作者: Leo A. Goodman
DOI: 10.1086/223266
关键词: Period (music) 、 Econometrics 、 Panel data 、 Psychology 、 Processes of change 、 Process (engineering) 、 Convention 、 Attitude change 、 Democracy
摘要: Methods for making predictions and testing various hypotheses concerning processes of change are presented here. Applying these methods to the reanalysis panel data on pre-election vote intention, we find that tests given here permit more detailed analysis than did applied earlier suggested process attitude in July-August period (during which time Democratic convention was held) differed from August-September period, indicates corresponding differences between two periods were not statistically significant except relating those who "Didn't Know" July August respectively; is, apparent over-all can be attributed Know."