Network to measure consensus among researchers is growing
23 January 2025
Jonas Roslund
In 2023, Linköping University joined a worldwide pilot project to systematically measure the extent of consensus among researchers on burning scientific issues. After a successful test, the number of higher education institutions that have registered interest has more than doubled. However, the project has its critics.
Professor Harald Wiltsche is happy with how the project turned outPhotographer: Magnus Johansson
“A common criticism is that it’s not relevant to measure consensus because science is about truth, not what the majority thinks. But that misses the point of the study,” says Professor Harald Wiltsche at the Department of Culture and Society.
He has been coordinator at LiU for the project led by Professor Peter Vickers from the University of Durham in the UK.
A simple question
In May 2023, more than 20,000 researchers in the natural sciences at 30 higher education institutions were invited to take a position on the following statement:
Science has put it beyond reasonable doubt that COVID-19 is caused by a virus.
The first question, which was about covid-19, was just a test.The answer was to be given on a scale of one to five. The purpose was really just to test the technology and methodology. Would it be possible to create a working network of experts in different fields and from different parts of the world? Would it be possible to avoid problems of too low response rates and long response times by keeping the question short and simple? And what degree of consensus could be expected?
An evaluation has now been presented in an article in the scientific journal PLOS One. According to Harald Wiltsche, the project has very much lived up to expectations.
The response rate was comparatively high: 33.9%. The answers were also evenly distributed among researchers in different fields of science. However, some parts of the world were under-represented. The survey itself was completed and the results compiled in a month.
A surprising result
What caused a little trouble was the final measurement result. It turned out that only 93.2% of the participants agreed that COVID-19 was caused by a virus. A worryingly low number, given that the claim is uncontroversial, the researchers admit. The question is how to interpret it?
Of course, there may actually be people who disagree that a virus caused COVID-19, but there may also be other explanations. Perhaps some participants think that nothing in science can be “beyond reasonable doubt” and that the statement is simplistic. Others may just like to be contrary or have answered without thinking.
The researchers behind the project believe that ultimately the result is not a weakness. Rather, it is to be expected from a survey that is sent out in many different languages and to many different cultural areas. The result could thus be used as a baseline to compare with in future surveys on more contested issues.
Those critical of the project use Einstein as an argument.Consensus and truth
Determining whether there is scientific consensus on a particular question is important, says Harald Wiltsche. It can help decision-makers and also prevent disinformation. However, as stated, there are critics. They point out how prominent scientists such as Galilei and Einstein, on the contrary, advanced science by opposing what the majority considered to be true.
“This is true, but their results only became significant when they eventually received support from their research colleagues. Consensus occurs when many experts independently study the same question and achieve the same result. We certainly do not mean that scientific truth is the same as consensus. But, if it’s well-founded, it’s the best indicator we have of the current state of knowledge in an area. This is precisely why it’s so important to develop methods for assessing consensus,” says Harald Wiltsche.
A growing interest
The research community has shown great interest in measuring agreement. After the first trial, the network of higher education institutions has grown significantly – from 30 to 80. This means that it now includes about 50,000 researchers. But what will happen in the future remains unclear. Efforts to find funding to make the project permanent are ongoing.
“I’m sure it will be successful. And, since I believe strongly in this, I’ll do my utmost to ensure that LiU continues to be involved when the next step is taken,” says Harald Wiltsche
