Investigating the Reliability and Validity of Text Leveling

• In Event: 35.083 - Poster Session 7
  In Poster Session: 35.083-1 - Reading Comprehension and Assessment

Abstract

We explored whether educators trained or not to use a particular text leveling rubric agree on the text level to assign to books and on the features of text related to those levels. We expected to find that, if book level is a meaningful entity, raters would agree on the level and that this agreement would be related to the expertise the raters had with the rubric. Results showed that within each group of raters there was consistency of level assignment and that greater expertise led to greater convergence. These findings support the construct of “book level,” and practice of leveling and although this approach is less objective than readability formulas, it can be conducted in a reliable manner.

Authors

• Emily M. Rodgers, The Ohio State University

• Jerome V. D'Agostino, The Ohio State University

• Katherine Brownfield, The Ohio State University - Columbus

• Sinead J Harmey, Queens College

• Robert H Kelly, The Ohio State University - Columbus