You can add location information to your Tweets, such as your city or precise location, from the web and via third-party applications. You always have the option to delete your Tweet location history. Learn more
Yes, that's part of the point. When you view these results as "failures to show this works," rather than "evidence this does not work," policymakers keep trying, and ignore the (implictly non-existent) evidence. tl;dr - Evidence-based policy should include negative evidence.
A Bayesian might say that this is due to lack of Bayesian reasoning. The way frequentist NHST is usually presented, when one finds p > alpha, there is no conclusion to draw except "failures to show this works", whereas really one can sometimes infer "this does not work".
Agreed, a Bayesian might say that ;) Still, a frequentist can (and should) use the same data to derive a confidence interval for the effect size, which allows rejecting the hypothesis that the effect is "clinically" significant.
The more hardcore Bayesians will reject the use of conf. intervals as substitute cred. intervals, but that just shows their own odd philosophy of probability IMO, and not any mathematical error. I'm happy to interpret confidence intervals as credibility intervals (w. flat prior).
I'm trying hard not to get baited into debating this. (But it's so tempting to do so, it's convincing me to close twitter and get work done.)
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.
