The CYRM-12: a brief measure of resilience.

Liebenberg, Linda ; Ungar, Michael ; LeBlanc, John C. 等

Children and youth develop into mature adults depending on the extent of intrinsic assets such as perseverance, efficacy, self-esteem, and active avoidance of risk-taking behaviours, and extrinsic assets such as living in a nurturing environment with supportive parents, having a non-delinquent peer group and experiencing a healthy school climate. (1-3) When faced with adversity and risk, some youth will survive and even thrive while others will succumb to risky and possibly self-destructive behaviour. Those who thrive under adversity (e.g., poverty, maltreatment, loss of a parent) exhibit engagement in processes described as resilience. (4,5) Measurement of resilience can enable identification of modifiable factors that can be used to inform research and policy initiatives to help youth develop the capacity they require to cope with adversity during normative and non-normative developmental transitions. (6)

The complexity of resilience as a construct, however, makes it challenging to measure. Resilience can be defined as an individual's capacity to navigate to health-enhancing resources that nurture individual, relational, and community assets, as well as the capacity of individuals to negotiate with others for these resources to be provided to them in culturally meaningful ways. (7) This socio-ecological definition implies that individual-, peer-, family-, school- and community-level resources protect and promote good outcomes by helping individuals engage in interactive processes within complex, multi-level environments that make it possible for them to avoid potential threats to their development. (8) Positive development, however, is contextual since a youth may thrive under one adverse circumstance but succumb under another. (9) As well, a youth's ability to cope over time may vary, (1,2,10) particularly during growth and development, and when processes associated with resilience interact with specific risk factors associated with culture, ethnoracial status, ability, gender, and socio-economic status. (11)

The 28-item Child and Health Youth Resilience Measure (CYRM) (12,13) was designed to measure youth resilience while accounting for diverse social contexts across numerous cultures. The CYRM-28 is a self-report instrument validated originally with a purposeful sample of 1,451 youth growing up facing diverse forms of adversity in 11 countries (Canada, USA, Colombia, China, India, Russia, Palestine, Israel, Tanzania, the Gambia, and South Africa). Items are rated on a 5-point scale from 1=does not describe me at all to 5=describes me a lot. Higher scores indicate higher levels of resilience. The final validated CYRM incorporates both cultural homogeneity and heterogeneity in how individuals, families and communities support successful development among youth aged 13-23. (13)

This article details the reduction of the CYRM-28 to a 12-item measure that is better suited for use in omnibus surveys with youth where the full 28-item version of the measure may be unacceptably long. In their recent review of measures of resilience, done before publication of the full validation of the CYRM-28, Windle, Bennet and Noye (14) found 15 published measures (including the CYRM) that captured processes related to resistance to risk impact.

The best-performing measures were all adult-focused. None, Windle et al. note, had included the range of cultural diversity in their development reflected in the CYRM. Most overemphasize individual characteristic without adequately addressing the quality of the social ecology around individuals that provides the trigger for the realization of latent capacity or the development of new coping skills. It is particularly noteworthy that even the best-performing of the measures provided only moderately good validity scores and that most showed a lack of theoretical rationale for their selection of items. Many were developed to measure strengths across an entire population, both those at risk and those not at risk, and are therefore measures of developmental assets rather than resilience. For the most part, the measures are also quite long, frequently with more than 25 items.

METHODS

Two samples of youth were included in the analysis. The first sample of youth consisted of 122 multiple-service-using youth participating in the Pathways to Resilience Study (www.resilienceresearch.org), a cross-sectional, multi-site Atlantic Canadian investigation of youth who use multiple services (child welfare, mental health, juvenile justice, community programs, and special educational supports) and are nominated to the Pathways study by their service providers (see Table 1). Thirty-seven percent of the sample were female and participants were between the ages of 14 and 22 (M = 18 years; SD = 2.017). Data from this sample were used to establish a 12-item version with acceptable validity, using exploratory factor analysis (EFA).

