• Develop intermediate data analysis skills in R
• Learn programming basics
• Drink reproducibility Kool-Aid
• Pick up some handy statistical techniques

• Savery 117 on Wednesdays, 3:30-5:20
• Format: lecture -> short break -> lab exercise (approximately)
• Materials: http://rebeccaferrell.github.io/CSSS508
• Office hours: TBD, will be in Padelford C-14A
• Grading:
  • Final grade: Cr/NC, 60% to get Cr
  • Homework most weeks (70% of grade), combo of reading and programming
  • Participation = peer grading the HW (30% of grade)
  • Grades and discussion forums maintained on Canvas page

Yes, peer grading…

• Write your reports better knowing others will see them
• Learn new tricks from reading others' code, “appreciate the little things”
• Randomly assigned peers, turnaround time: 1 week (due before next class)
• You'll get a rubric and are expected to leave constructive comments
• High scoring assignments will be adapted into keys (with your permission)
• Talk to me if you want more feedback

Don't be Goofus:

tried lm(y~x) but it iddn't work wat do i do

Instead, be Gallant:

y <- seq(1:10) + rnorm(10)
x <- seq(0:10)
model <- lm(y ~ x)

Running the block above gives me the following error, anyone know why?

Error in model.frame.default(formula = y ~ x, drop.unused.levels = TRUE) : 
  variable lengths differ (found for 'x')

1. Intro to R, RStudio, and R Markdown
2. Making awesome plots with ggplot2
3. Summarizing and combining data: dplyr
4. R data structures
5. Data cleaning: import/export, tidyr, dates
6. Programming and functions
7. More programming and functions
8. String/text processing
9. Taking requests for final weeks!

Do you have a source of data you'd like to see used in this class? Talk to me: rferrell@uw.edu

• Must have no restrictions on public use
• Pertinent to social science, public policy, global health
• Examples:
  • Census public use microdata sample
  • City of Seattle, King County, State of WA
  • World Health Organization, World Bank, United Nations
  • Data made public from studies

is a programming language and software environment for statistics. You will learn how to write R code in this course.

is a front-end for R. You do not have to use it to use R, but it will make your life easier:

• Organizes your code, output, and plotting windows
• Code auto-completion and syntax highlighting
• GUI for viewing data, objects in your environment
• Much easier to make R Markdown documents

Killer feature of R/RStudio is ease of making R Markdown files:

• Document analyses as you go with integrated text, code, and output
  • No rerunning, recopying, repasting
  • Easy for collaborators to understand
  • Show as little or as much of the guts as you want
• Make presentations (like this one!), webpages, reports in html, Word formats
• Works with LaTeX for PDFs, more formatting control

Open up RStudio now (CSSCR lab instructions), and choose File > New File > R Script. Let's get oriented with the interface:

• Top left: code editor pane, viewing data (tabbed)
• Bottom left: console for running code (> prompt)
• Top right: list of objects in environment, code history tab
• Bottom right: tabs for browsing files, viewing plots, managing packages, and viewing help files

You can change the layout in Preferences > Pane Layout.

There are several ways to run R code in RStudio:

1. Highlight lines in the editor window and click Run or hit Cntl-Enter (Command-Enter on a Mac) to run them all
2. Type individual lines in the console and press Enter
3. In R Markdown documents, click within a code block (“chunk”) and click on a choice in the Chunks dropdown

The console will show the lines you ran followed by any printed output. If you mess up (e.g. leave off a parenthesis), R might show a + sign prompting you to finish the command:

> (11 - 2
+

Finish the command or hit Esc to get out of this.

In the console, type 123 + 456 + 789 and hit Enter.

123 + 456 + 789

[1] 1368

Now in your blank R document in the editor, try typing the line sqrt(400) and either clicking Run or hitting Cntl-Enter.

sqrt(400)

[1] 20

sqrt is an example of a function in R.

If we didn't have a good guess as to what sqrt will do, we can type ?sqrt in the console and look at the Help panel on the right.

?sqrt

Arguments are the inputs to a function. In this case, the only argument to sqrt is a number or set of numbers x.

Let's try making an R Markdown file:

1. Choose File > New File > R Markdown…
2. Make sure HTML output is selected and click OK
3. Save the file somewhere, call it my_first_Rmd.Rmd
4. Click the Knit HTML button
5. Watch the progress in the R Markdown pane, then gaze upon your result! You may also open up the file in your computer's browser if you so desire.

The header of an .Rmd file is a YAML block (Yet Another Markup Language) code, and everything else is part of the main document.

---
title: "Untitled"
author: "Rebecca Ferrell"
date: "March 30, 2016"
output: html_document
---

To mess with global formatting, you can modify the header:

output:
  html_document:
    theme: readable

Now try editing and re-knitting my_first_Rmd.Rmd to give it some pizzazz: add sections, have it number the sections, make a list, throw in the UW logo, change the theme, insert a gratuitous equation…

• Ways to modify the overall document appearance
• Ways to format parts of your document

To keep R Markdown dead-simple, it lacks some features you might occasionally want to use. Your options for fancier documents are:

• Use HTML/CSS for custom formatting
• Use LaTeX and .Rnw files instead of .Rmd

For day-to-day use, plain vanilla R Markdown does the job.

