Access Type

Open Access Dissertation

Date of Award

January 2018

Degree Type

Dissertation

Degree Name

Ph.D.

Department

Education Evaluation and Research

First Advisor

Shlomo Sawilowsky

Abstract

The problem that exists when using one or more readability indexes to ascertain a text grade level is the varied outcomes received on any given text from readability indexes that purport to measure the same construct. This study aims to provide practitioners with data to make informed decisions regarding interchangeability of readability indexes. A total of n = 244 narrative (n = 116) and expository texts (n = 128) passages from grades 1-5 were evaluated using the following readability indexes: Flesch-Kincaid Grade Level, Fry Graph, Spache, Dale-Chall, Gunning Fog, and Smog. Fifteen (15) comparison sets were analyzed using Bland-Altman method to assess for agreement. An a priori set standard of 1.5 grade levels was used as an acceptable difference. Other considerations for agreement included narrow limits of agreement, low proportional error, and a Bland-Altman plot where data points clustered around the bias line. Of the fifteen (15) comparison sets, nine (9) resulted in agreement, or near agreement. Based on the findings of the study and the subjectivity of the Bland-Altman method, it is recommended that practitioners select one readability index for text evaluation and use it exclusively. No particular index was recommended for use. The use of readability indexes should be one of several means of evaluating a text.

Share

COinS