International Society of Political Psychology

Conference Abstracts

Daily Temperature Deviations Affect Beliefs about Climate Change:  The Political Implications of Loc

Lisa Zaval, Columbia University; Ye Li, Columbia University; Eric Johnson, Columbia University

Conference: ISPP 2011
Affiliation: Columbia University - Teachers College
Research Area: Political decision making

Despite widespread public awareness of global warming, there is reason to suspect that beliefs about climate change are constructed in response to questions and therefore malleable. Specifically, reports of these beliefs might include irrelevant information such as the highly variable day-to-day deviation in temperature. This attribute substitution occurs when a simple and very accessible judgment replaces a more complex and less accessible one. In three studies, we asked residents of the United States and Australia to report their opinions about global warming and if their cities were warmer or cooler than usual today.  The results show that respondents who thought today was much warmer than usual reported more belief in and concern about global warming than those who thought today was much colder than usual. They also donated more money to a climate-change charity when today seemed warmer than usual.  Additional regressions revealed that these effects remained significant after controlling for objective temperature, demographics, political affiliation, geographical region, and environmental attitude. Question order had no impact on effect size, suggesting that the potential for conscious awareness of the effect by respondents is limited. We use instrumental variable regression to rule out many alternative explanations. It is, perhaps, the issue’s very complexity, and people’s ambivalence toward global warming, that causes them to draw only temporary conclusions and to reconsider their belief each time they are surveyed or polled. These results raise important questions about the role of labile public opinion in the formulation of policy in complex areas.