توافقنامه آگاهی متقابل از مقیاس رتبه رفتاری و عاطفی برای جوانان در تنظیمات بهداشت روانی

Comprehensive assessment of youths' emotional and behavioral functioning includes obtaining data from multiple sources, such as parents and youth. Despite the shift in focus on youths' strengths and the increased availability of strength-based assessments, few studies have examined the cross-informant agreement between multiple raters of youths' behavioral and emotional strengths. Thus, the purpose of this study was to evaluate the cross-informant agreement between parent and youth ratings on the Behavioral and Emotional Rating Scale-Second Edition (BERS-2). The current study extends previous cross-informant research by examining the cross-informant agreement between parent and self-report ratings for youth served in community mental health centers and whether differences in cross-informant agreement exist between youth with and without a school-identified disability. Results indicated that cross-informant agreement on youths' strengths was acceptable, as most obtained correlations were greater than those typically reported on cross-informant agreement on deficit-based instruments. Furthermore, small but significant differences in cross-informant agreement for youth with and without a school-identified disability were observed for the BERS-2 Affective Strengths and School Functioning subscales. Overall, findings provide support for the reliability of multiple informants' ratings on the BERS-2 for measuring the strengths of youth referred for community mental health services.

Assessment of youths' behavioral and emotional functioning within community mental health organizations typically involves the administration of behavior ratings using multiple informants, such as parent report and youth self-report (De Los Reyes and Kazdin, 2005 and Renk, 2005). The benefit of having multiple perspectives is that clinicians are able to obtain a holistic understanding of the youth's functioning, as there is no one “gold standard” for measuring youths' emotional and behavioral functioning (De Los Reyes & Kazdin, 2005). By having a comprehensive view of the youth, assessment data can better inform treatment decisions and better outcome evaluation (Renk, 2005). However, when clinicians use information from multiple perspectives they must also consider the level of cross-informant agreement, or the similarity between reports by individuals with different perspectives, experiences, and information (Achenbach, 2006). Research on the cross-informant agreement of deficit-based behavior rating scales suggests that the degree to which multiple raters agree is typically modest. In their seminal meta-analysis, Achenbach, McConaughy, and Howell (1987) investigated the extent to which multiple informants agreed in their ratings of youths' behavioral and emotional problems. Their findings indicated that average agreement between self-report and ratings by another individual (as measured by weighted correlation coefficients) was .22. Furthermore, average agreement for raters with similar roles (r = .60; e.g., two teachers) was higher than for ratings between informants in different roles (r = .28; e.g., caregiver and youth). These findings have been replicated in several studies, demonstrating that there are more differences than similarities in informants' report of the emotional and behavioral deficits of youth (e.g., Gresham et al., 2010, Rescorla et al., 2013, Salbach-Andrae et al., 2009 and Youngstrom et al., 2000). Despite that agreement among raters is modest, researchers have concluded that this not due to the unreliability of the information raters provide, but due to more substantively meaningful differences regarding how informants process information and make decisions about rating items. Based on the Attribution Bias Context Model (De Los Reyes & Kazdin, 2005), informant discrepancies arise from three main sources: (a) the perceived cause of problematic behavior, (b) the threshold for how severe a problem must become before intervention is warranted, and (c) the context in which the informant observes the behavior of interest. These underlying factors are highly likely to differ across youth and caregiver informants, leading to the modest cross-informant agreement repeatedly observed in the literature. Thus, informant discrepancies may have an impact on the assessment and treatment of mental health difficulties by influencing decisions concerning service eligibility, progress monitoring, and treatment efficacy (De Los Reyes, 2011). Within community mental health settings, youth and their caregiver(s) may be the only informants to which providers have access. However, relying solely on one informant's ratings can lead to incomplete conclusions about a youth's functioning (De Los Reyes & Kazdin, 2005). It is for this reason that best clinical practice is to collect data from a multi-source, multi-method, multi-setting approach in which self-report measures completed by youth are an essential component of assessment (Whitcomb & Merrell, 2013). Although the cross-informant agreement of deficit-based measures is widely studied, much less is known about multiple raters' agreement