Assessing Teachers' Social-Emotional Practices: A Comparison of Measures that Capture Specific Social-Emotional Practices and General Teacher-Child Interactions

• In Event: Conducting Observations at Scale: Adapting a Measure of Teacher Practice to Support Children’s SEL Development

Abstract

Introduction
Evidence suggests that teachers are effective at improving the social and emotional readiness of low-income children. However, few measures are available to assess teachers' use of specific social-emotional practices within their classrooms. This presentation's overarching purpose is to examine the context of Head Start classrooms from both a general teacher-child interaction quality (CLASS) and teachers' use of social-emotional practices (adapted-TSRS) perspective. This study adds to the research by highlighting each measure's unique and complementary contributions and provides evidence to support the use of both tools in understanding the social-emotional environments of ECE classrooms.
Hypothesis
We posit that the general classroom climate domains in the CLASS will be related to the relevant social-emotional practices in the Adapted-TSRS - for example, Emotional Support (ES; CLASS) will be related to Social-Emotional Instruction (Adapted-TSRS), the Classroom Organization (CO; CLASS) will be most related to the Classroom Management (Adapted-TSRS) domain, and finally, the Instructional Support (IS; CLASS) domain will be related to the Scaffolding domain (Adapted-TSRS.) Finally, we hypothesize that concurrent associations are stronger between measures than cross-time associations between and across measures.
Study population
This sample from this study includes 17 Head Start grantees within 10 states across the United States. Approximately 307 Head Start lead teachers from 100 Head Start centers participated. The teachers were predominately female (96%, n = 295), had at least a bachelor’s degree (62%, n = 190), and were, on average, 43 years of age (SD = 11.84). Teachers were primarily African American (38%) or Hispanic (29%). One hundred and fifty-two classrooms served only 4-year-olds (referred to as "4-year-olds classrooms"), and 155 classrooms served both 3-year-olds and 4-year-olds (called "mixed-age classrooms" in this presentation).
Methods
The adapted-TSRS provides information on teachers' social-emotional practices using 14 items on a five-point scale, clustered in three domains (Social-Emotional Instruction, Classroom Management, Scaffolding). The CLASS provides global, seven-point Likert scores about all adults and children in the classroom across 10 items, clustered into three domains (ES, CO, IS). Each CLASS cycle consisted of 20 minutes of observations and 10 minutes of coding. Following the two CLASS cycles, observers coded a cycle of the Adapted-TSRS representing what they had observed during the two CLASS. This procedure was repeated (two CLASS observations followed by the Adapted-TSRS), resulting in 4 CLASS cycles and 2 Adapted-TSRS cycles.
Results
The Adapted-TSRS domains were correlated with the three domains of the CLASS (see Table 1) at baseline for both 4-year-old and mixed-age classrooms. Correlations varied depending on the type of classroom. Correlations across time were examined (see Table 2); the Adapted-TSRS Social-Emotional Instruction domain at baseline was significantly positively correlated with follow-up CLASS Emotional Support (r = 0.28, p < .10) and CLASS Instructional Support (r = 0.30, p < .05). However, baseline Adapted-TSRS Classroom Management and Scaffolding were not correlated with follow-up CLASS ES, IS, CO domains. Results suggest that the Adapted-TSRS, a behaviorally anchored, a checklist-based measure of teacher social-emotional practice, is associated with, but not duplicative of, the CLASS.

Authors

• Natalia Rojas, New York University School of Medicine
  Presenting Author

• Shira Kolnik Mattera, MDRC (Manpower Demonstration Research Corporation) - New York City
  Non-Presenting Author