Key points are not available for this paper at this time.
This article expresses a variety of theories of public opinion change in terms of a formal model for analyzing public opinion data. Statistical criteria are proposed whereby models of varying complexity can be accepted or rejected. The observation is made that certain substantive theories of change are more parsimonious than others in terms of their statistical predictions. The model for analyzing change presented here is based on the premise that the most parsimonious theories should be accepted or rejected first.
D. Garth Taylor (Tue,) studied this question.