Recently there has been keen interest shown in the applications of "Cloze Procedure" proposed by Taylor (1953) to the field of teaching of English as a second or foreign language. Though originally conceived to be an alternative measure of readability, the procedure has also come to be highly regarded as a reliable measure of reading comprehension and as a valid device for testing overall language proficiency.
In order to determine whether or not university students who differ in English proficiency have different strategies for English reading comprehension, we conducted the following experiment. Two groups of second-year students at Hiroshima University formed Groups A and B of twenty-six and thirty-one students, respectively. Both Group A and B were given CELT V-A Form (vocabulary test), S-A Form (stucture test), Davis's Diagnostic test, and three doze tests. The three doze tests, each of which included forty items, were constructed by deleting every sixth word from different passages extracted from three Oxford Progressive Readers (Grade 1, 2, and 3).
The results of the two proficiency tests and the Diagnostic test showed that Group A scored significantly higher than Grout B. This was true for the three doze tests, showing a significant difference in reading comprehension ability between the two groups. The correlations between the V-A Form, S-A Form and Diagnostic tests in Group A were fairly high, ranging from 0.64 to 0.72 while in Group B they were generally low, ranging from 0.40 to 0.59.
A total of one hundred and twenty items included in the doze tests were classified into cohesive and non-cohesive according to Halliday and Hasan (1976), in order to compare the scores within each group as well as between the groups. Both Group A and B were considerably more successful in guessing the cohesive items than the noncohesive ones. But the difference in the scores on cohesive items between the groups proved to be not significant.
On the basis of the results we conclude that university students who differ in English proficiency and reading comprehension ability use essentially the same strategies in the comprehension process.