The importance of the teacher-child relationship for teachers’ perceptions of students’ executive functions.

• In Event: 1-053 - An integrative approach to understanding executive function development in young children.

Integrative Statement

Teachers’ reports of executive functions (EFs) are a widely-used, pragmatic alternative to directly assessing EFs using computer tasks. Further, since teachers have valuable information about children’s EFs in the classroom context, teacher-reported EFs have more ecological validity than direct assessments (Toplak et al., 2013). However, there is evidence of biases in teacher-reported EFs by students’ gender and ethnicity (Garcia et al., 2018), suggesting that teacher-reported EFs are influenced by factors other than the child's EF skills.

The quality of the teacher-child relationship impacts how teachers view students and has also been linked to children’s EF development. In particular, studies highlight that teacher-child conflict is associated with lower gains in directly-assessed EFs in elementary school students (Berry, 2012; de Wilde et al., 2016; McKinnon & Blair, 2018). The current study tests whether teacher-child relationship quality is more strongly associated with gains in teacher-reported or directly-assessed EFs. Second, we examine whether teacher-child relationship quality predicts teacher-reported EFs after accounting for directly-assessed EFs.

The sample includes 317 third-, fourth-, and fifth graders from 33 classrooms. EFs were assessed in the fall and spring of a single academic year. Directly-assessed EFs were measured with four standardized tablet-computer tasks (Hearts and Flowers, Flanker, Digit Span Backward, Multi-Source Interference Task) and combined to create a composite. Direct assessments of EFs were conducted simultaneously with all students in the classroom at the beginning of the school year. Teacher-reported EFs were captured by asking teachers to rank the EFs of the children in their classroom relative to one another. Teachers reported on teacher-child closeness and conflict using the Student Teacher Relationship Scale. Child gender, age, ethnicity, and parental education were included as covariates. All analyses include school fixed effects to eliminate any school-level differences in teacher or student characteristics and multiple imputation was used to address missing data.

After accounting for the clustering of students within classrooms, teacher-child conflict was associated with smaller EF gains on both directly-assessed and teacher-reported EFs. However, the negative association between conflict and EFs was significantly larger for teacher-reported EFs compared to directly-assessed EFs (χ2(1) = 11.64, p < .001). Controlling for directly-assessed EFs in the spring, teachers rated students as having worse EFs when they also reported having more conflict with the student. Teacher-child closeness was not a significant predictor in any of the models.

Teacher-child conflict may interfere with children’s opportunities to improve and practice their EFs as more instructional time is spent on behavioral management than learning activities (La Paro et al., 2004). The negative association between conflict and EF growth was larger for teacher-reported EFs than for directly-assessed EFs. Further, we found that teachers who reported more conflict also rated students as having lower EFs, controlling for directly-assessed EFs and a robust set of covariates. Together these findings suggest that teacher-child conflict may bias teachers’ reports of EFs. With increases in the use of teacher-reported EF measures for school accountability, it is important to understand that teachers’ relationships with their students affects their perceptions of student EFs.

Authors

• Jenna E. Finch, University of Nebraska-Lincoln
  Presenting Author

• Jelena Obradovic, Stanford University
  Non-Presenting Author

• Elisa B Garcia, SRI International
  Non-Presenting Author