Stability of aggressive reaction patterns in males: A review.

Abstract

Considered in the review are 16 studies on the stability of aggressive behavior and reaction patterns. There is great variation among the studies in sample composition, in definition of variables, in method of data collection, and in the ages and intervals studied. Generally, the size of a (disattenuated) stability coefficient tends to decrease linearly as the interval between the two times of measurement (T2— 7\) increases. Furthermore, the degree of stability can be broadly described as a positive linear function of the interval covered and the subject's age at the time of first measurement, expressed in the age ratio T-^/T^,. The degree of stability that exists in the area of aggression was found to be quite substantial; it was, in fact, not much lower than the stability typically found in the domain of intelligence testing. Marked individual differences in habitual aggression level manifest themselves early in life, certainly by the age of 3. It was generally concluded that (a) the degree of longitudinal consistency in aggressive behavior patterns is much greater than has been maintained by proponents of a behavioral specificity position, and (b) important determinants of the observed longitudinal consistency are to be found in relatively stable, individual-differentiating reaction tendencies or motive systems (personality variables) within individuals.