PUB TYPE Reports- Research/Technical (143)-- Speeches /Conference Papers (150)

To investigate the temporally ordered process in image and issue learning as it affects the voting decision, a study conducted a secondary analysis of panel survey data from Wisconsin during the 1976 presidential elections. The study hypothesized that, because of the difficulty in understanding complex issues, voters rely on image evaluations for initial candidate preference--issue positions are then learned to support the image-based choice. The survey consisted of random telephone interviews: the first set of interviews was completed shortly before the first presidential debates between incumbent Gerald Ford and Jimmy Carter; those who agreed to be reinterviewed were called again between the first and second debates. The third interviews occurred after the four debates but before the election; and final interviews followed the election. A total of 164 resperdents continued throughout the interviews. Measurements included vote intention, candidates ' personal image, respondents ' position on four political issues, ideological differences, and party identification and leaning. Precommitted or "one- issue " voters were eliminated, leaving 123 respondents for analysis. Causal analysis revealed that for Carter, images at an initial time period determined later candidate preference and issue learning. Also, image evaluations for Carter occurred at the beginning of the decision-making process. For Ford, however, images did not predict later issue learning and candidate preference--initial issues were the predictor. Also for Ford, vote intention played a significant role in later issue learning. (Four tables of data, 4 figures, and 49 references are appended.) (MM