Voter Bias in the Associated Press College Football Poll

We investigate multiple biases in the individual weekly ballots submitted by the 65 voters in the Associated Press college football poll in 2007. Using censored tobit modeling, we find evidence of bias toward teams (1) from the voter’s state, (2) in conferences represented in the voter’s state, (3) in selected Bowl Championship Series conferences, and (4) that played in televised games, particularly on relatively prominent networks. We also find evidence of inordinate bias toward simplistic performance measures – number of losses, and losing in the preceding week – even after controlling for performance using mean team strength derived from 16 so-called computer rankings.