Lines of R code make up what are called chunks. Code is sandwiched between sets of three backticks and {r}.

```{r}
summary(cars)
```

summary(cars)

     speed           dist       
 Min.   : 4.0   Min.   :  2.00  
 1st Qu.:12.0   1st Qu.: 26.00  
 Median :15.0   Median : 36.00  
 Mean   :15.4   Mean   : 42.98  
 3rd Qu.:19.0   3rd Qu.: 56.00  
 Max.   :25.0   Max.   :120.00  

Chunks have options that control what happens with its code, such as:

• echo=FALSE: keeps R code from being shown in the document
• results='hide': hides R's (non-plot) output from the document
• warning=FALSE, message=FALSE: suppresses R warnings or messages from the document, e.g. about versions
• eval=FALSE: show R code in the document without running it
• cache=TRUE: saves results of running that chunk so if it takes a while, you won't have to re-run it each time you re-knit the document
• fig.height=5, fig.width=5: modify the dimensions of any plots that are generated in the chunk (units are in inches)

Try adding or changing the chunk options (separated by commas) for the two chunks in my_first_Rmd.Rmd and re-knitting to check what happens.

You can also name your chunks by putting something after the r before the chunk options.

```{r summarize_cars, echo=FALSE}
summary(cars)
```

After you name your chunks, look what happens in the dropdown on the bottom left of your editor pane.

Sometimes we want to insert a value directly into our text. We do that using code in single backticks starting off with r.

Four score and seven years ago is the same as `r 4*20 + 7` years.

Four score and seven years ago is the same as 87 years.

Maybe we've saved a variable in a chunk we want to reference in the text:

x <- sqrt(77) # <- is how we assign variables

The value of `x` rounded to the nearest two decimals is `r round(x, 2)`.

The value of x rounded to the nearest two decimals is 8.77.

Having R dump values directly into your document protects you from silly mistakes:

• Never wonder “how did I come up with this quantity?” ever again — just look at your formula in your .Rmd file
• Consistency, no “find/replace” mishaps: reference a variable in-line throughout your document without manually updating if the calculation changes (e.g. reporting sample sizes)
• You are more likely to make a typo in a “hard-coded” number than you are to write R code that somehow runs but gives you the wrong thing

In your YAML header, make the date come from R's Sys.time() function by changing:

date: "March 30, 2016"

to:

date: " `r Sys.time()` "

Fancier option: use format(Sys.Date(), format="%B %d, %Y") instead to take today's date and make it read nicely.

In the sample R Markdown document you are working on, cars is a built-in data frame. You can look at it in a couple of ways:

View(cars): pops up a spreadsheet-looking viewer (“interactive” use only, don't put in R Markdown document!) or

head(cars) # prints first 6 rows, see tail() too

  speed dist
1     4    2
2     4   10
3     7    4
4     7   22
5     8   16
6     9   10

str(cars) # str[ucture]

'data.frame':   50 obs. of  2 variables:
 $ speed: num  4 4 7 7 8 9 10 10 10 11 ...
 $ dist : num  2 10 4 22 16 10 18 26 34 17 ...

summary(cars)

     speed           dist       
 Min.   : 4.0   Min.   :  2.00  
 1st Qu.:12.0   1st Qu.: 26.00  
 Median :15.0   Median : 36.00  
 Mean   :15.4   Mean   : 42.98  
 3rd Qu.:19.0   3rd Qu.: 56.00  
 Max.   :25.0   Max.   :120.00  

If you put an assignment <- in parentheses (), R will print the output of the assignment out for you in your document. Otherwise, it won't show the value.

( dist_mean <- mean(cars$dist) )

[1] 42.98

( speed_mean <- mean(cars$speed) )

[1] 15.4

Let's switch gears to the swiss data frame built in to R.

First, use ?swiss to see what things mean.

Add chunks to your R Markdown document inspecting swiss, defining variables, doing some exploratory plots using hist or plot. You can experiment with colors and shapes.

( f_ae_reg <- lm(Fertility ~ Agriculture + Education, data = swiss) )

Call:
lm(formula = Fertility ~ Agriculture + Education, data = swiss)

Coefficients:
(Intercept)  Agriculture    Education  
   84.08005     -0.06648     -0.96276  

Try running summary(f_ae_reg).

Let's make a table that looks a little less code-y in the output. To do this, we'll want to install a package called pander.

In the console: install.packages("pander", repos = "http://cran.us.r-project.org").

• You can leave off the repos argument if you configure your default repository settings: Preferences > Packages > CRAN mirror. I suggest the Hutch.

• Once you install a package, you don't need to re-install it until you update R.

library(pander) # loads pander, do once in your session
pander(summary(swiss), style = "rmarkdown", split.tables = 120)

pander(head(swiss, 5), style = "rmarkdown", split.tables = 120)

Write up a .Rmd file showing some exploratory analyses of the Swiss fertility data. Upload both the .Rmd file and the .html file to Canvas.

Mix in-line R calculations, tables, R output, and plots with text describing the relationships you see. You must use in-line R calculations/references at least once (e.g. with functions like nrow, mean, sd, cor, median, min, quantile) and may not hard-code any numbers referenced in your text.

Your document should be pleasant for a peer to look at, with some organization using sections or lists, and all plots labeled clearly. Use chunk options echo and results to limit the code/output you show in the .html. Discussion of specific values should be summarized in sentences in your text — not as printed code and output — and rounded so as not to be absurdly precise.

• 0: Didn't turn anything in
• 1: Turned in but low effort, ignoring many directions
• 2: Decent effort, followed directions with some minor issues
• 3: Nailed it