Coding and Model Specifications
James F. Adams, Samuel Merrill, Bernard Grofman · 2005
Party identification is scored at one if the respondent identified with the candidate's party and zero otherwise; ideological distance is the negative of the squared distance between the respondent's self-placement and the candidate's position along the liberal–conservative scale; policy distance is the negative of the mean squared distance between the respondent's self-placements and the candidate's positions along the policy scales; candidate character is calculated as the average score the respondent assigned to the candidate on attributes such as intelligence, honesty, and leadership ability; race is scored at one if the respondent was black and zero otherwise; retrospective evaluation of the economy is coded from –2 (much worse) to 2 (much better). Each variable was included in the specifications for both the Democratic and Republican candidates' utilities except for race, which was omitted from the Republican candidate's utility. This was necessary in order to identify the model, since, as discussed in section 8.3, race was also included in the specifications for respondents' indifference and alienation thresholds. Education is coded on a seven-point scale ranging from less than high school education to post baccalaureate degree; political efficacy is calculated as the citizen's mean self-placement on the ANES political efficacy scales (recoded on a scale from zero to one representing low to high efficacy); previous vote is scored at one if the respondent reported having voted in the previous presidential election and zero otherwise; perceived election closeness is scored at one if the respondent believed the presidential election would be close and zero otherwise. Each variable was included in the specifications for both indifference and alienation except for election closeness, which was included only in the indifference specification.