Here’s the same data, but with trust broken down by political views (circles are trust among people on the left, +s the right).

It’s not just that the BBC is widely consumed — it also has solid trust on both left & right, whereas trust in the biggest US media brands is hugely polarised.

And I think, in those cases, the easiest solution that the field has found is to reportd STD (perhaps implicitly) as an upper bound.

I also find this often very confusing! But in some cases, a key issue is that evaluating SE over correlated measurements (e.g., as in K-fold cross-validation) brings numerous technical challenges, e.g.,
1. www.jmlr.org/papers/v5/gr...
2. doi.org/10.1080/0162...

Sure! econtent.hogrefe.com/doi/10.1027/...

I originally found it from this review, maybe that’s also interesting: www.frontiersin.org/journals/psy...

Thanks a lot! :)

This one is so interesting! Would you mind sharing the reference? :)