Validity and Reliability of the Cloze Procedure as a Measure of Readability and Comprehension for Prelingually, Profoundly Deaf Students

Abstract

The purpose of this study was to determine the validity and reliability of the cloze procedure as a measure of passage readability and the comprehension of prelingually, profoundly deaf students. Also under investigation was the use of the Fry (1968) and Dale-Chall (1948) readability formulas to determine if the cloze scores reflect relative passage difficulty for deaf students. Subjects for the study consisted of 18 prelingually, profoundly deaf students between the ages of 14-18 years and 27 hearing students between the ages of 9-10 years. The deaf subjects had attained scores of 4.0-7.6 on the Stanford Achievement Test, Intermediate I, Form B, Reading Comprehension subtest and the hearing subjects attained above average reading scores on the Iowa Test of Basic Skills. Cloze tests were constructed for each of the four reading passages consisting approximately of 250 words each. The Fry readability formula was used to compute the reading difficulty of the 3rd-grade passage. The Dale-Chall readability formula was used to compute the reading difficulty of the two 5th-grade and one 7th-grade passages. Both the deaf and hearing populations took all four passages. Statistical procedures used to analyze the data included the Pearson Product-Moment Coefficient of Correlation and the Kuder-Richardson Formula 20 for internal consistency. Within the limitations of this study, the following conclusions can be drawn: 1. The findings of this study suggest that, due to a lack of reliability, the cloze procedure is not a valid measure of readability and comprehension. 2. The findings failed to show that the cloze tests are measuring the same thing as the Stanford Achievement Test or the Iowa Test of Basic Skills. 3. The findings revealed that the deaf students found the two fifth-grade passages easiest, the seventh-grade passage next easiest, with the third-grade passage the most difficult.