Intended for healthcare professionals

Rapid response to:


StatsDirect—Statistical Software for Medical Research in the 21st Century

BMJ 2000; 321 doi: (Published 16 December 2000) Cite this as: BMJ 2000;321:1536

Rapid Response:


The 'four star' review of Statsdirect was poor. By the
end of the article I had not been enlightened as to the
feature set of the software or how it might compare with
traditional commercial statistical packages,
shareware/freeware programs or even the proliferation
of web based statistical packages (e.g. the Interactive
Statistical Calculation pages at
Webstat 2.0 at
). Does the
software employ a GUI or is it command line? How
easy is it to use? Can it import data from external
programs? Is it fast and accurate? Does it include
graphing funtionality suitable for presentation or

Indeed I was none the wiser as to whether I could even
run the software on my computer at all as no mimimum
specifications or operating system requirements were
mentioned. Suspicious, I went on line and checked the
StatsDirect website, there I was enticed to download
the software for a trial but still no mention of the
program's requirements anywhere. Luckily a link to a
more thorough JAMA review (
) confirmed what I had suspected through this
conspiracy of silence and prejudiced assumption -
Windows only. So, no joy for this MacUser - null
points/no stars in my humble opinion, not acceptable in
the age of Java and cross-platform development tools.

Please, if you are going to review something do it
thoroughly or not at all so that us poor consumers can
be properly informed and see some basis to the rather
arbitrary 'star ratings' attributed.

Kind Regards

Dr Giles Roditi

Competing interests: No competing interests

19 December 2000
Giles Roditi
Consultant Radiologist
Glasgow Royal Infirmary