Are Standardized Tests Sensitive to ESL Students' Growth?

Abstract

This study reports an experiment conducted on 110 ESL university students comparing a standardized traditional reading test with a multiple-choice cloze test and a short-context test with a criterion of teacher ratings as measures of growth in reading ability. Results indi cate no evidence that the cloze test or the short-context test performed in a superior manner to the traditional test. Questions are raised as to the nature of growth in reading skills and also as to the validity of the cloze procedure as a measure of across-time gains in reading and of reading proficiency.