"Whenever we measure anything, whether in the physical, the biological, or the social sciences, that measurement contains a certain amount of chance error....Two sets of measurements of the same features of the same individuals will never exactly duplicate each other....However, at the same time, repeated measurements of a series of objects or individuals will ordinarily show some consistency."

So wrote Robert L. Thorndike in Lindquist's "Educational Measurement" (1951). Traditionally, research into measurement error has dealt with whether or not the test items fit. But over the last 15 or so years, we have seen mounting interest in whether or not the people who answer the items fit. Most of the interest has centered on people whose responses do not fit the typical pattern.