Taking Time Seriously: Time-Series-Cross-Section Analysis with a Binary Dependent Variable

Researchers typically analyze time-series-cross-section data with a binary dependent (BTSCS) using ordinary logit or probit. However, BTSCS observations are to violate the independence assumption of the ordinary logit or probit statistical. It is well known that if the observations are temporally related that the results of ordinary logit or probit analysis may be misleading. In this paper, we provide a simple for temporal dependence and a simple remedy. Our remedy is based on the that BTSCS data are identical to grouped duration data. This remedy does not require BTSCS analyst to acquire any further methodological skills, and it can be easily in any standard statistical software package. While our approach is suitable any type of BTSCS data, we provide examples and applications from the field of Relations, where BTSCS data are frequently used. We use our methodology reassess Oneal and Russett's (1997) findings regarding the relationship between economic, democracy, and peace. Our analyses show that (1) their finding economic interdependence is associated with peace is an artifact of their failure to for temporal dependence yet (2) their finding that democracy inhibits conflict is even taking duration dependence into account.