A second sample, drawn from the Survey of Resilience and Risky Behaviours, included 1,574 students from rural and urban public schools in one Atlantic Canadian province, attending grades 7 to 12 and aged between 10 and 18 years (M = 15 years; SD = 1.715); 862 (53%) were girls. Data from this sample were used to conduct a confirmatory factor analysis.

In the Pathways to Resilience study, all youth completed the Pathways to Resilience Youth Measure (PRYM), comprising a battery of validated measures of risk, resilience, service use history, and experiences with caregivers. The CYRM was included. In all instances, the PRYM was administered in a face-to-face interview where each question was read out loud to participants, with explanation of ambiguous terms where necessary and giving participants the option of filling in their responses themselves or having the researcher do it for them. Participants were also given the opportunity to ask for clarification before responding to questions.

In the Survey of Resilience and Risky Behaviours among Youth, all consenting students in class on the day of the survey were administered the Student Drug Use Survey in the Atlantic Provinces (SDUSAP) and the 12-item version of the CYRM developed using EFA on the first sample of youth.

In both studies, Research Ethics Board (REB) approval was obtained from the host institution as well as all participating services and school boards. In all instances, informed consent was obtained in accordance

with the requirements of the host organization. For youth nominated to the Pathways to Resilience study through nominating services where the state was acting guardian of the youth, only informed consent from youth was required. In all other instances, active consent of the parent/legal guardian of youth was obtained in addition to that of the youth.

A multi-step procedure that included review of non-response rates, item variance and EFA with unrotated solutions was conducted repeatedly on the first sample of youth in a process of item reduction. Three iterations of this process were conducted on the first sample of youth, with unrotated EFAs being used in the first two iterations, and varimax rotation during the third iteration. This was followed by a confirmatory factor analysis (CFA) of the remaining 12 items using maximum likelihood estimation and multiple fit indices on a second sample of youth. All analysis was conducted using SPSS (Version 12), PASW Statistics (Version 18) and AMOS (Version 18) for Windows.

RESULTS

In the first iteration of the EFA procedure (n=122), six items were identified as having unacceptably high non-response rates (>10%; see Table 2). A further five items were then identified for elimination due to their lack of variance (see Table 2). An additional five items with extreme means were also identified for deletion. Using the remaining 12 questions, the Kaiser-Meyer-Olkin (KMO) statistic (.652) indicated adequacy of the sample size, and the Bartlett's test was significant (p<0.001) for factor analysis. All 12 items had communality of at least .423 and above. An unrotated solution was used to identify the 10 best-performing items from the original 28-item version of the CYRM (using cut-off values of .45; [alpha]=0.780; see Table 3). While there was good statistical justification for these 10 items, they did not address all domains in our conceptual model of resilience as captured by the CYRM-28. Specifically, none of the 10 items captured concepts related to family and culture-prominent dimensions in the CYRM-28. As such, we reviewed data used in the initial analysis in order to identify why these domains may have been excluded from the EFA. Thirteen youth responded to family-oriented questions as not applicable and all 13 indicated that they lived either on their own or with friends. Recognizing the relationship between these youth and the rates of missing data, these cases were removed from the data set and the data were reanalyzed.

Once again, preliminary data from the 28-item version of the CYRM as administered in the Pathways to Resilience Study (n=122) was used in the analysis with the replacement of the 13 youth with 13 demographically matched participants who had answered the family-oriented questions. Again, non-response and variance on the 28 items was explored. No items could be identified for elimination due to non-response. However, six questions were removed due to their lack of variance, and a further six questions were removed due to extreme means (see Table 2). Using the remaining 16 questions, the KMO statistic(.761) again indicated adequacy of the sample size, and the Bartlett's test was significant (p<0.001) for factor analysis. While 15 of the 16 remaining items have communality of at least .443 and above (see Table 3), "I have people I look up to" could potentially have been considered for elimination as its communality criterion is .332; the item was however included in the analysis. An unrotated factor solution was again used on the remaining 16 items. From this, 10 questions were identified (using cut-off values of .45; [alpha]=0.845; see Table 3) for inclusion in the measure.

