64.037 - Innovative Approaches in Estimating Seasonal Achievement: Implications for Methods, Policy, and Practice

Session Type: Symposium

Abstract

Educational data are often seasonal, with increases in achievement test scores between fall and spring accompanied by decreases during the summer when students are not in school. Yet, these seasonal patterns are often ignored when estimating parameters relevant to education policy and practice. In this symposium, we describe a statistical model that can be used to simultaneously estimate between- and within-year growth, helping account for seasonality. We then show how this model can be used to understand achievement gaps and school contributions to student growth in the presence of summer learning loss. Results suggest that our model fits seasonal data better than standard polynomial growth models, and that the policy and practice metrics we estimate appear sensitive to seasonality.

Sub Unit

• Division D - Measurement and Research Methodologies / Division D - Section 1: Educational Measurement, Psychometrics, and Assessment

Chairs

• Mark R. Wilson, University of California - Berkeley

• James Soland, NWEA

Papers

• Modeling Growth by Adding Curves: The Compound Polynomial for Seasonal Time Series - Yeow Thum, NWEA

• Seasonal Comparisons of the Black-White Achievement Gap - Megan Kuhfeld, NWEA

• Estimating School Contributions to Student Growth in the Presence of Seasonality - James Soland, NWEA