Spatial models are ubiquitous within political science. Whenever we confront spatial models with data, we need to measure policy distances in political space. The most commonly used measure is the Euclidean measure. This choice however can hardly be justified from a political behavior perspective. After all, why should we expect the individuals to ‘behave’ in accordance with some geometric basis?
In this paper, I develop a general, latent utility based spatial voting framework to look into this question. Using survey data for different country-elections, I test whether the choice of Euclidean measure is justified. I show that this is hardly the case. Furthermore, I show that it is possible to estimate the distance measure parameter (2, in the case of Euclidean metric) using individual level survey data. In fact this estimate can provide us valuable insights regarding to the underlying political behavior
