Innovations in Strength-Based Social-Emotional Assessment: Factor Analysis, Psychometric Analysis, and Cross-Informant Comparisons with the SEARS-T

• Main Author: Cohn, Bradley P., 1983-
• Language: en_US
• Published: University of Oregon 2012
• Subjects: Educational psychology; Clinical psychology; Education; Psychology; Resilience; Strength-based assessment
• Online Access: http://hdl.handle.net/1794/11925

xiv, 110 p. A print copy of this thesis is available through the UO Libraries. Search the library catalog for the location and call number. === Many youth under the age of 18 experience high levels of mental health problems, and very few of those youth receive the necessary services to combat those problems. Historically, assessment of behavior and social and emotional functioning and subsequent design of interventions occur using deficit-based measures and tools. Another method of assessing behavior and social and emotional functioning that is receiving more attention over the last decade is strength-based assessment and service delivery. The Social Emotional Assets and Resiliency Scales (SEARS) is a new multi-informant strength- based behavior-rating system currently being developed and researched at the University of Oregon. To assess the factor structure, psychometric properties, and cross-informant correlation of the teacher version of the SEARS, data were gathered from elementary, middle, and high schools throughout the United States. Teachers ( n = 1673) were asked to rate students in their classes in several domains of social and emotional functioning (e.g., problem solving, social skills, empathy, and self-regulation). Results of the exploratory and confirmatory factor analysis indicate that the SEARS-T is made up of four strong factors--Responsibility, Self-Regulation, Social Competence, and Empathy. Analysis of reliability of total scores reflects very strong internal consistency (α = .98) and test-retest reliability ( r = .94). Reliability of factor scores also reflects strong internal consistency and test-retest reliability. Cross-informant reliability with the SEARS-T indicates relatively weak correlations between teacher reports and child self-reports based on the Pearson-product moment correlation ( r = .37). Analyses of group differences were carried out for grade, student gender, rater gender, disability status, ethnicity, rater setting, and teacher categorization of academic performance. Results indicate teacher ratings differed based on student gender, disability status, rater setting, and academic performance. Results from this study indicate the SEARS-T is a psychometrically sound measure with a solid factor structure. With an understanding of the need for continued research, the SEARS-T appears to be culturally valid and useful for research and applied purposes. === Committee in charge: Kenneth Merrell, Chairperson, Special Education and Clinical Sciences; Krista Chronister, Member, Counseling Psychology and Human Services; Deanne UnrUh, Member, Special Education and Clinical Sciences; Philip Fisher, Outside Member, Psychology