Dynamic documents in Rreproducible research with R Markdown2020-08-221 / 51

Artwork by @allison_horst

2 / 51

R Markdown

3 / 51

R Markdown

Authoring framework: code and text in same document

3 / 51

R Markdown

Authoring framework: code and text in same document

Reproducible: re-run your analysis

3 / 51

R Markdown

Authoring framework: code and text in same document

Reproducible: re-run your analysis

Flexible: Output to different formats easily

3 / 51

knitting

4 / 51

Your turn 1Create a new R Markdown file. Go to File > New File > R Markdown. Press OK. Save the file and press the "Knit" button above.5 / 51

6 / 51

R MarkdownProseCodeMetadata7 / 51

R MarkdownProse = MarkdownCodeMetadata8 / 51

Basic Markdown Syntax

*italic*   **bold**
_italic_   __bold__

9 / 51

Basic Markdown Syntax

# Header 1
## Header 2
### Header 3

10 / 51

Basic Markdown Syntax

* Item 1
* Item 2
    + Item 2a
    + Item 2b
1. Item 1
2. Item 2

11 / 51

Basic Markdown Syntax

http://example.com
[linked phrase](http://example.com)

12 / 51

Basic Markdown Syntax

![](http://example.com/logo.png)
![optional caption text](figures/img.png)

13 / 51

Basic Markdown Syntax

$equation$
$$ equation $$

14 / 51

Basic Markdown Syntax

superscript^2^
~~strikethrough~~

15 / 51

Your turn 2

Do the ten-twenty minute tutorial on markdown at https://commonmark.org/help/tutorial. Let us know if you need help!

16 / 51

Your turn 3

Use Markdown syntax to stylize the text from the Gapminder website below. Experiment with bolding, italicizing, making lists, etc.

17 / 51

R MarkdownProseCode = R code chunksMetadata18 / 51

Code chunks

19 / 51

Code chunks

20 / 51

Code chunks

21 / 51

Chunk options

Option
Effect


include = FALSE
run the code but don't print it or results

eval = FALSE
don't evaluate the code

echo = FALSE
run the code and output but don't print code

message = FALSE
don't print messages (e.g. from a function)

warning = FALSE
don't print warnings

fig.cap = "Figure 1
caption output plot with "Figure 1"

22 / 51

Chunk options

Option	Effect
`include = FALSE`	run the code but don't print it or results
`eval = FALSE`	don't evaluate the code
`echo = FALSE`	run the code and output but don't print code
`message = FALSE`	don't print messages (e.g. from a function)
`warning = FALSE`	don't print warnings
`fig.cap = "Figure 1`	caption output plot with "Figure 1"

See the knitr web page

22 / 51

Engines52! Including Python, Julia, C++, SQL, SAS, and Stata23 / 51

Insert code chunks with `cmd/ctrl + alt/option + I`

24 / 51

Edit code chunk options

25 / 51

Your turn 4 (open `exercises.Rmd`)

Create a code chunk. You can type it in manually, use the keyboard short-cut (Cmd/Ctrl + Option/Alt + I), or use the "Insert" button above. Put the following code in it:

gapminder %>% 
  slice(1:5) %>% 
  knitr::kable()

Knit the document

26 / 51

Your turn 5Add echo = FALSE to the code chunk above and re-knitRemove echo = FALSE from the code chunk and move it to knitr::opts_chunk$set() in the setup code chunk. Re-knit. What's different about this?27 / 51

Your turn 5Add echo = FALSE to the code chunk above and re-knitRemove echo = FALSE from the code chunk and move it to knitr::opts_chunk$set() in the setup code chunk. Re-knit. What's different about this?Make sure to remove knitr::opts_chunk$set(echo = FALSE)27 / 51

Inline Code

28 / 51

Inline Code

29 / 51

Your turn 6Remove eval = FALSE so that R Markdown evaluates the code.Use summarize() and n_distinct() to get the the number of unique years in gapminder and save the results as n_years.Use inline code to describe the data set in the text below the code chunk and re-knit.30 / 51

