

Ungar

(2011) The social ecology of resilience. Addressing contextual and cultural ambiguity of a nascent construct. American Journal of Orthopsychiatry 81: 1–17.

29.

Ungar

(2012) Social ecologies and their contribution to resilience. In: Ungar

(eds) The social ecology of resilience: A handbook of theory and practice, New York, NY: Springer, pp. 13–32.

30.

Ungar

Liebenberg

(2011) Assessing resilience across cultures using mixed-methods: Construction of the Child and Youth Resilience Measure-28. Journal of Mixed-Methods Research 5(2): 126–149. doi:10.1177/1558689811400607.

31.

Ungar, M., Liebenberg, L., Armstrong, M., Dudding, P., & van de Vijver, F.J. R. (in press). Patterns of service use, individual and contextual risk factors, and resilience among adolescents using multiple psychosocial services. Child Abuse & Neglect.

32.

Voelkl

K. E.

(1997) Identification with school. American Journal of Education 105: 294–318.

Variable	Model 1			Model 2			Model 3
Variable	B	SE β	β	B	SE E	β	B	SE E	Β
Constant	11.402	2.274	14.391	2.160	15.176	2.884
Resilience
Individual	−0.063	0.061	−0.077	−0.008	0.057	−0.010	−0.004	0.055	−0.005
Primary relationships	0.114	0.061	0.131	0.081	0.056	0.094	0.077	0.055	0.089
Context	0.368	0.070	0.416**	0.276	0.067	0.312**	0.248	0.065	0.281**
Service Use
Mental health	0.019	0.133	0.009	0.098	0.135	0.047
Corrections	−0.669	0.117	−0.370**	−0.348	0.134	−0.193*
Risk
4HSQ delinquency	−0.353	0.075	−0.337**
CES-D-12-NLSCY	0.014	0.045	0.019
Sense of community danger	0.030	0.107	0.016
R ²	0.205	0.328	0.392
F for change in R²	18.457**	19.569**	7.387**

Variable	Model 1			Model 2			Model 3
Variable	B	SE β	β	B	SE E	β	B	SE E	β
Constant	5.449	2.058	8.770	2.189	13.290	2.946
Resilience
Individual	0.209	0.057	0.250**	0.175	0.057	.209*	0.126	0.056	0.150*
Primary relationships	0.072	0.058	0.081	0.055	0.057	.062	0.026	0.057	0.029
Context	0.169	0.060	0.196*	0.184	0.059	.214*	0.164	0.057	0.191*
Service use
Mental health	−0.211	0.114	−.103	−0.047	0.119	−0.023
Corrections	−0.345	0.106	−.179**	−0.096	0.127	−0.050
Risk
4HSQ delinquency	−0.307	0.071	−0.288**
CES-D-12-NLSCY	−0.139	0.046	−0.187*
Sense of community danger	0.099	0.085	0.068
R ²	0.195	0.239	0.320
F for change in R²	21.347**	7.569**	10.190**