ABSTRACT
A persistent concern in the analysis of public opinion data is the "meaning" that one can ascribe to the observed distributions and trends—and to the positions taken by particular individuals and segments of the population. In a certain sense, the need for more detailed information about opinions be met by improvements and refinements in the methodology of opinion assessment. But, no matter how refined the techniques, they do not provide direct information about the meaning of the opinions and do not permit automatic predictions to subsequent behavior: the investigator still has to make inferences from the data. Social influence has been a central area of concern for experimental social psychology almost since its beginnings. It can also be observed, for example, in the context of socialization of children, where the taking over of parental attitudes and actions is a normal, and probably essential, part of personality development.
