Elections under biased candidate endorsements — an experimental study

We construct an election game to study the electoral impacts of biased candidate endorsements. derive a set testable predictions. test these in laboratory experiment and find that observed outcomes vote shares are well predicted. no support, however, for our prediction relationship between outcome endorser's bias is non-monotonic; i.e., ex ante, candidate's winning probability will first increase then decrease as endorser becomes more towards her. Voter turnout much less responsive than argue voting behavior can be explained, substantial extent, by three behavioral mechanisms: (a) distinct levels rationality choice decisions, (b) conservative belief updating, (c) ‘partial competition neglect’, where voters underestimate correlation information released endorsement closeness elections.