Resilient, overcontrolled, and undercontroleed personality prototypes in childhood: Replicability, predictive power, and the trait-type issue.

Abstract

In a longitudinal study, Q-sort patterns of German preschool children were analyzed for personality prototypes and related to developmental outcomes up to age 12. Q-factor analyses confirmed 3 prototypic patterns that showed a high continuity and cross-judge consistency; were similar to those found for North American, Dutch, and Icelandic children; and can be interpreted as resilient, overcontrolled, and undercontrolled. Relations reported by R. W. Robins, O. P. John, A. Caspi, T. E. Moffitt, & M. Stouthamer-Loeber (1996) between these 3 patterns and the Big Five were fully replicated. Growth curve analyses showed that the 3 patterns predicted important developmental outcomes in both the social and the cognitive domains. Evidence was found for both traits and types: A continuous dimension of resiliency bifurcates in its lower part into two relatively discrete personality types, overcontrollers and undercontrollers.

Head‐to‐head comparison of the predictive validity of personality types and dimensions

“…In two studies of Big Five self-ratings in adulthood and one study of Big Five parental ratings of their children, Asendorpf et al (2001) found that only three clusters were replicable. These types showed substantial consistency, not only across the three clustering studies but also with the Q-sort types derived by Asendorpf and van Aken (1999), and could once again be interpreted as resilient, overcontrolled, and undercontrolled.…”

Co‐occurrence of depression and delinquency in personality types

“…This pattern of opposites could be explained as follows. Hart et al (2005) found that although overcontrollers and undercontrollers are remarkably similar in terms of physiological and cognitive processes, which could be due to their low levels of ego-resiliency (Asendorpf & van Aken, 1999;Robins et al, 1996), they differ radically at the behaviour level (Hart et al, 2005), which could be due to their markedly different levels of ego-control, namely high for overcontrollers and low for undercontrollers (Asendorpf & van Aken, 1999;Robins et al, 1996). In this respect, ego-control may play an important role in explaining the opposite patterns of the problem behaviours over time, especially for the personality groups that do not respond flexibly to their environment, such as the overcontrollers and undercontrollers.…”

“…Consequently, Dubas et al (2002) investigated whether the personality types could be constructed directly on the basis of the Big Five dimensions: this appeared to be possible. The personality types of Block and Block (1980) have been replicated in many studies using different informants, different methods and different statistical techniques (see: Asendorpf & van Aken, 1999).…”

“…We will use the personality typology of Block and Block (1980) to describe the adolescents' personality, since a series of studies have shown systematic differences in depression and delinquency between the three personality types that Block and Block distinguish (Akse et al, 2004;Asendorpf & van Aken, 1999;Dubas et al, 2002;Hart et al, 1997;Robins et al, 1996). Block and Block describe personality in terms of two continuous dimensions: ego-control and ego-resiliency.…”

Personality–relationship transaction in young adulthood.

“…This questionnaire was derived in a multistep procedure from a pool of 179 bipolar items used by Ostendorf (1990) for adults. The steps included deletion of items that were not clearly understood by 12-year old children in a pretest screening, selecting among the remaining items the 12 highest-loading items for each factor in a forced 5-factor solution of the original adult sample of Ostendorf (1990), assessing these 60 items in the LOGIC sample at age 12, and selecting 8 items per scale with regard to their factor loadings and cross-loadings in this assessment (see Asendorpf & van Aken, 1999, in press, for more details). The internal consistencies (Cronbach's a) of the resulting 8-item scales were sufficiently high with two exceptions (for age 12, mean a 5 .76, range .68 À .83; for age 17, mean a 5 .82, range .75 À .88).…”