Publication

This paper discusses the use of models with limited-dependent variables, which allow researchers to test important relationships in political science. Often, however, researchers employing such models fail to acknowledge that the violation of some basic assumptions can have different consequences in nonlinear models than in linear ones. The author demonstrates this for probit models in which the dependent variable is systematically miscoded. Contrary to the linear model, such misclassiffications affect not only the estimate of the intercept, but also those of the other coefficients. In a Monte-Carlo simulation, the author demonstrates that a model proposed by Hausman, Abrevaya and Scott-Morton (1998) allows for correcting such biases.