Tag
This paper consolidates the state of the art in supervised political scaling, investigating whether joint prediction of ideological scales and a middle ground between classification and regression can improve performance.