Personality and system influences on behavior in groups: Frequency models

Abstract

Social interaction depends on a great variety of simultaneous factors. Aggressive behavior of person A towards B e.g. commonly is explained by one or several of the following factors: (a) a personal disposition of A to behave aggressively towards others (the ‘aggression- potential’ of A); (b) a capability of B to elicit aggressive behavior from others (B's ‘cue value for aggression’); (c) an antipathy of A to B (the relationship between A and B). It is obvious that the constellation of simultaneous factors cannot be analysed by means of ‘pure’ observational methods only. Several principles for the use of theoretical concepts in explaining this kind of behavior are developed in the present paper. It is postulated that the proof of the validity of a theoretical model has to be independent of the (qualitative or) quantitative identification of the theoretical constructs of the model (thesis 3). The identification of theoretical constructs in turn has to be based on mutually independent parts of evidence present in the data (thesis 2). It is very unlikely that a complex constellation of social factors will occur on more than one occasion. Thus the required independent parts of evidence have to be derived from a single set of data. Formal models are suggested, which are in complete agreement with the above postulates. The model control and estimation of independent sets of parameters can be separated by conditional inference procedures in these models.