4 ms·
When people say scientists cannot be trusted, they’re talking about themselves and their biases. That is, if that critic was a scientist, then that critic would
by nis0s 2y ago
When people say scientists cannot be trusted, they’re talking about themselves and their biases. That is, if that critic was a scientist, then that critic would do whatever they could to advocate for an ideological cause, by lying about, or misrepresenting, facts or findings.
So it’s problematic to ignore such concerns as conspiracy or radicalization when these concerns, whether directly or indirectly, are admitting to and acknowledging a very human trait, tribalism. We all have our biases, and believe what we believe about what’s best for society, the world, and humanity.
Personally, I think the only way to bring trust back to science, outside of any concerns about politics, is to have double-blind third-party scientists use your raw data to validate your hypotheses. I think this will also help address the reproducibility crises. Am I wrong?
- bediger4000 2y agoYour first paragraph is interesting. I'll think about it. I'm not sure independent validation is good enough. The flat earth renaissance seems to have occurred even though the flat earthers themselves independently validated curved earth results much of the time. There's also the issue of interpreting the data. Some of the big data things like vaccine effectiveness seems to require knowledge of statistic techniques that's beyond the ken of most humans, myself included. How do you overcome resistance to a conclusion that requires more statistics than "average"?
- nis0s 2y agoI didn’t mean that the raw data analysis should be done by laymen, I meant it should be done by peer scientists, who are trained in the same way as anyone else in your field. It might also help sending such raw data to someone who isn’t a competitor, or directly related to your field.
- bediger4000 2y agoI had a vague sense this was often done. In the popular articles about some paper, there's often a quote from a researcher in the same field that the article you explicitly notes is not involved in the original paper
- nis0s 2y agoTraditionally, scientists provide their analyzed results and figures to peer reviewers, but raw data itself is seldom provided to reviewers to confirm or reject a hypothesis in a double-blind fashion.