Accuracy in identifying students’ miscues during oral reading: a taxonomy of scorers’ mismarkings

Abstract

Informal reading inventories (IRI) and curriculum-based measures of reading (CBM-R) have continued importance in instructional planning, but raters have exhibited difficulty in accurately identifying students’ miscues. To identify and tabulate scorers’ mismarkings, this study employed examiners and raters who scored 15,051 words from 108 passage readings by students in Grades 5 and 6. Word-by-word scoring from these individuals was compared with a consensus score obtained from the first author and two graduate students after repeated replaying of the audio from the passage readings. Microanalysis conducted on all discrepancies identified a cumulative total of 929 mismarkings (range = 1–37 per passage) that we categorized in 37 unique types. Examiners scoring live made significantly more mismarkings than raters scoring audio recordings, t(214) = 4.35, p =.0001, with an effect size of d = 0.59. In 98% of the passages, scorers disagreed on the number of words read correctly—the score used for screening and progress monitoring decisions. Results suggest that IRIs and CBM-Rs may not be accurate as diagnostic tools for determining students’ particular word-level difficulties.