3 ms·
I see these studies as a first step in a Bayesian inference. Trust comes with more evidence but you have to start somewhere. Edit: Actually this is the third
by ganzuul 2y ago
I see these studies as a first step in a Bayesian inference. Trust comes with more evidence but you have to start somewhere.
Edit:
Actually this is the third step. From the source:
https://www.ipsos.com/sites/default/files/ct/news/documents/2024-05/Ipsos%20report%20-%20Single-use%20plastics.pdf https://www.ipsos.com/sites/default/files/ct/news/documents/...
- SllX 2y agoLet me clarify: you cannot make useful judgements on the basis of an issue poll alone. The information they obscure is just as important as the information they purport to present. This is why we have legislators who in their offices actually investigate and discuss issues of lawmaking with their base, and can even contextualize it within a framework of privileged or classified information they may have access to rather than take an issue poll at face value that says “most of <nation’s population> support X on issue Y” and simply vote that way.