Measuring Emotion: Self-Reports vs. Physiological Indicators

While emotion has gained popularity in political science research, its measurement is still an open question. To date, most work on emotion and behavior relies on survey respondents to accurately and honestly report their emotional responses to relevant stimuli. While this strategy is relatively inexpensive and efficient, its accuracy has been called into question. More specifically, it has been suggested that people (1) have a difficult time pinpointing specific reasons for their attitudes, (2) hesitate to give honest answers when such answers are socially undesirable, and (3) rationalize their answers to such questions for various reasons. An alternative measurement strategy involves recording respondents' physiological responses to stimuli, which indicate activation of the autonomic nervous system. Unlike self-reports, physiological measures are not subject to social desirability bias, and they can capture aspects of emotional response that are beyond respondents' conscious control. Despite a wealth of research from both political science and psychology on the measurement characteristics of emotion self-reports and a large body of psychological research on physiological measures, there is a lack of understanding on the degree to which the two types of measures "converge." Our purpose here is two-fold. First, we examine the extent to which physiological reactions to stimuli influence self-reports. Second, we look for conditions under which one measure outperforms the other in terms of explanation/prediction. In short, we find that self reports are better predictors of political attitudes, but also, self-reports may provide biased estimates of emotions' effects on attitudes in respondents that score high on emotion regulation scales.