THE STRUCTURE OF EMOTIONAL RESPONSE: 1984 PRESIDENTIAL CANDIDATES GEORGE E. MARCUS Williams College Over the past two decadespsychologicalmodels of affect have changed from valence (one-dimensional) models to multiple-dimensional models. The most recentmodels, circumplex models, are two-dimensional. Feeling thermometer measures, which derivetheirtheoretical logic from earlier (valence) modelsof emotional appraisal, are shown to be confounded.Underlying the variationobtainedusingfeeling thermometer measuresare two dimensions of emotional response, mastery (positive emotionality) and threat(negative emotionality).Analysis of the 1984 NESsurvey sug- gests that positive emotional response is twice as influential as negative emotional responsein predicting presidential candidate vote dispositionto the presidential candi- dates. Reliance on emotional response is shown to be uniformly influential across various strata of the electorate. Policy considerations have little direct influenceon vote disposition, though policy considerations are indirectly relatedto vote dispositionthroughthe influenceof issues on the degreeof feelings of threatevoked by the candidates. It has been 15 years since it was first recognized that asking people about their feelingstoward presidential candidates was equivalentto ascertaining theirvote disposition(Brody and Page 1973). Only recentlyhave feel- ings-emotions-begun to be the subject of serious study by political scientists. The recent studies on emotions and poli- tics have shown that feelings have strong predictive effects on candidate appraisal (Abelson et al. 1982), on issue appraisal (Conoverand Feldman 1986), on political participation (Marcus 1985) and on polit- ical communication (Roseman,Abelson, and Ewing 1986). The primary concernof the initial studies has been to establish that emotions are worthy of study and that they add explanatorypower to our understandingof a variety of political phenomena.' While the studies cited above have established the value of studying emo- tions and in redressing the imbalance resulting from the dominanceof the cog- nitive approach to human behavior, it must also be noted that thesestudieshave failed to provide a sound theoretical focus. The recent attention to affect has been marked by contradictory and anom- alous findings; so that even when the cor- rect structure of emotional response has been identified, its significance and rele- vance have not been properly understood (e.g., Abelson et al. 1982). Contradictory findings have been reported in different AMERICAN POLITICAL SCIENCE REVIEW VOL. 82 NO. 3 SEPTEMBER 1988 This content downloaded from 137.165.4.15 on Sat, 13 Jul 2013 07:44:42 AM All use subject to JSTOR Terms and Conditions