The same trend can be seen in media trust: https://news.gallup.com/poll/267047/americans-trust-mass-media-edges-down.aspx https://news.gallup.com/poll/267047/a

The same trend can be seen in media trust: https://news.gallup.com/poll/267047/americans-trust-mass-media-edges-down.aspx https://news.gallup.com/poll/267047/americans-trust-mass-med... Basically, "science" here means the social institutions of science like universities not the abstract notion of it, like how "the media" means the actual concrete media institutions and not the general concept of the press. And US Democrats trust institutions dramatically more when they realize that those institutions have become full-blown ideological allies rather than neutral. Post-2016 US media stopped pretending it was politically neutral and became full time water carriers for the Democrats. Trust on the left surged as a result. COVID made it clear to everyone that the same is true of public health - it's just blatantly ideologically compromised by leftists to an extent that people hadn't realized before, and thus leftists now trust it a whole lot more than they used to. For science it really means academia. The professoriate has become much more partisan over time. If you go back to the 1960s or so, the balance between right and left wing professors was far better. Then the left went on their famous Long March, which is basically code for finding reasons to harass or exclude non-left wing people, and universities became ever more dominated by ideology. Along the way, science became "science". Modern science is totally debased. The sort of people who most loudly beat the drum for Believe The Science™ tend not to have actually read the science they're promoting. Certainly, journalists never do. They just take it as axiomatic that because universities are filled with lefties they're all honest, intelligent and pure. Then they present some papers pushing blatantly ideological propaganda as "science", and when others read it reject it, label them anti-science. A good example of this is the papers pouring out of university social science departments on the theme of: why are conservatives so stupid and misinformed? The entire research theme is itself ideological, but then you discover their experiments are done on random recruits from TaskRabbit type sites (i.e. they have no idea who they are experimenting on).