R MarkdownProseCodeMetadata = YAML31 / 51

YAML Metadata

---
author: Malcolm Barrett
title: Quarterly Report
output: 
  html_document: default
  pdf_document:
    toc: true
---

32 / 51

33 / 51

34 / 51

35 / 51

36 / 51

37 / 51

Output formats

Function
Outputs


html_document()
HTML

pdf_document()
PDF

word_document()
Word .docx

odt_document()
.odt

rtf_document()
.rtf

md_document()
Markdown

slidy_presentation()
Slidy Slides (HTML)

beamer_presentation()
Beamer Slides (PDF)

ioslides_presentation()
ioslides (HTML)

powerpoint_presentation()
Powerpoint Slides

38 / 51

Function	Outputs
`html_document()`	HTML
`pdf_document()`	PDF
`word_document()`	Word .docx
`odt_document()`	.odt
`rtf_document()`	.rtf
`md_document()`	Markdown
`slidy_presentation()`	Slidy Slides (HTML)
`beamer_presentation()`	Beamer Slides (PDF)
`ioslides_presentation()`	ioslides (HTML)
`powerpoint_presentation()`	Powerpoint Slides

Your turn 7Set figure chunk options such as dpi, fig.width, and fig.height.  Run knitr::opts_chunk$get() in the console to see the defaults.Change the YAML header above from output: html_document to another output type like pdf_document or word_document.Add your name to the YAML header using author: Your Name.39 / 51

ymlthis

check out the ymlthis package for tools and documentation for working with YAML

https://r-lib.github.io/ymlthis/

40 / 51

Parameters

---
params: 
  param1: x
  param2: y
  data: df
---

41 / 51

Parameters

---
params: 
  param1: x
  param2: y
  data: df
---

params$param1
params$param2
params$data

41 / 51

Your turn 8

Change the `params` option in the YAML header to use a different continent. Re-knit

gapminder %>% 
  filter(continent == params$continent) %>% 
  ggplot(aes(x = year, y = lifeExp, group = country, color = country)) +
  geom_line(lwd = 1, show.legend = FALSE) + 
  scale_color_manual(values = country_colors) +
  theme_minimal(14) + 
  theme(strip.text = element_text(size = rel(1.1))) +
  ggtitle(paste("Continent:", params$continent))

42 / 51

Bibliographies and citations43 / 51

Bibliographies and citationsBibliography files: .bib, End Note, others44 / 51

Bibliographies and citationsBibliography files: .bib, End Note, othersCitation styles: .csl45 / 51

Bibliographies and citationsBibliography files: .bib, End Note, othersCitation styles: .csl[@citation-label]46 / 51

Including bibliography files in YAML

---
bibliography: file.bib
csl: file.csl
---

47 / 51

Your turn 9Cite the Causal Inference book in text below in the format [@citation-label]. The label for the citation is hernan_causal_2019Add the American Journal of Epidemiology CSL to the YAML using csl: aje.csl48 / 51

Check out the citr package for easy citation insertion and .bib management49 / 51

Make cool stuff in R Markdown!bookdownblogdownthese slides!50 / 51

Resources

R Markdown: A comprehensive but friendly introduction to R Markdown and friends. Free online.

R for Data Science: A comprehensive but friendly introduction to the tidyverse. Free online.

R Markdown for Scientists: R Markdown for Scientists workshop material.

51 / 51

Help

Keyboard shortcuts

↑, ←, Pg Up, k

Go to previous slide

↓, →, Pg Dn, Space, j

Go to next slide

Home

Go to first slide

End

Go to last slide

Number + Return

Go to specific slide

b / m / f

Toggle blackout / mirrored / fullscreen mode

Clone slideshow

Toggle presenter mode

Restart the presentation timer

?, h

Toggle this help

Dynamic documents in R

reproducible research with R Markdown

2020-08-22

R Markdown

R Markdown

Authoring framework: code and text in same document

R Markdown

Authoring framework: code and text in same document

Reproducible: re-run your analysis

R Markdown

Authoring framework: code and text in same document

Reproducible: re-run your analysis

Flexible: Output to different formats easily

knitting

Your turn 1

Create a new R Markdown file. Go to File > New File > R Markdown. Press OK. Save the file and press the "Knit" button above.

