This study evaluated the Minnesota Multiphasic Personality Inventory (MMPI) as a predictor of employment status with male, physically handicapped vocational rehabilitants of employable age who had been patients at centers at the University of North Dakota and the University of Minnesota; and compared Katzell's Double Cross-Validation (DCV) with a conventional, unilateral procedure (SCV) used on the same sample. Principal results and conclusions were: (1) the MMPI contains items that significantly differentiate the employed and those in training from the unemployed; (2) two final scales were devised and cross-validated by the DCV procedure, but their predictive validity is uncertain; (3) two other scales obtained by conventional item analysis and cross-validation were not significant; (4) the final DCV scales are tentatively valid predictors of employment status for the male rehabilitants, but differences between the final scale and the two contributing subscales may make it inappropriate to cross-validate the former by cross-validating the latter; (5) differences between scales derived by the two cross-validation and by the conventional method indicated that the same items are not necessarily extracted by these differing procedures. (author/ly)