Predicting personality in adulthood from college MMPI scores: implications for follow-up studies in psychosomatic medicine.

To assess the long-term predictive utility of Minnesota Multiphasic Personality Inventory (MMPI) content scales, 1,960 individuals who had completed the MMPI in college in 1964 or 1965 were administered two measures of adult personality, the NEO Personality Inventory (NEO-PI) and the Cook and Medley MMPI Hostility scale, in 1988. A comparison group of 274 men and women in the Baltimore Longitudinal Study of Aging were given both MMPI and NEO-PI between 1981 and 1987. Predictive correlations between MMPI scales and NEO-PI factors were qualitatively similar to concurrent correlations, but approximately half as large in magnitude. Theoretically, these correlations were interpreted to mean that about half the variance in basic dimensions of personality is stable from college age into middle adulthood. Practically, the relatively modest correlations suggest that predictive studies of medical outcomes probably require large samples, and that baseline data from adults (e.g., over age 30) may be more useful for future studies. The combination of stability and change suggests that the decade of the 20s may be a particularly fruitful time to conduct research on interventions to alter personality and their effects on health outcomes.