Experimental results about linguistic voting

In this paper, we describe the results of experiments in which about 7000 voters in the Netherlands were asked in three different waves to give their most favored party and to give an evaluation on a scale of 0 till 10 of eleven major Dutch parties. We have applied five different voting rules to determine the number of seats each party would obtain in Parliament. Different from what one might think, in general voters had no problem to give an evaluation of eleven major Dutch parties. Interestingly, many voters gave the same evaluation to two or more parties, something they cannot do if they can only vote for one party. Although Majority Judgment has not been designed for a seat distribution in parliament, we describe two possible ways which enable such a distribution.