Exclusion of subjects with F scores at or above 16 in MMPI research on alcoholism

The results of these comparisons that used the MacAndrew and Holmes scales suggest that the practice of excluding Ss with F scores at or above 16 may result in the discarding of data on alcoholic and problem-drinking Ss with no appreciable gain in the statistical differentiation of the remaining Ss. In the case of the Hampton scale, application of the rule results in nonsignificant statistical results. Furthermore, the F scale differentiated the groups more effectively than did the Hampton scale. It is suggested that MMPI investigators who use the MacAndrew and other alcoholism scales reexamine the appropriateness of MacAndrew's exclusion rule for their particular samples.