The Classification Accuracy of Measurement Decision Theory. Paper presented at the annual meeting of the National Council on Measurement in Education, Chicago. Retrieved December 25, 2003 from: http://edres.org/mdt/papers/ncme2003c.pdf Rudner, L

Item responses on an operational state assessment were calibrated and analyzed using decision theory with the size of the calibration sample being the primary manipulated variable and score-group classification accuracy being the primary assessment goal. Simple decision theory was shown to be highly accurate in terms of placing individuals into the appropriate score categories. If the intent of an assessment is to classify individuals into discrete categories or determine the proportions of examinees within each score category, then decision theory provides an attractive alternative to classical or modern measurement theory. Classical measurement theory and item response theory are concerned primarily with rank ordering examinees across an ability continuum. But one is often interested in just classifying examinees into one of a finite number of discrete categories, such as pass / fail or below-basic / basic / proficient / excellent. Sometimes, one is interested in an eve