Maximal measurement or meaningful measurement: the interpretive challenges of the MMPI-2 Restructured Clinical (RC) Scales.

The persisting correlations of the Restructured Clinical (RC; Tellegen et al., 2003) Scales with independent measures of first-factor variance (A, NEGE) suggest that trying to extract demoralization did not work well, and it may not be a clinically realistic task. Nichols's (2006/this issue) description of the clinical scales as "syndromal" and "multivariate" is seen as accurately reflecting their multicore natures so that the emphasis on the extraction of a single core unduly distorts the complexity and richness of the scales. Meehl and Hathaway (1946) cited factorially derived scales as "maximal" measurement, specifically measuring something very well; empirical scales based on criterion groups provide "meaningful" measurement as to the similarities of an individual's behavior to that of the group. I discuss this narrowness of factorial scales, such as the RC Scales, as a limitation of their generalization. I discuss problems with interpreting scales on the basis of the face validity of their items; assuming that self-perception and self report are accurate is a potentially misleading and undependable basis for assessment and intervention. I contrast the narrowness of the interpretive specificity of the RC Scales with the much broader and surprisingly specific interpretations offered by study of the individuals aggregated by the MMPI (Hathaway & McKinley, 1943) and MMPI-2 (Butcher, Dahlstrom, Graham, Tellegen, & Kaemmer, 1989) code types.