[1] After receiving his doctorate, Collier taught at Indiana University, Bloomington, where he was promoted from Assistant to Associate Professor in 1975.
Collier has played an active role in building the Institute for Qualitative and Multi-Method Research (IQMR), an international training program held annually at Syracuse University.
Focusing on the fact that the most industrialized Latin American countries were not the more democratic ones, it articulated several critiques of the prevailing view of Third World politics put forth by modernization theorists (e.g., Seymour Martin Lipset).
Thus, at a time when comparative politics sought to provide a general theory of politics in a way that largely disregarded insights developed by regional experts, Collier's New Authoritarianism provided an alternative, demonstrating that area studies could be a site of creative theorizing.
Moreover, the applications that draw on Collier's core ideas about critical junctures and path dependence go beyond political science to economics.
This argument was examined through an ambitious comparative historical analysis of eight countries over a period of five decades that relied on four paired comparisons.
Nowadays it has become commonplace to see Brazil and Chile as exemplifying a common trajectory with regard to political and economic development.
In brief, Shaping the Political Arena makes a major theoretical contribution through its elaboration of a critical juncture and path dependency model and also stands as the most ambitious and systematic work on Latin America in the tradition of comparative-historical analysis.
[18] Collier shows that typologies, when used carefully and systematically, can help form the key concepts in substantive research and are also an essential tool for theorizing.
Here, central thrusts of Collier's work have been to put ideas about process tracing on a far more secure footing, and more broadly to codify procedures for qualitative causal inference.
2010, coedited with Henry Brady), and in articles in leading journals, challenges the conventional view that statistical tools are always more powerful for causal inference.
These may be defined as pieces of data that provide information about context and mechanisms, and that offer distinctive leverage in making causal inferences.
Another direction of Collier's work, which springs from the recognition of the shared challenges faced by quantitative and qualitative researchers, has been the exploration of multi-method strategies.