Institutions Matter?
Author:
- Adam PrzeworskiAPAdam Przeworski
- 1 January 2004
- journal article
- Published by Cambridge University Press (CUP) in Government and Opposition
- Vol. 39 (4) , 527-540
Abstract
The main methodological problem in assessing the impact of political institutions on any kind of performance stems from the possibility that institutions may be endogenous. As a result, institutions cannot be matched for the conditions under which they function. Inferences from such non-experimental observations are subject to several biases and, in the end, our conclusions may not be robust. One should not be confident, therefore, that any institutions would function in the same way under conditions different from those from which they are transplanted.Keywords
This publication has 5 references indexed in Scilit:
- Causal Inference without CounterfactualsJournal of the American Statistical Association, 2000
- THE ESTIMATION OF CAUSAL EFFECTS FROM OBSERVATIONAL DATAAnnual Review of Sociology, 1999
- Instrumental Variables: A Study of Implicit Behavioral Assumptions Used in Making Program EvaluationsThe Journal of Human Resources, 1997
- Counterfactuals and Hypothesis Testing in Political ScienceWorld Politics, 1991
- Statistics and Causal InferenceJournal of the American Statistical Association, 1986