"Does Congruence Between Student and Observer Ratings of Classroom Interactions Matter for Student Behavior and Relationships?"
AERA 2018 Annual Meeting: “The Dreams, Possibilities, and Necessity of Public Education”
April 13-17, 2018
Title: Does Congruence Between Student and Observer Ratings of Classroom Interactions Matter for Student Behavior and Relationships?
Authors: Catherine Corbin, Erik Ruzek (PhD Alumnus), Jason Downer
Teacher-student interactions within the classroom have been shown to predict social-emotional and academic achievement (Allen et al., 2013; Downer, Sabol, & Hamre, 2010). Most of these findings come by way of observational measures, many of which describe the average experience of students within a classroom (Pianta, Hamre, & Mintz, 2012). Though such information is useful when considering the general climate of a classroom, it is possible that such measures are masking heterogeneity that may exist among students’ individual experiences within that classroom. Investigating how this heterogeneity contributes to students’ classroom relationships and behavior holds promise for providing a more nuanced understanding of the environments in which learning takes place.
Thus, the present study utilizes data from a racially/ethnically diverse sample of 2,047 third and fourth grade students in 145 classrooms in a large, northeastern city to determine whether congruence between student perceptions and observer ratings of classroom interactions predicts students’ classroom relationships and behavior (concurrently and over time).
Data for this study come from a large-scale evaluation of 4Rs+MTP, a social-emotional intervention paired with a teacher coaching model. All data referenced presently were collected at the beginning and end of the academic year. Quality of classroom interactions was measured via student report with the Learning About Teacher-Student Interactions survey (LATSI; Downer, Stuhlman, Schweig, Martinez, & Ruzek, 2014) and live classroom observations using the Classroom Assessment Scoring System (CLASS; Pianta et al., 2012); both measures tap into interactions that provide emotional support, classroom organization, and instructional support. Teachers reported on their closeness and conflict with individual students using the Student-Teacher Relationship Scale (STRS; Pianta, 2001) and on students’ aggressive behaviors and social competence using the Behavioral Assessment System for Children (BASC; Reynolds & Kamphaus, 1998).
Mean scores of all LATSI items and CLASS dimensions, respectively, were created and standardized. A congruence score was created by subtracting the standardized LATSI mean score from the standardized CLASS mean score, and taking the absolute value of that difference. Models were run in Mplus version 8 using type=complex to account for students nested within classrooms.
Preliminary results in the beginning of the year indicated that absolute congruence between student perceptions and observer ratings of classroom interactions significantly positively predicted conflict with students (β = .07, p < .05) and students’ aggressive behavior (β = .04, p < .05), with results trending in the opposite direction for social competence (β = -.05 p < .09). These results suggest that students who perceive classroom interactions differently compared to an observer are perceived by teachers to exhibit more aggressive and less socially competent behavior, and have a more conflictual relationship.
Prior to the conference, these models will be re-investigated using a multilevel framework, adding a test of directional congruence (student perceptions higher or lower than observations), and examining congruence in relation to changes in student outcomes. Final results will be discussed in terms of implications for approaches to individualized instruction and culturally responsive teaching practices, in addition to expanding the way that teacher effectiveness is conceptualized and measured.