A Controlled Experiment to Measure the Effects of Personal Contact Campaigning

This is a report of a controlled experiment performed in a camgaign for local office. The experiment was designed to measure the impact of door-to-door canvassing and telephone solicitation. Competing candidates in a partisan election agreed to campaign according to a pre-determined design in a pre-selected set of election districts. A post-election survey was executed in the test area in an attempt to discover intervening variables and as a check on candidate behavior. The data are analyzed with a regression model, and the findings are compared to other studies in the field. An explanation is offered to reconcile these findings with those of other studies.