A new Pew survey updates their data on American’s trust in scientists. The “good news” is that overall, trust remains high, with 77% saying they trust scientists a great deal or a fair amount, and only 23% not too much or not at all. Actually, when you think about it these numbers are still pretty bad, but they seem good because our expectations are so low. More than one in five people don’t trust scientists. For more perspective, that 77% figure is the same for the military. The highest rated group was medical scientists at 80%. Elected officials were at 28%.

These numbers are also fairly stable over time. Interestingly they did bump up a bit during the pandemic, but then quickly returned to their historical levels. Some argue that these numbers are pretty good and we shouldn’t “freak out about the minority.” I disagree – not that we should freak out, but we do need to take these numbers seriously, and they are not necessarily good news.

One reason I am still concerned about these numbers is that there is a pretty significant partisan divide. Recent years have Democrats at around 90% with Republicans around 63%. More than a third of one major political party does not trust scientists, and they seem to be the political center of the party. This gets even worse if you look at the question of whether or not scientists should play an active role in policy debates. Only 66% of Democrats say yes, and only 29% of Republicans (down from 75 and 43 respectively). This, to me, is very telling. It’s one thing to say you trust scientists, but what does that mean that you also don’t want them to play an active role in policy?

To me this is a clear sign of compartmentalization. People can maintain multiple contradictory beliefs at the same time. It’s one thing to say you have trust in scientists, but this means nothing until that trust is tested. When scientists are saying something that cuts against your political or religious beliefs, that is when we see if you really trust in the process of science and the experts who engage in that process.

There has been an enduring question among those doing this kind of research. Is the partisan divide a manifestation of some inherent personality difference between liberals and conservatives, or are they just situational (or some mix of both). In other words, do Republicans have lower trust in science because they are more religious, and more likely to hold views that contradicts science? Is it because come major conservative political views also happen to go against science (like global warming)? Would Democrats be just as antiscience if the evidence contradicted a closely held ideological view? The evidence is a little ambiguous on this question, and could be used to support either view.

Show me a liberal environmentalist who says – you know what, the evidence does support the safety of GMOs, so I have to change my opinion on that – and then I will show you someone who truly trusts science. Same for a conservative who accepts the science on climate change against party orthodoxy because that is what the scientists say. Trusting scientists when they align with your politics earns you no credit. Saying you trust scientists, but not with policy, is signaling that you want to selectively ignore the science when it is inconvenient to your policy positions.

Only 29% of Republicans want scientists engaged in policy – which means to me that for all practical purposes, the strong majority of Republicans want to ignore or contradict scientists when they conflict with their ideological preferences. This is nothing to celebrate, and saying that you support scientists in the abstract (but not with actual input into real decisions) means nothing. For Democrats the number is only 66% – meaning a full third of Democrats don’t want to be constrained by science when it comes to their preferred policy. Keep in mind – the question was about whether scientists “take an active role in policy debates about scientific issues”. This is not about giving them power to decide, and only applies to scientific issues. Think about it – 71% of Republicans don’t want to here from scientists about scientific issues related to policy.

I also think that, in practice, the numbers are actually worse than what people say on a survey. I think there is still an overall cultural respect for science and scientists, and people don’t want to seem anti-science. But in practice they will choose their tribe or ideology over what the scientists say. There are also many ready-made narratives to justify this – scientists are in the pocket of big industry, they are disconnected from real people, they are part of some conspiracy. And of course, scientists are people who make mistakes and have their own biases and frailty, so there will always be anecdotes to support being critical of scientists. And having a general trust in scientists in the aggregate does not mean having absolute faith in all scientists.

Another trend in this data is that higher levels of education are associated with greater trust in scientists. So one solution to this situation is to improve overall scientific literacy in the public. That won’t be enough by itself, but it could make a critical difference.