My solution is to think of R as a piece of equipment no different than any other piece of laboratory equipment. You need the model number and the manufacturer. In this case the "manufacturer" is a website: https://www.r-project.org/. You will also cite any shell that you used. For example if you used RStudio as a front end to R, then both RStudio and R will need a model number and web address.

Packages should also be acknowledged especially when they are not part of base R. If necessary use the URL for the documentation. In part the level of detail should be consistent with the importance of the package to your conclusions. So I might not identify which package was used to create a boxplot because this can be done in several ways with many different software programs. However, I would cite the binGroup package (and make sure that the journal does not "correct" the capitalization during editing) because there are few programs available that will estimate disease incidence from pooled samples. Likewise if all I do is ANOVA, then a simple citation of R is sufficient. If I do a complex nonlinear regression then I should cite the packages that make the analysis possible. I would also consider making the program available either in the methods section, an appendix, or supplemental material. An alternative would be to provide a flowchart for the program with enough detail that a person can evaluate if it is reasonable to assume that the program could produce the output shown.

How to cite anything depends on the style of the journal or thesis or book. A popular method is to store references in BiBTeX. The citation for R (from citation() ) in BibTeX 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 = {2018},

url = {https://www.R-project.org/},

}

Then whatever style you are using, should cite it appropriately. If it doesn't have anything for Manual, you can save a Miscellaneous.

ps. If you want to cite a package, do

citation("packagename")

and it will provide you with the BiBTeX version which is sometimes a paper, sometimes a manual. It also provides plain text suggestions for citing, but these depend on if you are using MLA, APA, etc., guidelines, so are less useful (unless it matches onto that).

e.g., citation("lme4") produces

To cite lme4 in publications use:

Douglas Bates, Martin Maechler, Ben Bolker, Steve Walker (2015).

Fitting Linear Mixed-Effects Models Using lme4. Journal of

Statistical Software, 67(1), 1-48. doi:10.18637/jss.v067.i01.

A BibTeX entry for LaTeX users is

@Article{,

title = {Fitting Linear Mixed-Effects Models Using {lme4}},

author = {Douglas Bates and Martin M{\"a}chler and Ben Bolker and Steve Walker},

journal = {Journal of Statistical Software},

year = {2015},

volume = {67},

number = {1},

pages = {1--48},

doi = {10.18637/jss.v067.i01},

}

Sometimes the BiBTeX version the authors give won't include some information, like authors' first names or DOI, so you might need to add a bit.

For example:

> citation("MASS")

To cite the MASS package in publications use:

Venables, W. N. & Ripley, B. D. (2002) Modern Applied Statistics with

S. Fourth Edition. Springer, New York. ISBN 0-387-95457-0

A BibTeX entry for LaTeX users is

@Book{,

title = {Modern Applied Statistics with S},

author = {W. N. Venables and B. D. Ripley},

publisher = {Springer},

edition = {Fourth},

address = {New York},

year = {2002},

note = {ISBN 0-387-95457-0},

url = {http://www.stats.ox.ac.uk/pub/MASS4},

}

With these you often have to add some bits. So for MASS, if you want the MASS letters to be capitalized and S to be in san serif, that line would be

title = {Modern {A}pplied {S}tatistics with \textsf{S}}, and add Bill (William) and Brian for names (which aren't used in some styles).