Squire and Smith use 1982 California sate supreme court confirmation contests and rely on survey data where the sample was split between people who were told the partisan affiliation of a candidate, and those who were not. Conclusion:
Partisan information increases the probability of an individual holding an opinion on the
elections, and results in votes which are based on the respondent's partisan identification and
opinion of the governor who appointed the justice.
Snyder and Lim, more recently, have strong findings. The sad news:
we find that incumbent judges' quality has little effect on their vote share or probability of winning in partisan general elections...
we fi nd that voting is highly partisan in partisan judicial elections -- i.e., there is a strong correlation between the Democratic "normal vote" and the Democratic vote share for judges --but not in non-partisan or retention elections.
The nice thing with Woodford's paper is that the cognitive constraint explains the decoy effect. Consider the example of Huber et al. two goods that differ on price an quality, one is low price-low quality and one is high price-high quality. In the Figure below, those are the "Competitor" and "Target" goods. Each good dominates the other on one dimension. Let's assume for our example that the Target good is the high-quality high-price one. You can look at the choice people make based on this choice set.
Partisan labels are useful because they provide an easy dimension of comparison between the two candidates. One extra problem is that the interpretation of party labels is not clear. But that's a question for another time...