Two issues became apparent when comparing the two reduced versions of the CYRM. First, the manner in which items loaded on the various factors was noticeably different when replacing the 13 youth who lived on their own or with friends and who indicated that caregiver questions were not relevant to their lives. While inclusion of these youth meant that family or caregiver questions were not included in the analysis, replacing them with 13 similarly matched youth with different constructions of family meant that these questions featured prominently in the factor analysis load ings. Interestingly, the question "I have people to look up to" appears to have replaced the family and caregiver questions for youth who do not identify caregivers in their lives. Second, questions relating to community supports and self-sufficiency featured more prominently in the factor loadings of the first group of youth (i.e., those not identifying caregivers). These questions include, "I am able to solve problems without harming myself or others (for example by using drugs and/or being violent)", "I think it is important to serve my community", "I am treated fairly in my community", and "I have opportunities to develop skills that will be useful later in life(like job skills and skills to care for family)." Conversely, where family questions did feature prominently in the factor analysis loadings, these community and self-sufficiency questions did not.

Comparing the two versions, it became evident that in order to establish a brief screening measure that would account for all four components of an ecological resilience model and for variations in youth connection to family and culture, a combination of questions from version 1 and version 2 would need to be included. A third version of the measure was therefore constructed containing three questions included in both version 1 and version 2("I know where to go to get help"; "Getting an education is important to me"; and "My friends stand by me during difficult times"). "I feel supported by my friends" was not included because it has a high correlation with "My friends stand by me during difficult times" in both version 1 and version 2(r=.779 and r=.795, respectively; contact the authors for more information regarding these findings). In order to account for variations in connection to family-specifically parents and caregivers-"I have people to look up to", "My parents/caregivers know a lot about me" and "My family will stand by me during difficult times" were included. "My parents/caregivers watch me closely" and "I feel safe when I am with my family" were excluded due to high correlations with "My parents/caregivers know a lot about me" and "My family will stand by me during difficult times". In addition, there were thematic overlaps. Finally, three questions from version 1 were included to measure connection to community: "I think it is important to serve my community", "I have opportunities to develop skills that will be useful later in life (like job skills and skills to care for family)", and "I am treated fairly in my community".

A varimax rotated factor analysis of the 12 items identified for inclusion in the third version resulted in a four-factor solution, with 10 of the items loading well(see Table 4). While communalities on three of the items are very low, they still share at least 23% of the variance with the extracted component. While the reliability of this third grouping([alpha]=0.754) is not as high as in version 2([alpha]=0.845), it is still satisfactory. Combined with the improved content-validity of the measure, it can be argued that version 3 represents a more sophisticated cross-cultural screener of resilience.

A CFA was then undertaken on the 12-item CYRM ("CYRM-12") using data from the second sample of youth who had participated in the Survey of Resilience and Risky Behaviours among Youth (n=1494). Given the requirement for a brief screener of resilience, the analysis was of a model with a single latent variable structure containing all 12 items. Maximum likelihood estimation was used together with multiple fit indices.

[FIGURE 1 OMITTED]

Modification Indices suggested allowing the variables "I am treated fairly in my community" and "I feel I belong at my school" to co-vary, as well as allowing "I have people I look up to" and "My parents/caregivers know a lot about me", and "My parents/caregivers know a lot about me" and "My family stands by me during difficult times" to co-vary (see Figure 1). Once these changes were made to the model, a satisfactory fit was obtained ([chi square](51, N=1540) = 255.419, p=0.0001; Adjusted Goodness of Fit Index = 0.960; Comparative Fit Index = 0.957; Root Mean Square Error of Approximation = 0.050). Cronbach's Alpha for the 12 items was also Satisfactory ([alpha]=0.840).

