0

Given the "replication crisis" and numerous examples of outright fraud in academia in the past decades, why do reputable journals still accept papers where the authors don't share 100% of their data and code along with the paper? This would make replication attempts a lot easier and somewhat deter fraudulent researchers.

Also see: Are there any proposals to make sharing code and data with publications a requirement?

Improve this question
9
  • 3
    Related / possible duplicate: Why are papers without code but with results accepted? (also this one)
    – cag51 ♦
    Commented Apr 18 at 13:50
  • @cag51 it's been a decade since, I think we need a fresh question
    – JonathanReez
    Commented Apr 18 at 13:52
  • 2
    @JonathanReez Looking it over, the reasons haven't changed, so why do we need a fresh question?
    – Bryan Krause ♦
    Commented Apr 18 at 14:07
  • 1
    If you are doing research on people, there are competing interests with replicability and privacy of the respondents. So there are legitimate cases where the data cannot be made public. The data can still be made available to people wanting to replicate it, but depending on how sensitive the data is that access could be more or less convenient.
    – Maarten Buis
    Commented Apr 18 at 15:05
  • 1
    @JonathanReez You are free to doubt, but it would be even better if you did your research. An easy, and possibly pleasant, way to do that is to just invite a member of the IRB in your institution to a coffee and discuss that with her/him/them.
    – Maarten Buis
    Commented Apr 19 at 8:13

0

Reset to default

Browse other questions tagged .