Date created: | Last Updated:

Category: Communication

Description: Exploratory methods for scaling text, votes, and other count data are widely used for modeling policy preferences and their common logic is naturally ex- tended to more than one policy dimension. It is well known amongst practitioners that the substantive interpretation of induced dimensions requires care and substantive knowledge. Less well known is that the combination of orthogonality constraints necessary to identify multiple dimensions in count data force a predictable but entirely artifactual geometrical structure into position estimates that may easily be mistaken for substantive variation. This note characterizes the nature of this geometrical artifact, suggests when it may be expected to appear, and briefly considers possible remedies.