The Factor Structure of the MMPI–2 Restructured Clinical (RC) Scales

Abstract

We investigated the MMPI–2 Restructured Clinical (RC) scales (Tellegen et al., 2003) to determine if they had a more differentiated factor structure than the MMPI–2 Clinical scales. When factored alone, the RC scales had a 5-dimensional structure; the Clinical scales had 3 dimensions. When factored in combination with the Content scales, both sets of scales produced 5 dimensions. However, the RC and Content factors generally provided more efficient and logical markers of psychopathology than the Clinical and Content factors. We discuss interpretive considerations.

Editor's Note: Roger Greene served as the Editor with full decision authority for this submission.

Notes

¹Scale 5 and Scale 0 do not have corresponding RC scales and are not believed to assess core components of psychopathology (Tellegen et al., 2003).

²All RV pattern matrices are available on request.

³Varimax rotation produced similar factors and levels of correspondence across solutions.

⁴The RV loadings ≥ |.40| on a fourth dimension were as follows: three solutions had one, four solutions had two, two solutions had three, and one solution had four. Of these salient RV loadings, 14 were between |.40| and |.50|, 5 were between |.50| and |.60|, and 2 were ≥ |.60|. Across iterations, only one genuine scale packet had a salient loading (.42) on a fourth dimension, whereas the remaining packets did not have loadings > |.34|.

^a Sign of factor is reversed.

⁵The random variable loadings ≥ |.40| per matrix were as follows: one solution had two, five had three, three had four, and the remaining solution had five loadings. Of the random variables, 16 had loadings between |.40| and |.50|, 15 were between |.50| and |.60|, and 3 were ≥ |.60|. No genuine scale packet had salient loadings on a sixth dimension (highest loading = |.26|).

^a Sign of factor is reversed.

⁶Of the 10 solutions, 3 had a random variable loading ≥ |.40| on a sixth factor. Five of the remaining solutions did not have 6th dimensions that were distinguishable from random variables, as no genuine packets had loadings ≥ |.40|. However, two matrices had notable genuine loadings. One had a genuine loading of .43, although the next two highest loadings were just .31 and .28. We did not consider this dimension interpretable because the three highest loadings from the random variables were nearly identical in size: .39, .35, and –.29, respectively. The remaining matrix had genuine loadings of –.61, –.51, and –.46, and the largest random variable loadings were .20, .19, and .16. On this iteration, a Scale 4 dimension was differentiated from random “noise.” However, because this dimension was observed only once across 10 iterations, we do not think it provides sufficient evidence to warrant retaining six Clinical and Content factors.

^a Sign of factor is reversed.

^a Sign of factor is reversed.

^a Sign of factor is reversed.

^a Sign of factor is reversed.

^a Sign of factor is reversed.