Seeing the same evidence differently: Biased assimilation and moral conviction in public evaluations of scientific expertise

Abstract Particularly in democracies like the United States, the effective use of expertise to inform better policy decisions depends on public buy‐in. One barrier to this is biased assimilation, wherein individuals evaluate expert‐based knowledge, and the experts who promote it, differently based on alignment with their existing policy attitudes. While biased assimilation effects are well‐established, less is known about whether attitude‐level attributes like moral conviction may moderate them, as well as whether this effect may spill over into more general attitudes toward science. Using a two‐wave survey experiment in a sample of U.S. adults, this study confirms biased assimilation effects, as well as a novel moderating effect such that biased assimilation is strongest when it comes to attitudes held with strong moral conviction. As the moralization of an attitude increases, so do evaluations of pro‐attitudinal scientific knowledge and expert recommendations, suggesting that moral conviction may make people less critical information consumers. However, I find little evidence that these dynamics carry over into general attitudes toward scientific knowledge and scientists, even among those with the strongest moral conviction. These findings should temper fears that moralization or the use of expertise in divisive policy issues will erode general public support for science.