초록 열기/닫기 버튼
In general, it is very hard to conceptualize the very nature of adaptive behaviors. One of the most crucial problems is the lack of relevant theoretical sources necessary for constructing a general theory of dissident-state interaction. Difficulties in collecting relevant data provide another problem. There always exists a certain degree of scarcity of relevant data that allows researchers to investigate the dynamic nature of dissident-state interaction over a long period of time in various dimensions. Thus, the primary purpose of this article is to validate the methodological applicability of dynamic time-series econometric modeling to the study of protest, in particular, to that of protest-state interaction dynamism—in other words, adaptation. This article examines the theoretical plausibility of applying an advanced time-series econometric model to the study of dissident-state interaction by introducing the Vector Auto-Regressions (VARs) model, which has been intermittently adopted by the students of terrorism studies. The results show that the direct use of “a Vector Auto-Regressions model” is helpful for explaining a dynamic model of interaction between dissidents and state, especially in such a democratic country as France. Another interesting finding is that the result of variance decompositions reveals—though that of Granger-causality test does not endorse any causal relation between the number of protest participants (LPRO) and the level of protest’s organizational power (LOSD)—that the two series exhibit important co-movements. The correlation coefficient between error terms is 0.586, which is relatively higher than any other coefficients in the model. This means that LPRO and the number of state force (LSF) are the dominant factors of protest-state interaction. Relative to their dominance, the other variables demonstrate relatively small impact on each other.
