2-037 - Improving Precision and Utility of the Early Childhood Environment Rating Scales

Session Type: Paper Symposium

Integrative Statement

Initiatives designed to improve experiences of young children in classrooms require psychometrically sound measures of quality that are linked to children’s gains. The Early Childhood Environment Rating Scale-Revised (ECERS-R; Harms, Clifford, & Cryer, 2005)—the predecessor to the ECERS-3 (2015)—is the most widely used observational assessment of the early childhood environment. ECERS-R scores are often correlated with changes in children’s early academic and social skills, but associations are generally small (Burchinal, Kainz, & Cai, 2011), possibly due to psychometric limitations.

This symposium will present three papers, each focused on identifying and addressing these potential limitations. These papers explore alternate ways of analyzing and scoring ECERS to increase precision and strengthen links to children’s gains.

The first paper investigates the role of geographic heterogeneity and family/child demographics as moderators of associations between ECERS-R and children’s receptive vocabulary, controlling for a prior measure of development. The second paper combines eight large data sets to explore disordering within the ECERS-R. The authors conclude that ECERS-R cutoffs used in many high-stakes systems are problematic for accurately estimating quality. The third paper presents a novel indicator-based scoring system -- first using the ECERS-R and then using the ECERS-3—as a means of creating subscales that measure specific aspects of classroom quality, such as math/science, socio-emotional development, and teaching.

An Austin-based professor of public policy will serve as discussant. She will draw on her extensive engagement with state and local decision-makers in discussing the importance of rigorous and feasible measures of early childhood classroom quality.

Sub Unit

• School Readiness/Childcare

Chair

• Diane Marie Early, University of North Carolina at Chapel Hill

Discussant

• Cynthia Osborne, University of Texas at Austin

Individual Presentations

• ECERS-R Quality and Children’s Vocabulary: Examining Variation by Geography and Demographics Using Large Scale Meta-Synthesis - Presenting Author: Rachel Gordon, UIC; Kerry G. Hofer, Abt Associates; Fang Peng, University of Illinois - Chicago; Diksha Gaur, University of Illinois - Chicago; Danny Lambouths, III, University of Illinois - Chicago

• Measuring Preschool Quality with the ECERS-R: Replicating and Extending Item Response Theory Results across Eight Datasets - Presenting Author: Ken A. Fujimoto, Loyola University Chicago; Rachel Gordon, UIC; Fang Peng, University of Illinois - Chicago; Kerry G. Hofer, Abt Associates

• Alternative Scoring of the ECERS-R and ECERS-3 to Improve Precision and Prediction - Presenting Author: John Sideris, University of North Carolina at Chapel Hill; Diane Marie Early, University of North Carolina at Chapel Hill; Jennifer Neitzel, UNC-Chapel Hill