DISCUSSION

Increasingly, definitions of resilience emphasize processes that occur at multiple systemic levels, with individual, relational, community and cultural factors interacting to produce positive developmental outcomes among populations facing significant adversity. (15) Screening for the most likely individual and contextual capacities that predict positive outcomes has not been possible due to a lack of validated measures that have demonstrated sufficient internal and external validity. Both the CYRM-28 and the briefer CYRM-12 address this gap in the research. Based on two separate samples, one at high risk, the other a population-based sample of school children, the CYRM-12 demonstrates sufficient validity to merit its use as a screener for key resilience characteristics among youth. While the full CYRM-28 provides a more comprehensive understanding of the multiple dimensions of resilience, (13) the 12-item version is well designed for inclusion in larger omnibus studies or smaller clinical trials where researchers seek to document the capacity of adolescents and their social ecologies. This in fact follows a practice seen with many instruments where, for administration in settings with limited resources, brief versions have been developed. For example, the original 93-item Conners Parent Rating Scale for behaviour problems in children (16) was reduced to a 10-item version (17) and the 16-item Kutcher Adolescent Depression Scale was reduced to 6 items. (18)

A strength of the CYRM-12 is that it has been validated on two distinct groups of youth. The first sample is comprised of youth exposed to adversity who have accessed some type of health or community service. One would expect measures designed to capture adversity and resilience to perform well in a sample where exposure to adversity is common. The second is a school-based sample with no attempt to sample based on adversity or access to care. Rates of adversity, mental health conditions, assets and resilience in this sample should be typical of the general population of North American youth since the schools, while not randomly selected, are typical of schools in the province of Nova Scotia. That the confirmatory factor analysis demonstrated good performance of an instrument developed in a clinical sample supports the use of the CYRM-12 in both clinical and non-clinical settings.

Further study will examine whether the CYRM-12 has the potential to inform studies of resilience and risk where the focus is on screening for processes that predict resistance to problem behaviours and other coping strategies. Further study is required, however, to investigate whether the CYRM-12 is appropriate for use with other youth populations across cultures and contexts internationally. While the overall age range in this analysis is 10 to 22, validation of the measure was only conducted on youth aged 10 to 18. Future studies should include a broader age range. A program of research is continuing to investigate these questions.

Acknowledgements: This research was supported by the Public Safety Canada's National Crime Prevention Centre (NCPC) and the Department of Health of Nova Scotia.

Conflict of Interest: None to declare.

REFERENCES

(1.) Armstrong MI, Birnie-Lefcovitch S, Ungar M. Pathways between social support, family well being, quality of parenting, and child resilience: What we know. J Child Fam Stud 2005;14(2):269-81.

(2.) Hanson T, Austin G. Student Health Risks, Resilience, and Academic Performance in California: Year 2 Report, Longitudinal Analyses. Los Alamitos, CA: WestEd, 2003.

(3.) Lerner RM, Dowling EM, Anderson PM. Positive youth development: Thriving as the basis of personhood and civil society. Appl Dev Sci 2003;7(3):172 80.

(4.) Ungar M. The social ecology of resilience: Addressing contextual and cultural ambiguity of a nascent construct. Am J Orthopsychiatry 2011;81(1):1-17.

(5.) Rutter M. Implication of resilience concepts for scientific understanding. In: Lester BM, Masten AS, McEwen B (Eds.), Resilience in Children. Boston, MA: Blackwell, 2006.

(6.) Masten AS, Wright MO. Resilience over the lifespan: Developmental perspectives on resistance, recovery, and transformation. In: Reich JW, Zautra AJ, Hall JS (Eds.), Handbook of Adult Resilience. New York, NY: Guilford, 2010.

(7.) Ungar M. Resilience across cultures. Br J Soc Work 2008;38(2):218-35.

(8.) Fergus S, Zimmerman MA. Adolescent resilience: A framework for understanding healthy development in the face of risk. Annu Rev Public Health 2005;26:399-419.

(9.) Werner EE, Smith RS. Vulnerable but Invincible: A Longitudinal Study of Resilient Children and Youth. New York: McGraw-Hill, 1982.

