The Modeling of Nonlinear Growth for Improving the Accuracy of Identification Decision Rules

• In Event: Fairmont Roundtable Session Seventeen
  In Roundtable Session: Multiple Linear Regression: The General Linear Model SIG Roundtable: Statistical Modeling and Applications to Educational Data (Table 10)

Abstract

Progress monitoring using curriculum based measures administered to a student at multiple points in time is common in educational and psychological settings. Results from these assessments help inform educators about individual student needs for special services. Recent research has demonstrated that common approaches for identifying individuals in need of special services are deleteriously impacted by a nonlinear relationship between time and test scores. The purpose of this study was to test a nonlinear regression model method to account for such nonlinearities, and thereby improve accuracy in identifying students for special services. Results of the statistical simulation demonstrated that use of this nonlinear model did indeed improve the accuracy of common methods for identifying students in need of special services.

Authors

• W. Holmes Finch, Ball State University
  Presenting Author

• Maria E. Hernandez Finch, Ball State University
  Non-Presenting Author