Searching for the Open Access corpus is done with the ScienceOpen search. This is the URL used on March 30, 2017 to search for meta-analyses that mention something about publication bias.
It says 422 results, but when browsing through it on the page gives only 370 results (at least that's what ScienceOpen shows...). One article was left out, resulting in 369 articles (this article didn't have a DOI).
- Extend this to flowcharts to analyze CONSORT or PRISMA flows
-
Upon manual inspection (20170704) of 10.5812_ircmj.40061 I noticed that points are extracted twice (all of them; 60 extracted, 30 present).
-
Extracted data contain x coordinates in column 1, y coordinates in column 2.
-
Scale in 10.2147_BCTT.S94617 not linear so data incorrect (but relative scale is correct, so can be transformed)