Humphrey D H, Dahlstrom W G
Dorothea Dix Hospital, Raleigh, North Carolina, USA.
J Pers Assess. 1995 Jun;64(3):428-39. doi: 10.1207/s15327752jpa6403_3.
Fifty-three individuals under court review at a forensic unit of a state hospital were administered both the original Minnesota Multiphasic Personality Inventory (MMPI; Hathaway & McKinley, 1943) and the MMPI-2 (Butcher, Dahlstrom, Graham, Tellegen, & Kaemmer, 1989) within an interval of a few days. The test-retest stability of the raw scores from each administration was determined by computing Pearson product-moment correlations for both the individual scales in the profile and for the pattern of scores on the two instruments for each subject. The stability of the T-score patterns was analyzed by means of total codes of the pairs of profiles, tabulations of the two-point high-point combinations, and correlations of the T-score profiles of each subject on the two instruments. The raw scores from the two administrations were highly stable on retest. The patterns of the raw scores for each subject were also very stable. However, when the raw scores were transformed into T-scores on their respective norms, the patterning was often drastically different, indicating that the bases for clinical interpretation derived from the MMPI and the MMPI-2 profiles were sufficiently at variance to require different conclusions. Until the correlate base of the MMPI-2 is better established, it is recommended that two separate profiles be drawn, one from the original norms and the other from the restandardized norms, and that each be interpreted separately to determine their differences and similarities.
