Key points are not available for this paper at this time.
The effect of response format on diagnostic as-sessment of students ’ performance on an algebra test was investigated. Two sets of parallel, open-ended (OE) items and a set of multiple-choice (MC) items-which were stem-equivalent to one of the OE item sets-were compared using two diagnostic approaches: a dquo;bugdquo; analysis and a rule-space analysis. Items with identical format (parallel OE items) were more similar than items with different formats (OE vs. ntC). Index terms: bug analysis, diagnostic assessment, free-response, item format, multiple-choice, rule space. Response formats of assessment measures vary. There are two broad categories of response-constructed response and choice response-with various types of formats sub-sumed under each. In a constructed-response (also known as free-response) format, an ex-aminee is required to generate an answer to an open-ended (OE) item; in a choice-response for-mat, an examinee is required to select one or more answers from a short list of options. The most common item type in this category is the multiple-choice (lvtc) item. Numerous studies have compared the two for-mats with respect to different domains and from different perspectives. (For recent reviews of the literature see Bennett, 1991; Traub MacRury, 1990.) Response format comparisons have in-cluded theoretical considerations of the cognitive processing requirements of the two formats; empirical investigations concerning the psycho-metric properties of the two formats; examination of interaction effects of factors such as gender, race/ethnicity, test anxiety, test wiseness, and examinees ’ attitudes toward the formats; and examination of the effects of format expectancy on test preparation and test performance. Some of the psychometric properties on which the two response formats have been extensively contrasted include item difficulty and test relia-
Birenbaum et al. (Tue,) studied this question.