MAC scale scores of three samples of men under conditions of conventional versus independent scale administration.

Abstract

The MacAndrew Alcoholism (MAC) Scale, a 49-item scale derived from the MMPI [Minnesota Multiphasic Personality Inventory], was administered both independently and in the context of the MMPI to 53 men alcoholics in a Veterans Administration hospital, to 49 men charged with drunken driving and to 79 male college students. A 1-2 wk interval elapsed between the 2 testings of each sample, and the college students were retested under both conditions approximately 11 wk later. Comparison of the scores obtained under the 2 testing conditions indicated that the groups'' mean scores were stable. When the 6 possible correlations between the students'' 4 arrays of scores were examined, none of the differences between any of the 15 possible pairings of the 6 correlations were significant, indicating that the subjects'' relative score positions were not affected by the passage of time between testings. Applications of the MAC scale''s standardization group-derived decision rule to the scores of the 53 hospitalized alcoholics resulted in virtually identical accuracy rates, 89% being identified as alcoholics when the complete MMPI was administered and 87% when the MAC scale was administered separately.