---

Cloze tests have been widely used for measuring reading comprehension, readability and language proficiency. There is still much controversy on what it really is that cloze measures. The result of much correlational research is contradictory and very unsatisfactory. Thus, with a qualitative orientation, this study attempts to look at the judgmental validity of cloze as a test of reading comprehension. To this end, a group of 32 native and non-native speakers of English sat a standard cloze test. The participants were expected to complete most of the blanks correctly if cloze measured reading comprehension properly, because the text had been intended for undergraduates while cloze-takers were all either PhD students or members of academic staff with a PhD. Surprisingly, the results indicated that none of the participants reached the minimum native speaker performance criterion of 70%. Invited to reflect on what they thought they were doing when reading the blanked text, most cloze-takers felt that the text they read was a puzzle or a guessing game. Provided with the deleted words and asked to re-read the text, they confessed that cloze reading was very different from the second reading. Further findings and implications for future research are discussed in the paper.