Context in Reading and Listening: Variations in Approach to Cloze Tasks

130 ENGLISH SUBJECTS beginning their fifth year of school were tested on cloze tests of reading and on the same tests presented as either cloze listening tests or as restricted cloze reading tests. In the latter total sentence context was not available. When the sample was dichotomized about the median of the normal reading test scores and when results for the 3 versions of the test were compared for the poorer readers, there were no significant differences; for the better readers, there was a significant difference (p < .01) between the normal cloze scores and either the listening scores or the restricted reading scores. An examination of incorrect responses showed that all subjects seemed to be aware of grammatical and semantic constraints, but the poorer readers tended to make mistakes which were suitable only to preceding context, on all tests. The better readers did so chiefly on the listening and restricted reading tests. It was concluded that these poorer readers were using strategies inappropriate to silent reading.