of strength-based measures. Strength-based assessment refers to the evaluation of the emotional and behavioral characteristics, skills, and competencies that strengthen an individual's ability to manage stress and adversity, foster valuable relationships with others, and develop feelings of personal achievement (Epstein, 2004). A strength-based approach is consistent with the belief that (a) youth with deficits also have strengths, (b) youth and families may be more motivated and empowered during treatment if strengths are incorporated, and (c) the absence of a strength in an area is not indicative of a weakness or deficit in that domain (Epstein, 2004). Furthermore, research indicates that when youths' strengths are measured, clinicians can better develop, implement, and monitor interventions (Tedeschi & Kilmer, 2005). For example, the results of a strength-based assessment can be used to inform an approach to treatment wherein youth are taught strategies to use their strengths to cope with difficult situations (Rashid & Ostermann, 2009). Several assessments are available to measure youths' strengths; however, one of the most commonly used standardized measures (Nickerson & Fishman, 2013) is the Behavioral and Emotional Rating Scale-Second Edition (BERS-2; Epstein, 2004). The BERS-2 is a nationally normed rating scale that measures the emotional and behavioral strengths of youth. The BERS-2 can be completed by teachers, parents, and youth ages 11–18 years in approximately 10 min. There are 52 positively worded items that measure youths' Interpersonal Strengths (ability to interact with others), Family Involvement (relationships with family members), Intrapersonal Strengths (youths' perception of his/her accomplishments), Affective Strengths (ability to give and receive affection), and School Functioning (competence in school). Scores from the five subscales combine to produce the Total Strength Index, which is an estimate of the youth's overall emotional and behavioral strengths. The psychometric properties of the BERS-2 scores are well established ( Benner et al., 2008, Epstein, 2004 and Walrath et al., 2004). However, few studies have investigated the cross-informant agreement of the BERS-2, and those that do exist only included youth within school settings. The cross-informant agreement research on the BERS-2 has generally revealed greater levels of agreement than what is typically found with deficit-based measures. For instance, correlation coefficients measuring the degree of agreement between parents and teachers of 20 youth with school-identified serious emotional disturbance were .54 to .67 for all subscales except Intrapersonal Strengths (r = .20, ns; Friedman, Leone, & Friedman, 1999). Similarly, correlation coefficients measuring the extent of agreement between parents and self-report ratings for youth in general education were .50 to .63 for all BERS-2 scales ( Synhorst, Buckley, Reid, Epstein, & Ryser, 2005). Findings from two studies evaluating the cross-informant agreement of the Finnish version of the BERS-2 revealed most agreement coefficients were in the moderate to large range for parent–teacher, parent–youth, and teacher–youth dyads in elementary school settings ( Sointu et al., 2012a and Sointu et al., 2012b). However, research has yet to evaluate the cross-informant agreement between parent and self-report ratings for youth with identified mental health problems served within community mental health settings. Furthermore, the BERS-2 was normed on a representative sample drawn from the United States (US) school population, which consisted of primarily typically-developing youth and a much smaller percentage of youth with school-identified disabilities. Well-accepted test standards established by professional organizations indicate that when an instrument is used with a population not included in the original standardization, its reliability and validity need to be re-established ( American Educational Research Association, American Psychological Association, & National Council on Measurement in Education, 1999). Extant research suggests greater agreement on the majority of the Finnish BERS-2 subscales when informants rated youth who received special education services and supports than for those youth who did not receive special education supports (Sointu et al., 2012b). The potential moderating role of a youth's special education disability status may have implications for the interpretation of assessment data gathered by mental health practitioners. However, there are no studies evaluating differences in cross-informant agreement youth referred for community mental health services based on the presence or absence of a school-identified disability in the US. The purpose of the current study was to evaluate the cross-informant agreement of the BERS-2 for youth in community mental health settings. More specifically, we investigated the degree to which parent ratings and youth self-report ratings were similar. A second purpose of this study was to examine potential differences in cross-informant agreement on the BERS-2 based on whether youth received special education services due to a school-identified disability.