Variation in Young Children’s Self-regulation Skills Across Early Childhood Classroom Contexts

• In Event: 1-144 - Executive Functions in Flux: Contextualising our Approaches for the Real World

Integrative Statement

Researchers and practitioners increasingly seek to support the development of young children’s executive function and self-regulatory skills. Understanding how these skills vary allows teachers and caregivers to anticipate when children may need extra supports. While existing measures capture adults’ holistic perceptions of children’s skills (e.g. BRIEF; Gioia, Isquith, Guy, & Kenworthy, 2000) or measure performance on specific, computer-based tasks (e.g. NIH Toolbox; Zelazo et al., 2013), little is known about how children’s skills vary across everyday situations, either randomly or systematically. This study uses results from an observational measure of regulation-related skills (e.g., attention, impulse control, planning) to explore within-individual variation across different situations in early childhood classrooms. We describe how much variation exists at the child-level, then examine if these skills vary by time of day or the social and instructional context of activities (e.g. paired vs. independent activities).

Research assistants collected videos of 38 consented children in six preschool/kindergarten classrooms in suburban Massachusetts (55% male, average 5.24 years-old, range= 4-6). Segments of 4-7 minutes were selected to show children completing a teacher- or student-directed task, followed by a transition to a new activity. Research assistants trained on the use of the RRSM coded 139 video segments (average 3.66 per child, range = 1 to 8). An additional round of data was collected this school-year, which will approximately double the sample size for the final analysis.

The Regulation-Related Skills Measure (RRSM) is an ecologically valid, observational tool which uses 16 items to assess a child’s use of regulation-related skills in early childhood classrooms over a 4-7-minute period. It shows adequate internal consistency and inter-rater reliability, as well as positive, moderate correlations with existing measures (e.g. BRIEF, NIH Toolbox; McCoy, Jones, Hemenway, Koepp, & Wilder-Smith, 2017).

We use hierarchical linear models (Raudenbush & Bryk, 2002) with observations nested in children to describe the amount of variation in regulation-related skills at the child-level, as well as the amount variation predicted by contextual factors at the observation-level. RRSM total scores are predicted by dichotomous variables indicating the timing of the activity (morning or afternoon), whether the activity is led by a teacher or child, and whether the activity is completed individually, in small groups, or as a whole class. Additional analyses will examine the content of activities and amount of instructional support the teacher provided by teachers.

A simple random-intercepts model indicates that about half of the variation in regulation-related skills is explained by characteristics at the child-level (intra-class correlation = .57), whereas the rest reflects within-child variability. Preliminary results show that children demonstrated stronger regulation-related skills during transitions (p<0.1). Although some descriptive differences emerged across social and instructional contexts, these were not statistically significant (see Table 1).

Preliminary results indicate that these skills are only moderately stable within children across different observations, with nearly half of the variation left unexplained by characteristics at the child-level. A larger sample will help to determine if differences in regulation-related skills across contexts are due to chance rather than specific contextual factors.

Authors

• Andrew Koepp, The University of Texas at Austin
  Presenting Author

• Dana McCoy, Harvard Graduate School of Education
  Non-Presenting Author

• Stephanie Jones, Harvard Graduate School of Education
  Non-Presenting Author

• Elena Bodrova, Tools of the Mind
  Non-Presenting Author

• Deborah Leong, Tools of the Mind
  Non-Presenting Author

• Abigail Hemenway, Harvard Graduate School of Education
  Non-Presenting Author