EDUCATIONAL AND PSYCHOLOGICAL MEASUREMENT 1968, 28, 297-315. How do subjects go about answering an item on a personality questionnaire? Are characteristics of these responding processes related to the psychometric adequacy of the item? Interest in these questions stems from considering the relatively unsatisfactory state of personality measurement by questionnaires. Even in the better instruments, a large proportion of the total variance is remainder variance, variance not associated with differences between the scores of persons or differences between item means (Fiske, 1963(Fiske, , 1966. While part of this remainder variance can be considered an artifact associated with dichotomous or other restricted scoring, much of it must be attributed to person-item interaction: the person's response to the item differs from that predicted from his mean and the item's mean. What accounts for such idiosyncrasy of response?Only a few studies have investigated this interaction and the internal process in the subject as he selects his response. Benton (1935) demonstrated that the same item is given a variety of interpretations by subjects. But Eisenberg (1941) found that subjects may give different responses with the same reason and the 1 We are indebted to Thomas Tyler for originating the approach on which this study was based, to Barbara Turner for serving as a rater and helping in many other ways, and to Dr. Vivian Lipman and Dr. Eli Lipman for providing the subjects.