Abstract
The authors investigated whether scores on the Cloze procedure as a test of plagiarism would be significantly affected by the readability of text. Undergraduates were asked to either paraphrase or plagiarize legal documents. Approximately half of the participants in the paraphrase condition received documents that were difficult to read, whereas the other half received versions that were easy to read. About 2 weeks later, participants completed Cloze tests that were based on their rewritten paraphrased or plagiarized versions of the documents. As predicted, Cloze scores from the plagiarize condition were significantly lower than those from the paraphrase condition. However, difficult-to-read documents, whether they had been paraphrased or plagiarized, yielded significantly lower Cloze scores than their easier-to-read counterparts. In spite of some methodological shortcomings, the results of this study suggest caution in the use of the Cloze procedure as a test of plagiarism.