Presents study designed to examine trait structure of a cloze test using confirmatory factor analysis. Results suggest that a modified cloze passage, using rational deletions, is capable of measuring syntactic- and discourse-level relationships in a text, and this advantage may outweigh considerations of reduced redundancy which underlie random…

Describes a study that developed criteria for rationally selecting the words to delete in developing a cloze test. The study also tried to determine the extent to which performance on a cloze test with rational deletions differs from that on a cloze test developed by the fixed-ratio deletion procedure. (SED)

A theoretical framework that distinguishes the knowledge, or competence, aspect of language proficiency from the skills aspect is outlined, and the factors in the language testing situation that affect performance on language tests are examined. The model is intended for use in assessing the construct validity of tests of language proficiency and…