R Markdown

Prose

Code

Metadata

R Markdown

Prose = Markdown

Code

Metadata

Basic Markdown Syntax

Basic Markdown Syntax

Basic Markdown Syntax

Basic Markdown Syntax

Basic Markdown Syntax

Basic Markdown Syntax

Basic Markdown Syntax

Your turn 2

Do the ten-twenty minute tutorial on markdown at https://commonmark.org/help/tutorial. Let us know if you need help!

Your turn 3

Use Markdown syntax to stylize the text from the Gapminder website below. Experiment with bolding, italicizing, making lists, etc.

R Markdown

Prose

Code = R code chunks

Metadata

Code chunks

Code chunks

Code chunks

Chunk options

Chunk options

See the knitr web page

Engines

52! Including Python, Julia, C++, SQL, SAS, and Stata

Insert code chunks with cmd/ctrl + alt/option + I

Edit code chunk options

Your turn 4 (open exercises.Rmd)

Create a code chunk. You can type it in manually, use the keyboard short-cut (Cmd/Ctrl + Option/Alt + I), or use the "Insert" button above. Put the following code in it:

Knit the document

Your turn 5

Add echo = FALSE to the code chunk above and re-knit

Remove echo = FALSE from the code chunk and move it to knitr::opts_chunk$set() in the setup code chunk. Re-knit. What's different about this?

Your turn 5

Add echo = FALSE to the code chunk above and re-knit

Remove echo = FALSE from the code chunk and move it to knitr::opts_chunk$set() in the setup code chunk. Re-knit. What's different about this?

Make sure to remove knitr::opts_chunk$set(echo = FALSE)

Inline Code

Inline Code

Your turn 6

Remove eval = FALSE so that R Markdown evaluates the code.

Use summarize() and n_distinct() to get the the number of unique years in gapminder and save the results as n_years.

Use inline code to describe the data set in the text below the code chunk and re-knit.

R Markdown

Prose

Code

Metadata = YAML

YAML Metadata

Output formats

Your turn 7

Set figure chunk options such as dpi, fig.width, and fig.height. Run knitr::opts_chunk$get() in the console to see the defaults.

Change the YAML header above from output: html_document to another output type like pdf_document or word_document.

Add your name to the YAML header using author: Your Name.

ymlthis

check out the ymlthis package for tools and documentation for working with YAML

https://r-lib.github.io/ymlthis/

Parameters

Parameters

Insert code chunks with `cmd/ctrl + alt/option + I`

Your turn 4 (open `exercises.Rmd`)

Add `echo = FALSE` to the code chunk above and re-knit

Remove `echo = FALSE` from the code chunk and move it to `knitr::opts_chunk$set()` in the `setup` code chunk. Re-knit. What's different about this?

Add `echo = FALSE` to the code chunk above and re-knit

Remove `echo = FALSE` from the code chunk and move it to `knitr::opts_chunk$set()` in the `setup` code chunk. Re-knit. What's different about this?

Make sure to remove `knitr::opts_chunk$set(echo = FALSE)`

Remove `eval = FALSE` so that R Markdown evaluates the code.

Use `summarize()` and `n_distinct()` to get the the number of unique years in gapminder and save the results as `n_years`.

Set figure chunk options such as `dpi`, `fig.width`, and `fig.height`. Run `knitr::opts_chunk$get()` in the console to see the defaults.

Change the YAML header above from `output: html_document` to another output type like `pdf_document` or `word_document`.

Add your name to the YAML header using `author: Your Name`.

Change the `params` option in the YAML header to use a different continent. Re-knit

Bibliography files: `.bib`, End Note, others

Bibliography files: `.bib`, End Note, others

Citation styles: `.csl`

Bibliography files: `.bib`, End Note, others

Citation styles: `.csl`

`[@citation-label]`

Cite the Causal Inference book in text below in the format `[@citation-label]`. The label for the citation is `hernan_causal_2019`

Add the American Journal of Epidemiology CSL to the YAML using `csl: aje.csl`