People Mistrust Science in General, But Not Specific Studies
Uncertainty decreases people's trust in science
Posted Feb 13, 2017
This has been a difficult era for science in the public eye. Debates about climate change treat the science related to the influence of humans on the climate as less strong than it is. Parents use discredited studies linking vaccines to autism to justify not vaccinating their children. And there have been high-profile cases in psychology of fraud and studies whose results did not replicate.
Part of the mistrust of science comes from difficulties nonscientists have in understanding that the results of every study are uncertain to some degree. For one thing, the measures people use in science are not perfect and may have some error. For another, some sciences like astronomy and geology observe the past and so they cannot experimentally manipulate the object of study. On top of that, there may be conflicting results within a field that can make it hard to determine what someone should believe.
How does this uncertainty affect the public’s perception of science? This question was explored in an interesting paper in the February, 2017 issue of the Journal of Experimental Psychology: General by Stephen Broomell and Patrick Kane.
In one study, they were interested in a general question about the sources of uncertainty in different sciences. They gathered ratings about 16 fields of study including psychology, climate science, nuclear physics, and forensics. They had participants rate these sciences for various sources of uncertainty including their perception of the quality of the evidence the science produces, the accuracy of measurements, the specificity of predictions the science makes and whether the scientists gather enough data to support their theories.
First, they used a technique called factor analysis to analyze the responses across all the questions for the fields of study. Factor analysis allows researchers to determine how the answers to each question tend to hang together. They found that the questions tended to reflect three dimensions of people’s beliefs about sciences. In the perception of many people, sciences vary in how precise they are, how much they use math to create broad abstract statements, and how much they study the present day versus the distant past.
The various fields of study differ along these dimensions. People view psychology as the least precise science, while forensic science is seen as most precise. Psychology is also seen as using very little math, while astrophysics and economics are seen as using a lot of math. Psychology studies the present day, while geology studies the distant past.
A second study looked just at psychology and manipulated the cover story to try to influence how uncertain people thought findings in the field were. One cover story focused on the difficulty of getting good measurements in psychology and the problems with understanding the complexity of the brain. The second cover story focused on advances in precision for measuring behavior and for imaging the active brain.
Participants were given one of the cover stories. They answered several questions about the precision of the science and also addressed additional questions like whether the research has social benefit and should be funded by the public. This study was conducted in the United States, and people’s political affiliation (Democrat or Republican) was also gathered, because some research suggests that party affiliation may affect the way people view science.
This study found that people who identified as Democrats were relatively unaffected by the manipulation of the cover story about precision. However, people who identified as Republicans felt psychology was less precise when given the low-certainty cover story than given the high-certainty cover story. Republicans were also less likely to see societal benefit for research or to agree with public funding of science when the science was described as more uncertain than when it was described as more certain.
So—at least for some people—seeing a science as less certain decreases its appeal as something that should be supported by public funds.
The final study looked at specific studies. The researchers gathered several studies on topics like teen crime that could have been done in psychology (which was seen as fairly imprecise) or forensics (which was seen as fairly precise). Some participants were told that the studies were psychology studies. Other participants were told that he studies were studies in forensics. Then, they read descriptions of 8 studies. They were told that some studies were repeatable studies, while others were studies whose findings did not replicate. For each study, they stated whether it was replicated or whether it had been overturned. (In fact, half of the studies chosen were ones that replicated and half were ones whose findings had been overturned.) Once again, political affiliation was measured.
In this study, participants judged about the same number of studies as replicable regardless of whether they were labeled as psychology studies or forensics studies. Political affiliation also had no impact on the results. People were paying attention to the study, though, because the experimenters also manipulated the descriptions so that some studies were described as having a high level of agreement among experts about it, while others were described as having a low level of agreement. Participants were more likely to think that a study had been overturned when it was described as having a low level of agreement among experts than a high level.
What does all of this mean?
People differ in their beliefs about the precision and certainty of different sciences in general. These general beliefs affect at least some people’s judgments about whether that science is worthwhile and whether it should be funded by the public. However, group differences based on political affiliation disappear when people start to focus on specific research rather than the science in general.
This pattern of results suggests that researchers may want to start by describing particular studies to people in order to help them understand the research that gets done in science. Then, they should relate the specific findings back to the area of science that it comes from to help people change their general beliefs about the quality of work done in those sciences. In this way, the scientific community can help the broader public to see the benefit and value of the research that gets done.
Broomell, S.B. & Kane, P.B. (2017). Public perception and communication of scientific uncertainty. Journal of Experimental Psychology: General, 146(2), 286-304.