Exploring Teachers’ Ratings of Students’ Math Ability, Students’ Perceptions of Teachers’ Practices, and Student Motivation

• In Event: Exploring the Situated Nature of Teachers’ Ability-Related Beliefs, Classroom Practices, and Well-Being

Abstract

Aims/Background
Building on work by Miele et al. (2019), the original aim of this study was to examine the moderating effects of elementary school teachers’ growth mindsets about intelligence on the relations between teachers’ ratings of individual students’ math ability, the practices they engage in when those students are struggling, and changes in students’ math motivation over the course of one school year. When examining the results, we were surprised to observe a positive association between students’ perceptions of teachers’ restrictive practices (e.g., being told to hurry up when struggling with a math problem) in the fall and their ratings of math value in the spring (controlling for baseline). This led us to further explore the cross-sectional and longitudinal relations between the constructs in our study, and to call into question the stability of these constructs over time. The purpose of this presentation is to relate the findings of this exploration to a central theme of the symposium: the importance of accounting for context-specificity when examining the associations between teacher beliefs, instructional practices, and student outcomes.

Method
Participants were 43 elementary school teachers from a U.S. school district and 570 of their students (see Table 1 for demographics). The measures examined in the present analyses (detailed in Table 2) were collected at three timepoints: August 2018 (T1; teachers’ mindsets only), November 2018 (T2; teachers’ ratings of each students’ math abilities, students’ perceptions of their teacher’s use of supportive/restrictive instructional practices, and students’ math expectancies/value), May-June 2019 (T3; same measures as T1 + T2).

Results/Discussion
To examine our hypotheses pertaining to the moderating effects of teachers’ growth mindsets, we estimated the multi-level path analyses depicted in Figure 1. The results of these analyses indicated that none of the effects involving teacher mindsets were significant, which may have been partly due to low stability in this variable over time. Indeed, the correlation between mindsets from T1-T3 was only .12.
One of the few longitudinal associations that was significant in these analyses was the aforementioned positive relation between students’ T2 perceptions of teachers’ restrictive practices and their T3 math value (controlling for T2 value). To better understand this surprising result, we examined versions of the Figure 1 model where the Level 1 variables were entirely cross-sectional (at either T2 or T3) and that included teacher mindsets as a covariate rather than moderator. As shown in Table 3, these models indicated that perceived restrictive practices were negatively associated with math values and math expectancies at each time point (as expected). A potential explanation for the existence of both a concurrent negative relation and a positive longitudinal relation is low variable stability over time and/or regression to the mean (Marsh & Hau, 2002). Indeed, the correlation between perceptions of teachers’ restrictive practices from T2-T3 was only .32. Additional findings of interest that will be discussed include concurrent (but not longitudinal) positive relations between students’ perceptions of teachers’ supportive practices and their math values/expectancies, and relations between teachers’ rating of students’ math ability and students’ values/expectancies.

Authors

• David B. Miele, Boston College

• Alexander S. Browman, College of the Holy Cross

• Sidney C. May, Dartmouth College

• Melissa McTernan, Boston College