Poster #200 - Variation and average teaching quality as a predictor of academic outcomes: extending CLASS

• In Event: 2-002 - Special Poster Session 05 with Continental Breakfast Reception
  In Poster Session: PS 05 - Policy Section

Integrative Statement

The Classroom Assessment Scoring System (CLASS) is an observational instrument used to assess the everyday interactions in educational settings.

The instrument has strong theoretical groundings; however, prior empirical validation of the CLASS has exposed some psychometric weaknesses and recent meta-analyses demonstrate limited ability to predict child outcomes.

CLASS consists of multiple observations of a classroom over time. Though the general approach is to take an average of these observations for analytical purposes, there may be important information within these multiple assessments.

In this project, we use observational data of 250 Early Childhood Education centres to test the hypothesis that variability in the quality of interaction between child and educators would be a greater predictor of child academic outcomes in grade 3 than the overall average CLASS score.

The analysis was performed using hierarchical random effects models to obtain room level trajectories of the CLASS scores over time. These trajectories were then were used to predict academic outcomes in a separate hierarchical random effect linear model.

Preliminary results suggest that variability in the quality of interactions between children and educators, particularly instructional support, is a stronger predictor of child academic outcomes than average CLASS scores. Results are discussed within the context of how educators respond to fluctuations in children’s learning across the day.

Authors

• Peter Rankin, University of Queensland
  Presenting Author

• Sally Staton, University of Queensland
  Non-Presenting Author

• Tony Beatton, University of Queensland
  Non-Presenting Author

• Karen Thorpe, University of Queensland
  Non-Presenting Author