Overgeneralization in Learning to Read

Open Access
Publisher Website
Google Scholar
Abstract
An acquisition model of decoding is proposed to explain the kinds of errors children make as they try to read. The model is based on data gathered from pronunciation errors of words presented in isolation. An analysis of the errors reveals that the kinds of errors made vary with reading competency. Less able readers make errors that appear to reflect reliance on a one-to-one correspondence between single letters and their sounds. More able readers make errors that suggest an understanding of letter-cluster-to-sound rules. The less able readers rely on less effective rules than do the more able readers. Differential understanding of the structure of letter-to-sound correspondences, particularly of vowel to sound patterns, is proposed to be the primary source of disparity in reading performance among beginning readers.

This publication has 2 references indexed in Scilit: