Stability of personality self-ratings over 30 years: evidence for an age/cohort interaction

An investigation was made of the hypothesis of Bloom (1964) and others that the stability of personality is greater among older than younger adults. The subjects were 459 men in two cohorts, followed from 1947 to 1977 as part of a longitudinal medical study. Self-ratings were obtained over the 30-year period on 15 item-factor scales from the Minnesota Multiphasic Personality Inventory and on two higher order factors. There was no selective attrition in the sample with respect to baseline personality scores. Thirty-year uncorrected stability coefficients ranged from .08 to .88 across the two cohorts. There was substantial evidence of greater stability on many traits in the older than in the younger age/cohort group. Furthermore, in late adulthood, traits related to the higher order factor Constraint were more stable than were traits related to Positive vs. Negative Affectivity. The study is not a definitive test of Bloom's hypothesis because cohort and age effects are confounded. However, the results are supportive of the hypothesis and the study improves on existing research in a number of ways.