MMPI Expert Dave Nichols Reviews the MMPI-RF on Listserv

Dave Nichols gave me permission to copy his recent post to a Rorschach Listserv. This is a great response, typical of Dave.

Dr. Edelson writes: “My first course of action in this situation would be to administer the MMPI-2 and/or MMPI-RF.”

Although Dr. Edelson may well not intend it so, her statement could be misleadingly read to indicate a rough equivalence between the MMPI-2 and the MMPI-2-RF. This would be unfortunate. The MMPI-2-RF is only tenuously related to the MMPI-2, amounting largely to the two tests sharing 338 items and the 1989 Restandardization norms. The use of the familiar MMPI acronym in the MMPI-2-RF, while understandable as a means of commercial promotion given the established reputation of the MMPI/MMPI-2, risks (or intends?) distraction from the differences between the two forms. These are substantial, including the elimination of the 10 standard Clinical Scales from the RF, and their substitution by the new Restructured Clinical (RC) scales.

A close examination of the literature of the MMPI-2-RF and the RC scales will reveal a level of arrogance and sloppiness in their construction that some may find disconcerting.

Arrogance: Rather than taking the necessary pains to fully describe the development of each of the 28 new scales (out of 50) in the MMPI-2-RF for the benefit of customers, users and, especially, researchers, the authors of the new form state: “In the following we do not report the particulars of scale derivation in the same detail as we have provided for the RC scales (noting, as we did in the case of the RC scales, that ultimately what is most important is the results, the content, structure, correlates, and functions of each new scale). Instead we offer a narrative summary.” (Tellegen & Ben-Porath, 2008, MMPI-2-RF Technical Manual, p. 18)

Possible translations: 1) Trust us. 2) None of your business.

Sloppiness: One possible reason for avoiding the detailed description of the development of these 28 new RF scales is the sloppiness that was evident in these authors’ previous description of the development of the RC scales. Examples:

1) Their failure to provide in the RC Monograph (2003) a complete scoring key for the preliminary Demoralization scale (Dem), the version of the scale used in their factor analyses of the original Clinical Scales in order to identify a core construct for each (Step 2).

2) Their failure to confirm the results of these analyses after having dropped 5 items from Dem and added 6 new items to create the revised and final version of the Demoralization scale (RCd).

3) The contamination that resulted from appending Dem to the items of Scales 2 and 7, respectively, in their Step 2, after having previously recruited the Dem items exclusively from these same two scales in Step 1, thereby essentially ruling out any items overlapping Dem and either of Scales 2 or 7 as candidates for the latter scales’ core constructs. Unlike the Step 2 procedures followed to determine the core constructs of Scales 1, 3, 4, 6, 8, & 9, applying the same procedure to Scales 2 & 7 would have the effect of extracting the very same factor these scales had earlier been recruited to enlist!

4) Their failure to factor the final RC scales in any of their developmental samples to confirm that the core construct for each scale as embodied in the seed scales selected from each parent Clinical Scale survived as the dominant factor in its RC counterpart, or at least to report such analyses.

More extensive critical analysis of the RC scales and, by extension, the MMPI-2-RF, may be found in: Nichols (2006). The trials of separating bath water from baby: A review and critique of the MMPI–2 Restructured Clinical scales. Journal of Personality Assessment, 87, 121-138; Rouse, Greene, Butcher, Nichols, & Williams (2008). What do the MMPI–2 Restructured Clinical scales reliably measure? Answers from multiple research settings. Journal of Personality Assessment, 90, 435-442; Greene, Rouse, Butcher, Nichols, & Williams (2009). The MMPI–2 Restructured Clinical (RC) scales and redundancy: Response to Tellegen, Ben-Porath, and Sellbom. Journal of Personality Assessment, 91, 222-226; and in Ranson, Nichols, Rouse, & Harrington (2009). Changing or replacing an established psychological assessment standard: Issues, goals, and problems with special reference to recent developments in the MMPI-2. In J. N. Butcher (Ed.), Oxford Handbook of Personality Assessment (pp. 112-139). New York: Oxford University Press.

Dave Nichols