Functionally, Rmarkdown provides a solution to the problem of separating our work (i.e., our R code) from the reporting of our results.

More specifically, it is a toolkit which allow us to tightly integrate any code used to solve a problem, the results of running that code, and the human-readable explanation of the methods and the solution.

By using the Rmarkdown toolkit, we can blend human language, computer language, and computer output in a single, executable file.

The final product of our work becomes a typeset PDF or HTML document explaining all of our work and results, rather than an arcane R script and console output with random copy-pasta to a word document.

If you're running a newer version* of Rstudio, then you should have Rmarkdown installed already. Other R package dependencies will be automatically pulled in as needed when you start using Rmarkdown.

In non-Rstudio environments, you must manually install the Rmarkdown package with install.packages("rmarkdown") . You'll also need to install pandoc, which is a standalone executable that converts document between formats.

*newer means newer than version 98.932 (released June 2014)

Since we can't exhaustively learn Rmarkdown in one class, you'll need to know where to look for help and refreshers.

• Be sure to look at all the pages at http://rmarkdown.rstudio.com/
• Under the "Help" menu in Rstudio, you can open a built-in markdown quick reference. This opens in the same pane as the help files for R functions.
• Thoroughly explore all the chunk options at http://yihui.name/knitr/options/ (more on chunk options later in the class).
• For a longer, more in depth intro checkout this tutorial by Rob Knell

We've created our first dynamic document, huzzah!

But we're going to learn even more than how to press some buttons, so grab this .Rmd file , open it up and let's get to making even better documents.

We'll work through the sections of this document together, and as we go through it, try to connect the dots between the format of the input document, and the resulting elements in the output document.

The section at the very top, separated from the rest of the document by the 3 dashes ---, is called the YAML header (YAML stands for YAML Aint Markup Language).

The YAML header stores info about your document in your document (i.e. metadata), and tells the interpreter about how to render the final document.

For example, the output: html_document option tells interpreter to construct an HTML document (as opposed to a PDF)

• For inline equations, enclose your equations in single $ like so: $equation$
• To give your equation its own centered "paragraph", use double $ with spaces, like so: $$ equation $$

## Warning: package 'knitr' was built under R version 3.4.4

1. Create a new .Rmd file, set up to compile to HTML
2. Use an inline R expression to make the date stamp update each time the document is compiled. Do this by enclosing the function Sys.Date() with backticks in the "date" YAML field . Be sure to also keep the enclosing double quotes!
3. Make a center-aligned boxplot of the distribution of count values for each value of spray in the InsectSprays data frame.
   • Hint: Remember you can use the ~ operator in the call to boxplot()
4. Below the figure, write a one sentence description of the plot (e.g. a short figure caption).
5. Add a proper table summarizing the InsectSprays data frame in your document using kable. Keep the lines of R code loading the knitr package and generating the table out of the final document.

Your resulting HTML document can be extensively customized by including styling information via CSS (Cascading Style Sheets).

Teaching you HTML + CSS is definitely outside the scope of this class, but there are approximately 10²⁰ tutorials and classes available online that you can use.

When you have written a style sheet you want applied to your HTML document, you can include it via the css YAML header like so:

css: "path/to/your/stylesheet.css"

You can take an existing Rmd file, change the output type in the YAML header to ioslides_presentaion, and then add slide separators everywhere you want to divide content up into discrete slides.

Or if you want to start a new Rmd file set up to compile to a slide show, go to File –> New File –> R Markdown, select "Presentation" from the menu on the left, and pick your format (I prefer ioslides)

It also easy to make your final document a PDF. Creating PDFs requires a \(\LaTeX\) installation. MikTeX is preferred for Window's, while MacTex is preferred for Mac.

Once \(\LaTeX\) is installed, just change the output YAML header value to pdf_document to set PDF as final format.

Because they are typeset with \(\LaTeX\), you have more powerful formatting options when compiling to PDF. But, having a paginated final product can mean more typsetting headaches, especially with table and figure placement, relative to HTML.