Leveled Reading Assessment: Documenting Reliability and Text Complexity Using Item Response Theory

• In Event: Roundtable Session 57
  In Roundtable Session: Investigating Characteristics of Assessment Measures and Data: Applications in K–12 Settings

Abstract

Leveled reading assessments are commonly used by teachers but often have limited empirical evidence documenting reliability, validity, and utility for important instructional decisions and a lack of transparency about the book leveling process. The current paper describes the development process and psychometric methodology employed to demonstrate evidence for both the reliability of the assessment and the text leveling process implemented during assessment development. Empirical results support the idea that text levels demonstrate monotonically increasing difficulty, a central characteristic and assumption of leveled book sets and leveled reading assessments.

Authors

• Matthew M. Gushta, Amplify Education, Inc.

• Danielle M. Parisi, University of Nebraska - Lincoln