Published online by Cambridge University Press: 17 February 2020
Studies on discrete emotions typically work to evoke one emotion at a time. Yet many political phenomena cause multiple emotions. Threats, for example, cause, anger, and fear, have diametrically opposing behavioral consequences. As a result, the effect of experimental treatments can be masked by the countervailing influence of emotions with similar affect. This issue is exacerbated by existing measures of negative emotions, such as the Positive and Negative Affect Schedule (PANAS). We show that the PANAS is contaminated by systematic measurement error, as negative affect produced by one emotion influences responses on the other. To overcome this, we develop an alternative version of the PANAS that allows respondents to select which emotions they are feeling, then rate the severity. This technique accurately captures respondent’s emotional reactions, reducing measurement error and thus decreasing the correlation between fear and anger. The tactics we developed have broad relevance for experimental researchers analyzing emotional responses to politics.
This research was funded through the Boise State School of Public Service. A previous version of this manuscript was presented at the 2019 Annual Meeting of the Midwest Political Science Association. The authors declare no conflicts of interest. The data, code, and any additional materials required to replicate all analyses in this article are available at the Journal of Experimental Political Science Dataverse within the Harvard Dataverse Network, at: doi: 10.7910/DVN/MDMVZ5.