The pivotal role of perceived scientific consensus

The embargo on my latest paper on the cognition of climate change, published in Nature Climate Change, has now been lifted. The paper and abstract are available using the doi: 10.1038/10.1038/NCLIMATE1720.

The paper is entitled “The pivotal role of perceived scientific consensus in acceptance of science”, and its abstract reads as follows:

Although most experts agree that CO2 emissions are causing anthropogenic global warming (AGW), public concern has been declining. One reason for this decline is the ‘manufacture of doubt’ by political and vested interests, which often challenge the existence of the scientific consensus. The role of perceived consensus in shaping public opinion is therefore of considerable interest: in particular, it is unknown whether consensus determines people’s beliefs causally. It is also unclear whether perception of consensus can override people’s ‘worldviews’, which are known to foster rejection of AGW. Study 1 shows that acceptance of several scientific propositions—from HIV/AIDS to AGW—is captured by a common factor that is correlated with another factor that captures perceived scientific consensus. Study 2 reveals a causal role of perceived consensus by showing that acceptance of AGW increases when consensus is highlighted. Consensus information also neutralizes the effect of worldview.

There are a number of interesting aspects to the results, and I am highlighting three that I find particularly noteworthy or intriguing.

The first one is the finding that people’s attitudes towards science are a mixture of specific opinions (i.e., how much people endorse climate science or the link between tobacco and lung cancer) as well as a general factor (i.e., how much people endorse scientific propositions in general). This is quite interesting because it means that there is something about science in general that (partially) determines people’s acceptance of scientific propositions about issues as diverse as tobacco, HIV, and climate.

We furthermore showed that this general factor capturing acceptance of science was correlated with another general factor that represented perceived consensus among scientists. That is, the public’s views on science are at least partially driven by the extent to which people perceive agreement among scientists.

This correlation is neither surprising nor entirely new, as other researchers have reported similar results for climate science (although no one has previously reported the involvement of a general factor). It also explains why climate deniers expend considerable effort to negate the existence of that consensus, using the usual array of deceptive techniques such as pseudo-experts, or pointing to unreviewed blog-posts as “evidence” for their contrarian positions.

The second intriguing finding is that when people were explicitly informed about the scientific consensus on climate change, they became significantly more likely to endorse the basic premise of global warming, and they attributed a larger share of the observed warming trend to human CO2 emissions, than people in a control condition who received no such information. This result suggests that consensus information causally contributes to people’s acceptance of scientific propositions.

The final intriguing finding was that the effect of the consensus information was particularly effective for people whose “free-market” worldview predisposed them to reject climate science. It has long been known that personal ideology or “worldview” is a major driver of people’s attitudes towards climate change: the more strongly people endorse a “fundamentalist” view of the free market, the more likely they are to reject climate science. The role of worldview presents a formidable challenge to science communicators because ideology may override any factual information. Worse yet, the provision of factual information may lead to “backfire” effects that reduce—rather than enhance—acceptance of science among people with extreme worldviews.

The fact that in our study, the provision of consensus information attenuated the role of worldview and increased acceptance particularly among people who maximally endorsed the free market may therefore present an avenue to overcome the communication challenge faced by climate scientists.