Social sciences: Nature’s endorsement of Biden assessed | Nature Human Behaviour | Nature Portfolio

The endorsement of Joe Biden for US president by the journal Nature reduced stated trust in the journal and scientists in general among supporters of Donald Trump, according to a study involving over 4,000 participants in the USA, published in Nature Human Behaviour.

Leading up to the 2020 US presidential election, several scientific publications — including Nature — endorsed the candidacy of Joe Biden for president. In a preregistered experiment conducted in late July and early August 2021, Floyd Zhang randomly assigned 4,260 US participants to receive either a message summarizing Nature’s endorsement of Joe Biden followed by a screenshot and link to the piece, or a control message, screenshot and link regarding Nature’s announcement of its new visual designs for its website and print copies.

The author found that the endorsement message was associated with reductions in stated trust in Nature among supporters of Trump. When prompted to acquire information about emerging COVID-19 variants and vaccine efficacy, those Trump supporters who had received information about the endorsement were 38% less likely than those in the control group to access stories from Nature’s website. The endorsement also reduced the trust of Trump supporters in US scientists in general. Biden supporters’ trust in Nature and scientists were not reliably affected by the endorsement. There was also little evidence that the endorsement changed the views of participants about Biden and Trump.

Although the experimental results suggest that the political endorsement may affect confidence in the journal among a particular group, the research does not show that the endorsement had a negative effect on those outside of the experimental conditions. It also cannot answer whether such effects would generalize to other scientific journals or scientific organizations. The author stresses that future research is needed to understand the longevity of the reported attitudinal and behavioural effects.