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
Rule 1 of #depression trials: For every drug, there is at least one trial with tiny N & ridulous effect size.
Take the study below: N=30, d=1.80 (6 times higher than best meta-analytic estimates). See the very strong conclusions drawn in the abstract? >
https://www.sciencedirect.com/science/article/pii/S0022395612000386?via%3Dihub&fbclid=IwAR2i6Q-BYSb2QHh5Ym9XLvRThT3FBxj04mpLceDuRTEJFRrXZHoJj1y3c1A …pic.twitter.com/rX2rfLg14n
These super large effects suggest data/code errors or fake data. Whoever heard of a real effect that large? Prior very low for being real. P vals don't suggest p hacking.
There are plenty of clinical depression trials with such huge effects. In many cases, probably a combination of questionable research practices, failed blinding, & publication bias. E.g. 1) d=1.2 https://bmcmedicine.biomedcentral.com/articles/10.1186/s12916-018-1221-5 … 2) d=2.2https://eiko-fried.com/hyperthermia-as-depression-treatment/ …
What's does the distribution of effect sizes look like among pre reg studies? That's my prior. Last big replication report found a mean d of 0.15 among the REAL effects IIRC, and nothing remotely that large. (on phone can't check)
Same. Best meta-analytic effect size estimates are around 0.3, possibly a bit smaller after controlling for pub bias.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.