(10.) Schoon I, Parsons S, Sacker A. Socioeconomic adversity, educational resilience, and subsequent levels of adult adaptation. J Adolesc Res 2004;19(4):383-404.

(11.) American Psychological Association, Task Force on Resilience and Strength in Black Children and Adolescents. Resilience in African American Children and Adolescents: A Vision for Optimal Development. Washington, DC: APA, 2008.

(12.) Ungar M, Liebenberg L. Assessing resilience across cultures using mixed methods: Construction of the Child and Youth Resilience Measure. J Multiple Methods Res 2011;5(2):126-49.

(13.) Liebenberg L, Ungar M, Van de Vijver FRR. Validation of the Child and Youth Resilience Measure-28(CYRM-28) among Canadian youth with complex needs. Res Soc Work Pract 2012;22(2):219-26.

(14.) Windle G, Bennet KM, Noye J. A methodological review of resilience measurement scales. Health Qual Life Outcomes 2011;9:8. doi: 10.1186/1477-75259-8.

(15.) Sanchez-Jankowski M. Cracks in the Pavement: Social Change and Resilience in Poor Neighborhoods. Los Angeles, CA: University of California Press, 2008.

(16.) Conners CK, Sitarenios G, Parker JDA, Epstein JN. The revised Conners' Parent Rating Scale (CPRS-R): Factor structure, reliability, and criterion validity. J Abnorm Child Psychol 1998;26(4):257-68.

(17.) Conners CK. The Conners Rating Scales: Use in clinical assessment, treatment planning and research. In: Maruish ME (Ed.), Use of Psychological Testing for Treatment Planning and Outcome Assessment. Hillsdale, NJ: Lawrence Erlbaum Associates, 1994.

(18.) LeBlanc JC, Almudevar A, Brooks SJ, Kutcher S. The Kutcher Adolescent Depression Scale. J Child Adolesc Psychopharmacol 2002;12(2):113-26.

Received: September 25, 2012

Accepted: January 24, 2013

Linda Liebenberg, PhD, [1] Michael Ungar, PhD, [2] John C. LeBlanc, MD, MSc [3]

Author Affiliations

[1.] Co-Director, Resilience Research Centre, Dalhousie University, Halifax, NS

[2.] Killam Professor of Social Work, Dalhousie University, Halifax, NS

[3.] Associate Professor, Departments of Pediatrics, Psychiatry, Community Health & Epidemiology, Dalhousie University, Halifax, NS

Correspondence: Linda Liebenberg, Co-Director, Resilience Research Centre, Dalhousie University, 1459 Oxford St., Halifax, NS B3H 4R2, Tel: 902-494-1357, Fax: 902-494-7728, E-mail: Linda.Liebenberg@dal.ca

Table 1. Youth Nominations of Sample One by Service Provider

                                   Frequency   Percent

Community-based service provider         47      38.5
Justice                                  51      41.8
Education                                 1       0.8
Child and Family Services                17      13.9
Mental Health and Addictions              6       4.9

Total                                   122     100.0 *

* Percent column does not add to 100% due to rounding.

