A PSYCHOLINGUISTIC ANALYSIS OF THE CLOZE RESPONSES OF SELECTED READERS OF VARYING ABILITIES DURING SILENT AND ORAL READING

This study analyzed the patterns of "non-prespecified" responses on a written and oral cloze test exhibited by selected above average, average, and below average ability readers in order to develop a profile of how readers of varying abilities utilized particular reading strategies. The study also investigated how particular strategies employed in the silent reading process related to particular reading strategies in the oral reading process for readers of varying abilities. The subjects were thirty-four sixth grade students and thirty-three eighth grade students. Scores from the Reading Comprehension Test of the Iowa Tests of Basic Skills were used to obtain the reading ability groups necessary to this study. The instruments used were every fifth word deleted cloze passages constructed from text materials not familiar to the students having a readability level, assessed by Dale-Chall and Fry Readability formulas, equal to the grade level tested. The subjects were administered a group written and an individually administered oral cloze test. Responses that were not identical to the deleted word in the cloze passages were analyzed using an adapted form of the Goodman Taxonomy of Reading Miscues and classified according to categories of correction, syntactic and semantic acceptability, and grammatical function. A comprehending score was also calculated. To determine whether or not there were statistical differences between readers of varying abilities, analysis of the variance (ANOVA) was employed to identify those sources of variation critical at the F = .05 level. For those sources--ability, grade, and mode of testing identified as significant by the ANOVA--the Duncan Multiple Range Test was used to distinguish those means exhibiting significant differences. The findings revealed that: (1) Above average, average, and below average readers exhibit significantly different reading strategies that reflect their varying abilities on a written cloze test. (2) Above average, average, and below average ability readers exhibit significantly different reading strategies that reflect their varying abilities on an oral cloze test. (3) Although the process of reading is similar for all readers, there is a significant difference in the way readers of varying abilities use the information available to them. The degree to which a reader used the information on a written and on an oral cloze test distinguished the particular reading strategies of the above average, average, and below average ability readers from one another. (4) The consistent use of the language cues by readers of varying abilities suggests that oral reading and silent reading appear to be similar processes. (5) From those taxonomic characteristics highlighted by the written cloze test and those highlighted by the oral cloze test it appears that the written would be a valuable screening device, while the oral would be a beneficial diagnostic test when analyzing the "non-prespecified" responses of readers of varying abilities. The following recommendations were made: (1) Research should be performed at different grade levels in order to further investigate the strategies of readers of varying abilities at different instructional levels. (2) Investigation of post-performance measures of comprehension would provide further insight into reading behaviors, on a written and oral cloze test. (3) Research should focus on the role that the correction strategy plays in the silent reading process. (4) Future research should investigate the validity and reliability of the cloze test as a diagnostic test of silent reading ability using some form of miscue analysis. (5) Research should be conducted with learning disabled students to assess their use of the syntactic and semantic cues on the printed page. This information would be an important addition to the Individualized Educational Program developed for each Special Education student.