Inducing Apparently Self-Interested Political Preferences

We argue that most previous empirical reports of the influence of personal economic situations upon political preferences (self-interest effects) in the mass public have been due to either of two item-order artifacts: political preferences may have been personalized by assessing immediately after the respondent's own economic situation has been made salient, or perceptions of personal economic situations may have been politicized by assessing them immediately after major political preferences have been made salient. Three sources of data were used to test the effects of these item orders on the consistency between personal economic situations and political attitudes: an experiment conducted within the 1979 National Election Studies (NES) pilot survey, and secondary analyses of NES and media surveys during the 1976 and 1980 presidential campaigns, as well as the full series of pre-1980 NES surveys. Self-interest effects prove to have occurred primarily in surveys with either personalizing or politicizing designs, and only rarely in uncontaminated surveys. We speculate that such survey-induced increases in consistency reflect temporary facework rather than a genuinely increased grounding of political preferences in selfinterest. Political campaigns may also find it difficult to induce a genuine link between self-interest and political preferences; NES data reveal little success for the Reagan campaign's effort to do so in 1980.