Citing software in research papers

January 10th, 2013 | Categories: math software, programming, R, Scientific Software | Tags:

R has a citation() command that recommends how to cite the use of R in your publications, information that is also included in R’s Frequently Asked Questions document.

To cite R in publications use:

  R Core Team (2012). R: A language and environment for
  statistical computing. R Foundation for Statistical
  Computing, Vienna, Austria. ISBN 3-900051-07-0, URL
  http://www.R-project.org/.

A BibTeX entry for LaTeX users is

  @Manual{,
    title = {R: A Language and Environment for Statistical Computing},
    author = {{R Core Team}},
    organization = {R Foundation for Statistical Computing},
    address = {Vienna, Austria},
    year = {2012},
    note = {{ISBN} 3-900051-07-0},
    url = {http://www.R-project.org/},
  }

We have invested a lot of time and effort in creating R, please
cite it when using it for data analysis. See also
‘citation("pkgname")’ for citing R packages

This led me to wonder how often people cite the software they use.  For example, if you publish the results of a simulation written in MATLAB do you cite MATLAB in any way?  How about if you used Origin or Excel to produce a curve fit, would you cite that?  Would you cite your plotting software, numerical libraries or even compiler?

The software sustainability institute (SSI), of which I recently became a fellow, has guidelines on how to cite software.

  1. ssam
    January 10th, 2013 at 18:40
    Reply | Quote | #1

    I have seen codes (freeware and commercial) that demand a citation as part of their licence agreement.

  2. JuanPi
    January 10th, 2013 at 19:21
    Reply | Quote | #2

    In GNU Octave we set this page to account and estimulate citation.

    http://wiki.octave.org/Publications_using_Octave

    Providing easy access to a well formatted citation is very important, lets see if we can follow R’s idea.

  3. AlBahtaar
    January 10th, 2013 at 19:30
    Reply | Quote | #3

    Did not reference compilers and plotting utilities in articles, but I should have thought about that for my PhD dissertation.
    I reckon that re. the first ones and sci. softs, mentionning the versions is important.

  4. Umberto
    January 10th, 2013 at 21:05
    Reply | Quote | #4

    Yes I always cite the software I use. Any serious researcher knows (or should know) that it is essential to allow reproducibility of e.g. simulation studies. For statisticians dealing with modelling or analysis of medical data it’s absolutely relevant to keep track (and cite) the software used, as well as its release version, as a bug in a procedure leading to incorrect results (and potentially serious consequences!) can at least be “justified” by the use of buggy software.

  5. January 10th, 2013 at 21:19
    Reply | Quote | #5

    Matplotlib recently added a ‘citation’ page to their website to encourage referencing when use of matplotlib leads to a scientific publication. See http://matplotlib.org/citing.html

  6. oversky
    January 11th, 2013 at 02:55
    Reply | Quote | #6

    I agree with Umberto. A good research paper should have enough information to be reproduced by others. However, many publications have page limit, and the author has to decide what is the most important to keep.

  7. Steve L
    January 11th, 2013 at 16:39
    Reply | Quote | #7

    If you’re using a submission from the MATLAB Central File Exchange, one of the members of the MATLAB Central team posted about how to cite it:

    http://blogs.mathworks.com/community/2010/12/13/citing-file-exchange-submissions/

    There have been some discussions on how to cite the documentation on MATLAB Answers. One such discussion with a link to an older discussion is:

    http://www.mathworks.com/matlabcentral/answers/43629-citing-a-matlab-document

    If you’re looking for an official answer on how to cite the documentation, I recommend contacting Customer Service. You can do so using the Contact Us link at the top of http://www.mathworks.com. If you’re using a sufficiently recent version of MATLAB, you can even contact CS from within MATLAB itself!

    http://blogs.mathworks.com/community/2011/06/20/requesting-technical-support-from-within-matlab/

    http://www.mathworks.com/help/matlab/matlab_env/contact-technical-support.html

  8. January 11th, 2013 at 18:52
    Reply | Quote | #8

    Thanks Steve. I contacted Customer Service and they gave me an answer. I asked if I could include it on this page and was told that they were checking with superiors which is fair enough. So, hopefully watch this space.

  9. January 13th, 2013 at 22:38
    Reply | Quote | #9

    I also cite all the software I use, even commercial tools. I don’t cite compilers.

  10. Martin Cohen
    January 14th, 2013 at 04:15

    I find that when I look at the technical reports I wrote many (30+) years ago, I really wish that I had included listings of the programs I had written and all the data produced , even if that would have doubled (or more – sometimes a factor of ten) the size of the report.

    In recent years, when the reports were stored digitally, I included the listings as appendices and, if a hard copy was required, I did not print the appendices, just the main body.

  11. January 17th, 2013 at 01:58

    Software should certainly be cited if it has anything to do with computed results. Seems to me that there are two issues: 1) Reproducibility and 2 ) Credit. Making things reproducible should be a requirement in scientific publications. And credit should be given where it’s due.

    I get a bit irked when people don’t cite our stuff, e.g. PLS_Toolbox. I’ve instances where I can tell published results were produced with our software and the authors write as if they wrote the software themselves.