Construct drift (Nichols, 2006) describes the possibility that the MMPI‐2 Reconstructed Clinical (RC) scales are less accurate measures of the primary traits than the original Clinical scales. The authors review the development of RC scales RC4, RC7, and RC9 and provide further evidence that these scales measure traits and behaviors that are not necessarily equivalent to those assessed by the original scales. Further, the development of the original Clinical Scale 7 is reviewed and evidence is provided that suggests that construct drift is not a new phenomenon, but may have been present in the development of the original scales. Suggestions for future research as well as cautions with regard to the clinical interpretation of the RC scales are provided. © 2011 Wiley Periodicals, Inc. J Clin Psychol 67:1–11, 2011.