Abstract
When using panel data important problems are often conveniently overlooked. These include model misspecification, asymptotic stability, unequally spaced panel waves, and the use of ordinal rather than metric data. While panel data may be useful to eliminate specification error, if the process generating data is in equilibrium, the problems of misspecification persist if a model with a lagged endogenous variable is formulated. Furthermore, the notion of asymptotic stability and its relation to the notion of inertia, which is often called stability by social scientists, is discussed. An important practical question is how to deal with data from panels with unequally spaced intervals between waves. Finally, the issue of the treatment of ordinal dependent variables is addressed. Using ordinal probit thresholds, the question of the variance of an ordinal variable across time is clarified and a quick estimator for this variance is given.
Get full access to this article
View all access options for this article.