Table 2. Descriptive Statistics for CYRM-12 Version One and Version
Two

                                    Version 1

                                 N         Mean

1.    I have people I look up   123   3.63
      to
2.    I cooperate with people   123   3.62
      around me
3.    Getting an education is   123   3.96
      important to me
4.    I know how to behave in   123   4.11 ([dagger])
      different social
      situations
5.    My caregiver(s) watch     107   2.94 ([dagger])
      me closely
6.    My caregiver(s) know a    109   3.44
      lot about me
7.    If I am hungry, there     118   3.79
      is enough to eat
8.    I try to finish what I    123   3.71
      start
9.    Spiritual beliefs are a   122   2.52 ([dagger])
      source of strength for
      me
10.   I am proud of my ethnic   123   4.12 ([dagger])
      background
11.   People think that I am    123   4.06 ([dagger])
      fun to be with
12.   I talk to my              109   2.86 ([dagger])
      caregiver(s) about how
      I feel
13.   I am able to solve
      problems without
      harming myself or         122   3.65
      others (for example by
      using drugs and/or
      being violent)
14.   I feel supported by my    123   3.76
      friends
15.   I know where to go in     123   3.94
      my community to get
      help
16.   I feel I belong at my     123   3.21
      school
17.   My caregiver(s)           107   3.51
      stand(s) by me during
      difficult times
18.   My friends stand by me    123   3.84
      during difficult times
19.   I am treated fairly in    122   3.66
      my community
20.   I am given
      opportunities to show     123   4.11 ([dagger])
      others that I am
      becoming an adult and
      can act responsibly
21.   I am aware of my own      123   3.85
      strengths
22.   I participate in          122   1.92 ([dagger])
      organized religious
      activities
23.   I think it is important   122   3.03
      to serve my community
24.   I feel safe when I am     110   3.62
      with my caregiver(s)
25.   I have opportunities to
      develop skills that       123   3.89
      will be useful later in
      life (like job skills
      and skills to care for
      others)
26.   I enjoy my                104   3.49
      caregiver(s)' cultural
      and family traditions
27.   I enjoy my community's    120   2.99 ([dagger])
      traditions
28.   I am proud to be a        123   4.57 ([dagger])
      citizen of Canada

                                           Version 1
                                    Std.
                                 Deviation           Missing

1.    I have people I look up   1.307                   0
      to
2.    I cooperate with people    .928 *                 0
      around me
3.    Getting an education is   1.162                   0
      important to me
4.    I know how to behave in   1.002                   0
      different social
      situations
5.    My caregiver(s) watch     1.459          16 ([double dagger])
      me closely
6.    My caregiver(s) know a    1.410          14 ([double dagger])
      lot about me
7.    If I am hungry, there     1.232                   5
      is enough to eat
8.    I try to finish what I    1.022                   0
      start
9.    Spiritual beliefs are a   1.300                   1
      source of strength for
      me
10.   I am proud of my ethnic   1.120                   0
      background
11.   People think that I am     .813 *                 0
      fun to be with
12.   I talk to my              1.494          14 ([double dagger])
      caregiver(s) about how
      I feel
13.   I am able to solve
      problems without
      harming myself or         1.272                   1
      others (for example by
      using drugs and/or
      being violent)
14.   I feel supported by my    1.064                   0
      friends
15.   I know where to go in     1.189                   0
      my community to get
      help
16.   I feel I belong at my     1.433                   0
      school
17.   My caregiver(s)           1.463          16 ([double dagger])
      stand(s) by me during
      difficult times
18.   My friends stand by me    1.112                   0
      during difficult times
19.   I am treated fairly in    1.218                   1
      my community
20.   I am given
      opportunities to show      .857 *                 0
      others that I am
      becoming an adult and
      can act responsibly
21.   I am aware of my own       .989 *                 0
      strengths
22.   I participate in          1.289                   1
      organized religious
      activities
23.   I think it is important   1.304                   1
      to serve my community
24.   I feel safe when I am     1.478          13 ([double dagger])
      with my caregiver(s)
25.   I have opportunities to
      develop skills that       1.115                   0
      will be useful later in
      life (like job skills
      and skills to care for
      others)
26.   I enjoy my                1.488          19 ([double dagger])
      caregiver(s)' cultural
      and family traditions
27.   I enjoy my community's    1.381                   3
      traditions
28.   I am proud to be a         .967                   0
      citizen of Canada

