Validity Evidence for the Behavioral and Emotional Rating Scale—3: Factor Analysis and Measurement Invariance Across Race and Ethnicity

• Published in: School mental health 2024-12, Vol.16 (4), p.1082-1093
• Main Authors: Lambert, Matthew C., Gonzalez, Jorge E.
• Format: Article
• Language: English
• Subjects: Agreements; Association (Psychology); Behavior disorders; Behavior Rating Scales; Behavioral Science and Psychology; Check Lists; Child Advocacy; Child and School Psychology; Child Behavior; Child Development; Child Health; Children & youth; Childrens Rights; Clinical Psychology; Competence; Developmental Psychology; Discriminant analysis; Education; Educational Assessment; Educational Research; Emotional Problems; Evidence Based Practice; Factor Analysis; Factor Structure; Families & family life; Family Involvement; Government School Relationship; Individualized education programs; Individualized Instruction; Individualized Programs; Interviews; Mental disorders; Mental health; Meta Analysis; Minority Groups; Psychology; Quantitative psychology; Ratings & rankings; Special education; Students; Validity
• ISSN: 1866-2625; 1866-2633
• DOI: 10.1007/s12310-024-09715-0

Over the past several decades, the education, mental health, and social service fields have witnessed a significant shift in the focus of assessment of children moving from a solely deficit-based approach to a model incorporating strengths and competencies, which has been referred to as strength-based assessment. The Behavioral and Emotional Rating Scale (BERS) is one of the first and most widely used strength-based assessments of behavioral and emotional functioning. While the BERS is widely used and researched, there is limited evidence supporting validity of test scores based on the internal structure or measurement invariance. The present study examined validity evidence based on internal structure and measurement invariance across White, Black, and Hispanic students for the BERS-3 Teacher Rating Scale using a national sample of 1,965 school-aged students. Results indicated empirical support for the hypothesized correlated factors structure as well as a bifactor structure with a strong general factor. Measurement invariance was also established for the correlated factors and bifactor CFA models, indicating that the scores consistently measured emotional and behavioral strength constructs across the three student groups, scores were comparable between groups, and scores were comparable to the same set of normative standards. Because of the test’s emphasis on strengths and invariant measurement across groups, implications for students of color are discussed.