Subjects were 1,313 inmates at a Federal Correctional Institution who were admitted over a 2-year period. Two thirds of the sample was used as a derivation sample and one third was used for cross-validation. The criteria of adjustment were ratings made by custodial personnel, work supervisors, and teachers, as well as disciplinary infractions, days in disciplinary segregation, and days in sick call. The results provided support for the construct validity of several MMPI and CPI scales, but the absolute magnitude of the associations was low. Combinations of MMPI and CPI scales produced higher multiple correlations but generally did not hold up on cross-validation. It was concluded that neither the MMPI or CPI scales or weighted linear combinations thereof provide enough accuracy to be used alone for individual actuarial prediction. (Publisher abstract)