                                    Version 2

                                 N         Mean

1.    I have people I look up   122   3.77
      to
2.    I cooperate with people   122   3.66
      around me
3.    Getting an education is   122   3.98
      important to me
4.    I know how to behave in   122   4.18
      different social
      situations
5.    My caregiver(s) watch     118   3.03
      me closely
6.    My caregiver(s) know a    121   3.46
      lot about me
7.    If I am hungry, there     122   3.84
      is enough to eat
8.    I try to finish what I    122   3.75
      start
9.    Spiritual beliefs are a   121   2.45 ([dagger])
      source of strength for
      me
10.   I am proud of my ethnic   122   4.12 ([dagger])
      background
11.   People think that I am    122   4.05
      fun to be with
12.   I talk to my              121   2.90 ([dagger])
      caregiver(s) about how
      I feel
13.   I am able to solve
      problems without
      harming myself or         121   3.65
      others (for example by
      using drugs and/or
      being violent)
14.   I feel supported by my    122   3.75
      friends
15.   I know where to go in     122   3.93
      my community to get
      help
16.   I feel I belong at my     122   3.38
      school
17.   My caregiver(s)           119   3.56
      stand(s) by me during
      difficult times
18.   My friends stand by me    122   3.84
      during difficult times
19.   I am treated fairly in    122   3.68
      my community
20.   I am given
      opportunities to show     122   4.11
      others that I am
      becoming an adult and
      can act responsibly
21.   I am aware of my own      122   3.83
      strengths
22.   I participate in          122   1.94 ([dagger])
      organized religious
      activities
23.   I think it is important   122   2.98 ([dagger])
      to serve my community
24.   I feel safe when I am     122   3.65
      with my caregiver(s)
25.   I have opportunities to
      develop skills that       122   3.89
      will be useful later in
      life (like job skills
      and skills to care for
      others)
26.   I enjoy my                114   3.5
      caregiver(s)' cultural
      and family traditions
27.   I enjoy my community's    119   2.98 ([dagger])
      traditions
28.   I am proud to be a        122   4.57
      citizen of Canada

                                      Version 2
                                    Std.
                                 Deviation     Missing

1.    I have people I look up   1.983             0
      to
2.    I cooperate with people    .879 *           0
      around me
3.    Getting an education is   1.178             0
      important to me
4.    I know how to behave in    .936 *           0
      different social
      situations
5.    My caregiver(s) watch     1.461             4
      me closely
6.    My caregiver(s) know a    1.414             1
      lot about me
7.    If I am hungry, there     1.213             0
      is enough to eat
8.    I try to finish what I     .990 *           0
      start
9.    Spiritual beliefs are a   1.323             1
      source of strength for
      me
10.   I am proud of my ethnic   1.154             0
      background
11.   People think that I am     .822 *           0
      fun to be with
12.   I talk to my              1.491             1
      caregiver(s) about how
      I feel
13.   I am able to solve
      problems without
      harming myself or         1.283             1
      others (for example by
      using drugs and/or
      being violent)
14.   I feel supported by my    1.103             0
      friends
15.   I know where to go in     1.179             0
      my community to get
      help
16.   I feel I belong at my     1.439             0
      school
17.   My caregiver(s)           1.459             3
      stand(s) by me during
      difficult times
18.   My friends stand by me    1.153             0
      during difficult times
19.   I am treated fairly in    1.201             0
      my community
20.   I am given
      opportunities to show      .855 *           0
      others that I am
      becoming an adult and
      can act responsibly
21.   I am aware of my own      1.042             0
      strengths
22.   I participate in          1.344             0
      organized religious
      activities
23.   I think it is important   1.298             0
      to serve my community
24.   I feel safe when I am     1.454             0
      with my caregiver(s)
25.   I have opportunities to
      develop skills that       1.122             0
      will be useful later in
      life (like job skills
      and skills to care for
      others)
26.   I enjoy my                1.489             8
      caregiver(s)' cultural
      and family traditions
27.   I enjoy my community's    1.402             3
      traditions
28.   I am proud to be a         .971 *           0
      citizen of Canada

* Items identified for elimination due to lack of variance.

([dagger]) Items identified for elimination due to extreme means.

([double dagger]) Items identified for deletion due to non-response
rates.

