LINEAR PROBABILITY-MODELS OF THE DEMAND FOR ATTRIBUTES WITH AN EMPIRICAL APPLICATION TO ESTIMATING THE PREFERENCES OF LEGISLATORS

We formulate and estimate a rigorously justified linear probability mo del of binary choices over alternatives characterized by unobserved at tributes. We apply the model to estimate preferences of congressmen as expressed in their votes on bills. The effective dimension of the att ribute space characterizing votes is larger than what has been estimat ed in recent influential studies of congressional voting by Poole and Rosenthal. Congressmen vote on more than ideology, Issue-specific attr ibutes are an important determinant of congressional voting patterns. The estimated dimension is too large for the median voter model to des cribe congressional voting.