Abstract
Though Rmarkdown is a powerful tool for integrating text with code for analyses in a single source document exportable to a variety of output formats, until now there has been no simple way to integrate the data behind analyses into Rmarkdown source documents. The knitrdata package makes it possible for arbitrary text and binary data to be integrated directly into Rmarkdown source documents via implementation of a new data chunk type. The package includes command-line and graphical tools that facilitate creating and inserting data chunks into Rmarkdown documents, and the treatment of data chunks is highly configurable via chunk options. These tools allow one to easily create fully standalone Rmarkdown source documents integrating data, ancillary formatting files, analysis code and text in a single file. Used properly, the package can facilitate open distribution of source documents that demonstrate computational reproducibility of scientific results.