Table 3. Communalities, Factor Loadings * and Cronbach's Alpha
Values for Version One and Version Two

                                               Version 1
                                                Factor     Cronbach's
                                  Extraction    Loading     Alpha if
                                                          Item Deleted
1.  I have people I look up to       .635        .684          .750
3.  Getting an education is          .658        .537          .767
      important to me
5.  My caregiver(s) watch me          -           -            -
      closely
6.  My caregiver(s) know a lot        -           -            -
      about me
7.  I eat enough most days           .784         -            -
8.  I try to finish what             .713         -            -
      I start
13. I solve problems without         .423        .470          .775
      drugs or alcohol
14. I feel supported by my           .764        .731          .746
      friends
15. I know where to go to            .606        .466          .773
      get help
16. I feel I belong at my            .442        .527          .771
      school
17. My caregiver(s) stand(s)           -          -            -
      by me during difficult
      times
18. My friends stand by me           .819        .708          .748
      during difficult times
19. I am treated fairly in my        .454        .575          .758
      community
21. I am aware of my own               -          -           -
      strengths
23. I think it is important          .458        .573          .762
      to serve my community
24. I feel safe when I am              -          -           -
      with my caregiver(s)
25. I have opportunities to          .620        .488          .764
      develop job skills
26. I enjoy my caregiver(s)'           -          -           -
      cultural and family
      traditions

                                               Version 2
                                                Factor     Cronbach's
                                  Extraction    Loading     Alpha if
                                                          Item Deleted
1.  I have people I look up to       .332        .354          -
3.  Getting an education is          .593        .539          .838
      important to me
5.  My caregiver(s) watch me         .629        .558          .837
      closely
6.  My caregiver(s) know a lot       .671        .725          .821
      about me
7.  I eat enough most days           .623        .426          -
8.  I try to finish what              -         -              -
      I start
13. I solve problems without         .583        .360          -
      drugs or alcohol
14. I feel supported by my           .710        .577          .837
      friends
15. I know where to go to            .618        .494          .846
      get help
16. I feel I belong at my            .443        .452          .852
      school
17. My caregiver(s) stand(s)         .801        .811          .806
      by me during difficult
      times
18. My friends stand by me           .753        .602          .836
      during difficult times
19. I am treated fairly in my        .429        .392          -
      community
21. I am aware of my own             .566        .419          -
      strengths
23. I think it is important           -           -            -
      to serve my community
24. I feel safe when I am            .818        .794          .807
      with my caregiver(s)
25. I have opportunities to          .666        .379          -
      develop job skills
26. I enjoy my caregiver(s)'         .677        .688          .822
      cultural and family
      traditions

* Extraction method: Principal component analysis.

Table 4. Communalities, Factor Loadings * ([dagger]) and Cronbach's
Alpha Values For Version Three ([double dagger])

                      Extra-          Compnent             Cronbach's
                      ction                                 Alpha if
                                1      2      3      4        Item
                                                            Deleted

1.  I have people      .996           .525                    .759
  I look up to
3.  Getting an         .616                  .752             .733
  education is
  important to
  me
6.  My parent(s)/      .739    .844                           .725
  caregiver(s)
  know a lot about
  me
8. I try to finish     .264           .462   .388             .759
  what I start
13. I solve prob-
  lems without
  harming myself
  or others (by        .528           .608                    .744
  using drugs and/
  or being violent)
15. I know where       .275                         .803      .740
  to go in my
  community to
  get help
16. I feel I           .830                  .816             .735
  belong(ed) at
  my school
17. My family          .820    .885                           .714
  will stand by me
  during difficult
  times
18. My friends         .535           .703                    .727
  stand by me
  during difficult
  times
19. I am treated       .549           .744                    .736
  fairly in my
  community
25. I have opportu-
  nities to develop
  skills that will     .233                         .769      .744
  be useful later
  in life
26. I enjoy my         .748    .773                           .724
  cultural and
  family traditions

* Extraction method: Principal component analysis.

([dagger]) Rotation method: Varimax with Kaiser normalization.

([double dagger]) Rotation converged in 6 iterations.