JASA demands code and data be supplied as a condition of publication

The Journal of the American Statistical Association (“JASA”) has announced in this month’s Amstat News that effective 1st September 2016 “… will require code and data as a minimum standard for reproducibility of statistical scientific research.” Trends were heading this way, but it is excellent to see a major journal insisting upon it as standard practice.

There appear to be some weasel words allowing publications having “proprietary data” to move forward, insisting upon code nevertheless. I can only imagine that publications opting for that path will be seen as less established, solid, or compelling.

About hypergeometric

See http://www.linkedin.com/in/deepdevelopment/ and https://667-per-cm.net/about
This entry was posted in American Association for the Advancement of Science, American Statistical Association, citizen science, engineering, ethics, evidence, new forms of scientific peer review, numerical software, planning, rationality, reasonableness, resiliency, science, statistics, stochastic algorithms, testing, the right to know. Bookmark the permalink.