Reproducibility is indeed important. From the point of view of those in
the statistics
community and in particular in the R community, the key issue is that
the data on which
a publication is based should be readily accessible so that others can
replicate and
possibly extend the analysis, and propose and present alternative analyses.
But to say that "most published scientific research is probably false"
is drivelous
nonsense. That sort of assertion is made by right-wing ideologues who
are afflicted
with envy of the academic community and slang off at it to alleviate
their own sense
of inadequacy.
cheers,
Rolf Turner
On 10/26/13 21:06, Liviu Andronic wrote:
Dear all,
I know that reproducibility is a big concern for the R community, so
it may be interesting to some of the readers on this list that The
Economist recently ran a series of articles denouncing the alarming
number of shoddy and non-reproducible published papers:
http://www.economist.com/news/leaders/21588069-scientific-research-has-changed-world-now-it-needs-change-itself-how-science-goes-wrong
http://www.economist.com/news/briefing/21588057-scientists-think-science-self-correcting-alarming-degree-it-not-trouble
They even went as far as stating that "most published scientific
research is probably false":
http://www.economist.com/blogs/graphicdetail/2013/10/daily-chart-2
Anyways, food for thought for the weekend.
______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
