An investigation of a proposed public transit was based on the assumptions that knowledge of school tax referenda should provide a framework for understanding voter behavior in other types of referenda, and that propositions generated in one type of research should hold up in another. It was anticipated that there would be spillover effects from past school referenda onto a proposed future transit referendum. In planning the survey instrument a strong similarity between a school levy and transit levy in the community was conceptualized. However, the data on public transit predicted a contested referendum but nothing even closely resembling a polarized electorate as had been anticipated. The mismatch of predispositions and the data suggested that a comparison was being made between two different political dynamics that were related to the question of high versus low saliency among an electorate that would temporarily assume the role of fiscal decision-maker. The difference in potential saliency of contested school versus transportation referenda is accounted for with a descriptive model of the dynamics of contested school fiscal referenda. (Author/MLF)