The effects of substitution miscues on the reading performance of selected grade four boys

Abstract

The purpose of this study was to determine to what extent substitution miscues affected the comprehension scores of grade four boys as they orally read a selected passage. The theory presented suggested that such mistakes made during oral reading would generaliy detract from comprehension, but that not all substitutions would detract equally. Further, it was suggested that if students reread to correct a substitution miscue, the original mistake would not detract from comprehension; rather the correction would add to the reader's comprehension and nothing would have been lost. But if a student did not go back to correct his error, there still may not have been a loss of comprehension, because the substitution may have contained as much meaning as the correct word. That is, if the uncorrected substitution was an acceptable one, both syntactically and semantically, it may have added to, rather than detracted from, the comprehension of the passage. Finally, it was suggested that only those substitutions which were syntac~ically-semantically unacceptable; that is, those that were grammatically incorrect and that were void of meaning, detracted from comprehension, and, hence, resulted in low comprehension scores. The sample for testing these hypothesized relationships consisted of 46 grade four boys who individually read the same passage for the investigator. Following the oral reading, each child "retold" the story in his own words and, based upon this recall and retelling, a comprehension score was established. Each session was tape-recorded; later on, substitution miscues were coded into one of three categories: (1) corrections, (2) syntactically-semantically acceptable miscues, or (3) syntactically-semantically non-acceptable m•i scues. Bivariate relationships were established between the three predictor variables and the outcome variable, reading comprehension. All associations were significant in the hypothesized direction. Regression analysis was then conducted on two predictor variables--proportion of corrections and proportion of syntactically-semantically acceptable miscues--to establish their effects on reading comprehension. It was found that corrections and acceptable miscues each independently affected reading performance and that the combined effects accounted for 38% of the variability in that reading performance. It was, therefore, concluded that not all oral reading errors detracted from comprehension; rather, that corrected errors and acceptable miscues added to the understanding of the passage, and. only unacceptable miscues detracted from understanding. The support for the hypotheses of the study and the theory from which they were derived served to enhance and reconfirm the theory underlying the Goodman-Goodma·n-Burke research.