Eva Vivalt points me to this. I don’t know anything about it, but I am intrigued by the idea of a meta-analysis being done outside of the usual channels.
Eva Vivalt points me to this. I don’t know anything about it, but I am intrigued by the idea of a meta-analysis being done outside of the usual channels.
Interesting, thet edit a bit
Have you ever wondered whether aid programs [something] actually work[s]?
Hmm, is that not the purpose/promise of the scientific method?
So I check the title of the first talk I gave on Meta-analysis.
“Meta-Analysis: Auditing Scientific Projects and Method. University of Toronto, Department of Statistics Colloquim, 1989”
Oops, 20 years too early.
But what is interesting is the online fundraising for the project
– that could enable/empower science.
Sounds interesting,… On the cynical side–I can see Evidence based political action committees (EB-PACS?)facing off in the courts/press for years… However, my biggest worry is that while they mention random effects affecting individual studies, it seems to me (admittedly an outsider both to social studies & meta-analysis) that such social/political studies may all contain similar systematic biases and other errors that could easily compound when aggregated. I hope they will also include details or sources for people to access the individual studies that go into their analysis as well.
The systematic error issue is increasingly becoming emphasized, it’s initially being largely ignored being punctuated by M Egger’s paper entitled Spurious Certainty.
An introduction to those issues (limited to “baby” stats methods) is available in Greenland S, O’ Rourke K: Meta-Analysis. In Modern Epidemiology, 3rd ed. Edited by Rothman KJ, Greenland S, Lash T. Lippincott Williams and Wilkins; 2008.