Screening for word reading and spelling problems in elementary school: An item response theory perspective

Abstract

The purpose of the present study was to explore whether the Item Response Theory (IRT) provides a suitable framework to screen for word reading and spelling problems during the elementary school period. The following issues were addressed from an IRT perspective: (a) the dimensionality of word reading and spelling ability, (b) the development of word reading and spelling ability throughout the elementary grades, and (c) measurement precision and differential item functioning. Two computer-based tests were administered to measure word reading and spelling ability. In the word reading test, a progressive demasking technique was used to present isolated words. In the spelling test, children used the computer keyboard to spell words. The children’s spelling ability was also tested using a paper and pencil test. Analyses showed unidimensionality for the children’s word reading and spelling abilities in grades 2 to 6. The effect sizes also showed the growth in word reading and spelling abilities to be substantial. Furthermore, the informative value of both the word reading and the spelling ability scale was satisfactory for all of the elementary grades and the results did not indicate that the children score systematically higher or lower on the computer-based spelling test than on the paper-based spelling test. The present results open up possibilities to map progress in the reading and spelling process systematically and accurately.