Sustainability-supporting laws: the curious lack of influence of scientific agreement about climate change
Another application of the agreement paradox to modern American society involves Americans’ beliefs about the environment. Despite the extremely high rate of consensus among expert climate scientists that climate change is occurring and caused by human activities (97 percent; for a review and synthesis of relevant literature, see Cook et al., 2016), the American public do not report holding corresponding beliefs. Indeed, far from reflecting the emerging scientific consensus, over the past 20 years Americans (if anything) have moved somewhat in the opposite direction (Saad & Jones, 2016).
A number of mechanisms have been employed in an attempt to provide insight into this discrepancy between expert and public opinion such as political bias (see e.g., Guber, 2012; McCright & Dunlap, 2011) and selective exposure (see e.g., Stroud, 2011). However, a recent series of studies (Conway & Repke, 2019) suggests part of the explanation involves perceived pressure for agreement producing reactance and informational contamination. In their first study, Conway and Repke (2019) measured the degree of informational contamination in participants’ assessment of the scientific consensus of climate change. Specifically, they gave participants a brief summary of a published study showing 97 percent scientific consensus on climate change and then asked several questions about participants’ attributions of the consensus (e.g., “I believe that the 97 percent agreement exists because there is a larger political agenda that puts pressure on scientists to conform to the views endorsed by the agenda”). Conway and Repke (2019) then used this informational contamination measurement to predict participants’ support for both governmental and civic action regarding climate change. They found that perceived informational contamination decreased the likelihood of supporting governmental and civic action in response to climate change — and, importantly, this effect remained even when controlling for participants’ ideology. In other words, informational contamination about scientific consensus eroded support for climate change-based action, even when ideologically Americans would be prone to support such action.
Two additional studies used experimental scenarios to test both informational contamination’s and reactance’s role in the political pressure rejection of governmental action path. In both studies, an agreement in favor of a prosustainability law emerged in a political context. The researchers manipulated whether or not a strong political leader (Study 2) or a highly visible law (Study 3) put pressure for public agreement. In both studies, they also measured informational contamination by asking questions about the degree that an apparent consensus for a pro-sustainability law was artificial versus genuine (e.g., “At the assembly where everyone spoke in favor of the law, to what degree do you think the unanimous discussion in favor of the law occurred because of the influence of the President?”). They measured reactance by asking questions about the degree that pressure to conform with the law made participants want to reestablish their freedom to choose (e.g.,“In the scenario, to what degree did the President’s discussion at the meeting make you want to do the opposite of what he said,just to show him that you could not be told what to do?”). In both studies, political pressure to engage in specific environmental policies operated through informational contamination and reactance to decrease support for governmental action - an indirect path that was overwhelmingly significant in both studies. Taken together, this set of studies suggests that part of the division in environmental sustainability attitudes in modern America isn’t just about pre-existing ideological differences; rather it is that many Americans perceive that there has been public pressure for agreement which has influenced science — and thus they both react against that emotionally and informationally discount that information.