EDUCATIONAL AND PSYCHOLOGICAL MEASUREMENT LUDLOW AND O’LEARY SCORING OMITTED AND NOT-REACHED ITEMS: PRACTICAL DATA ANALYSIS IMPLICATIONS

Missing data occur for a variety of reasons. Particularly problematic are those items an individual skipped or could not reach in time. This article focuses on the practical effects of using different statistical treatments with omitted and not-reached items in an item response theory application. The particular strategy that is selected for scoring such items has a considerable impact on the interpretation of results. This impact is evident for either individual results or group-level assessments. Missing data are an inherent feature of most assessments at the state, na-tional, or international level. In a general sense, missing data refer to any blank responses to the total set of items available to a test taker. These may oc-cur for a variety of reasons including the fact that in most such assessments, no single test taker attempts all possible items. This deliberate design situa-tion may be handled through balanced incomplete block (BIB) spiraling and plausible values technology (Mislevy, Beaton, Kaplan, & Sheehan, 1992) as championed by the Educational Testing Service, among others. In contrast, however, it is also a fact that test takers create missing data by skipping some items and not having sufficient time to complete others. This topic, of course, is not new. Particularly relevant discussions and ref