Spatial statistics of proximal factors on the political behavior of voters in a regional election: Case study of a mayoral election in Ako City, Japan

Abstract

This research applies spatial statistics to examine proximal factors affecting the political behavior of voters in a regional election in Japan, particularly, voter proximity to the election campaigns of the candidates. During the mayoral election in Ako City, Hyogo Prefecture, voters' political behavior, attitudes, and awareness of politics were measured using a social survey, the spatial location information relating to candidates' election campaigns being measured using GPS. Voters' favorable perception of a certain candidate was positively correlated to the degree of contact with his election campaign of voters themselves or that of their neighborhood, but not to spatial proximity with his campaign. On the other hand, both the degree of contact and spatial proximity with his election campaign of voters themselves led them to cast their votes for the candidate, even controlling for favorability. It was revealed that there is a possibility for proximal factors to be treated more precisely by applying spatial statistics.