Brief Book,Software, and Test Announcements

After reviewing literature detailing the ubiquity of error in research and assessment data, I describe mistakes found with MMPI-2 and Rorschach scores in an earlier publication (Meyer, 1993). The mistakes emerged from several sources, including hand-scoring the assessment measures, manually retrieving scores from patient files, and manually entering information for statistical analyses. The proportion of erroneous scores in the original study (MMPI-2 = 5.88%, Rorschach = 1.56%) was relatively small and reliability coefficients between the original and corrected scales were uniformly high (i.e., > .940). Consequently, the findings detailed in the original publication were not greatly affected by the mistakes that were made. Nonetheless, because this will not always be the case, I discuss the discrepancies as a way to sensitize researchers, clinicians, and students to the presence of error and then suggest several strategies for minimizing its impact on assessment data and research.