Getting Help with R

A good place to start is Rstudio’s guide for getting help with R

Asking for Help

When asking for help with R it is important to at least

  1. Include the exact code that you ran
  2. The error message that you got

Without the code and the error message it can be hard to impossible to diagnose the problem and provide a solution. Questions such as “This isn’t working” or even “I’m getting this error message” aren’t useful without the code and error message.

Even better is to provide a minimal reproducible answer. This is the smallest amount of code that someone else can run, that produces the same error that you are getting. This often means trying to reproduce your error running the same code on a different built-in data set and seeing if you still get the same error. See more instructions here. A benefit of this approach, is that often by trying to reproduce with minimal code you will be able to debug the code itself.

Finally, include the results of sessionInfo(). This provides information about the exact versions of R, loaded packages, and operating system you are using, which can be useful in debugging problems.

sessionInfo()
## R version 3.5.0 (2018-04-23)
## Platform: x86_64-apple-darwin15.6.0 (64-bit)
## Running under: macOS High Sierra 10.13.4
## 
## Matrix products: default
## BLAS: /Library/Frameworks/R.framework/Versions/3.5/Resources/lib/libRblas.0.dylib
## LAPACK: /Library/Frameworks/R.framework/Versions/3.5/Resources/lib/libRlapack.dylib
## 
## locale:
## [1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8
## 
## attached base packages:
## [1] stats     graphics  grDevices utils     datasets  methods   base     
## 
## loaded via a namespace (and not attached):
##  [1] compiler_3.5.0  backports_1.1.2 bookdown_0.7    magrittr_1.5   
##  [5] rprojroot_1.3-2 tools_3.5.0     htmltools_0.3.6 yaml_2.1.18    
##  [9] Rcpp_0.12.16    stringi_1.1.7   rmarkdown_1.9   blogdown_0.6   
## [13] knitr_1.20      stringr_1.3.0   digest_0.6.15   xfun_0.1       
## [17] evaluate_0.10.1

For further discussion on how to ask good questions:

Getting Help from Within R and RStudio

Help tab: The first place to look is to use the search box in the Help tab of RStudio.

Autocompletion: If you start typing in either the console or editor in RStudio and hit [TAB], the RStudio will show a list of completions as well as a short help for each.

There are several R functions to get help. You can use either ? or help. For example to find help for a function quantile, both

?quantile

and

help("quantile")

will display the help page for quantile. To view this documentation online in a searchable form visit Rdocumentation.org.

To find the help pages available for a package, use help with the package argument. For example,

help(package = "tidyr")

The help for R functions is only useful if you know what function you need want to know either what it is doing, or what the arguments are. The help files in R are not useful for understanding what a package does or how to use it. However, vignettes for R packages are often overviews of the functionality of packages that serve as tutorials or introductions to using the package. To get a list of all vignettes of the packages your have installed, run

browseVignettes()

To get the list of vignettes for a specific package, run

browseVignettes("dplyr")

You can also find vignettes for packages you have not installed on the packages’ CRAN page https://cran.r-project.org/web/packages/tidyr/index.html. There is also a vignette function, but you need the exact name of the vignette, which you generally do not know.

You can search for foo in help manuals and archived mailing lists

RSiteSearch("ggplot2")

Want to find out what data sets are included in the R packages that you have installed?

data()

To find data sets included in a specific package, use data with the package argument, For example,

data(package = "ggplot2")

Note that search() does not do what you expect it to do (unless you expect it to return the search path in which R looks for names of functions and objects).

Help on the Internet

Rdocumentation.org searches the documentation of all R packages on CRAN.

There are several documentation sites that are of particular use:

Google has gotten better with R searches, but R is a particularly unfriendly search term. However, if you are searching for “R” with other words in context, you are likely to have some success. An alternative is to use rseek.org instead. It is a Google search restricted to a curated set of sites related to R.

Many R packages have an article in either the Journal of Statistical Software or The R Journal These articles are similar to package does.

Stack Overflow is a question and answer website for computer programming questions. Search for questions about R using the “r” tag. You can do this by adding [r] to your search query. Other tags that may be useful are [ggplot2], [plyr]. However, generally you should not need to ask questions for this course. Someone has likely already asked your question before.

If you have a question about “statistics” or “data analysis” rather than “R”, then Cross Validated is a Q-and-A site like Stack Overflow, but for statistics and machine learning.

The traditional source of R help was the R-help mailing list. It is still a useful resource, though much of the traffic has moved over to StackOverflow.

R Cheatsheets

RStudio has man very good cheatsheets. The most relevant to this course are:

The Short R Reference Card is the classic cheatsheet. However, some of the data manipulation and plotting functions are not those we use in this course, but it does cover much of the statistical functions.

Books

Two references organized by task are The Cookbook for R and Quick-R.

Blogs and Twitter

R-bloggers aggregates R-related blogging.

Task Views

Do you want to know what packages exist for a given topic? CRAN Task Views provide subject matter groupings of R packages. Some potentially useful task-views to social scientists are: