Lab 03: BMI 5/625

class: center, middle, inverse, title-slide

.title[
# Lab 03: BMI 5/625
]
.subtitle[
## Working with Data
]
.author[
### Alison Hill, updates by Steven Bedrick
]

---

# Plan for today

- Refresher on Base R types and operators

- Tying it together with `dplyr`

- `ggplot` aesthetics and colors

- Lab workbook

---

# Data for today

We'll use data from [Wordbank](http://wordbank.stanford.edu)- an open source database of children's vocabulary development. The tool used to measure children's language and communicative development in this database is the [MacArthur-Bates Communicative Development Inventories (MB-CDI)](http://mb-cdi.stanford.edu). The MB-CDI is a parent-reported questionnaire.

- R package [`wordbankr`](https://cran.r-project.org/web/packages/wordbankr/index.html)
- [`wordbankr` vignette](https://cran.r-project.org/web/packages/wordbankr/vignettes/wordbankr.html)
- More about [Wordbank](http://wordbank.stanford.edu)
- More about [MB-CDI](http://mb-cdi.stanford.edu)

---
# Get the data

Use this code chunk to import my cleaned CSV file:

```r
library(readr)
sounds <- read_csv("http://bit.ly/cs631-meow")
```

---
class: inverse, middle, center

## RStudio Base R Cheatsheet

https://github.com/rstudio/cheatsheets/blob/master/base-r.pdf

---

## Know your data types

* Numeric (2 subtypes)
    - Integers (`1, 50`)
    - Double (`1.5, 50.25`, `?double`)
* Character (`"hello"`)
* Factor (`grade = "A" | grade = "B"`)
* Logical (`TRUE | FALSE`)

```r
typeof(sounds$age)
```

```
[1] "double"
```

```r
typeof(sounds$sound)
```

```
[1] "character"
```

```r
typeof(sounds$sound == "meow")
```

```
[1] "logical"
```

---
# Even better: `glimpse`

```r
glimpse(sounds)
```

```
Rows: 33
Columns: 4
$ age          <dbl> 8, 8, 8, 9, 9, 9, 10, 10, 10, 11, 11, 11, 12, 12, 12, 13,…
$ sound        <chr> "cockadoodledoo", "meow", "woof woof", "cockadoodledoo", …
$ kids_produce <dbl> 1, 0, 3, 0, 2, 2, 0, 5, 4, 0, 5, 12, 0, 12, 28, 9, 125, 2…
$ kids_respond <dbl> 35, 35, 35, 91, 93, 93, 139, 145, 143, 94, 94, 94, 141, 1…
```

---
# `sounds` (a subset)

- `age`: child age in months
- `sound`: a string describing a type of animal sound
- `kids_produce`: the number of parents who answered "yes, my child produces this animal sound" 
- `kids_respond`: the number of parents who responded to this question at all

<table>
 <thead>
  <tr>
   <th style="text-align:right;"> age </th>
   <th style="text-align:left;"> sound </th>
   <th style="text-align:right;"> kids_produce </th>
   <th style="text-align:right;"> kids_respond </th>
  </tr>
 </thead>
<tbody>
  <tr>
   <td style="text-align:right;"> 8 </td>
   <td style="text-align:left;"> cockadoodledoo </td>
   <td style="text-align:right;"> 1 </td>
   <td style="text-align:right;"> 35 </td>
  </tr>
  <tr>
   <td style="text-align:right;"> 8 </td>
   <td style="text-align:left;"> meow </td>
   <td style="text-align:right;"> 0 </td>
   <td style="text-align:right;"> 35 </td>
  </tr>
  <tr>
   <td style="text-align:right;"> 8 </td>
   <td style="text-align:left;"> woof woof </td>
   <td style="text-align:right;"> 3 </td>
   <td style="text-align:right;"> 35 </td>
  </tr>
  <tr>
   <td style="text-align:right;"> 9 </td>
   <td style="text-align:left;"> cockadoodledoo </td>
   <td style="text-align:right;"> 0 </td>
   <td style="text-align:right;"> 91 </td>
  </tr>
  <tr>
   <td style="text-align:right;"> 9 </td>
   <td style="text-align:left;"> meow </td>
   <td style="text-align:right;"> 2 </td>
   <td style="text-align:right;"> 93 </td>
  </tr>
  <tr>
   <td style="text-align:right;"> 9 </td>
   <td style="text-align:left;"> woof woof </td>
   <td style="text-align:right;"> 2 </td>
   <td style="text-align:right;"> 93 </td>
  </tr>
</tbody>
</table>

---
# Data types

---
# Lists? Vectors?

--
What is `2`, anyway?

```r
typeof(2)
```

```
[1] "double"
```

```r
length(2)
```

```
[1] 1
```

`2` is really a vector of length 1!

```r
is.vector(2)
```

```
[1] TRUE
```

---

# Lists? Vectors?

What about longer vectors?

```r
typeof(c(2,2))
```

```
[1] "double"
```

```r
length(c(2,2))
```

```
[1] 2
```

---

# Lists? Vectors?

_Atomic Vectors_ in `R` are of the same _type_ (all numbers, all logicals, etc.)

_Lists_ are vectors made up of multiple items of _different_ types:

```r
my.list <- list(c(1,2,3), c(TRUE, FALSE, TRUE))
my.list
```

```
[[1]]
[1] 1 2 3

[[2]]
[1]  TRUE FALSE  TRUE
```

---

# Lists? Vectors?

_Atomic Vectors_ in `R` are of the same _type_ (all numbers, all logicals, etc.)

_Lists_ are vectors made up of multiple items of _different_ types:

```r
my.list <- list(x=c(1,2,3), y=c(TRUE, FALSE, TRUE))
my.list
```

```
$x
[1] 1 2 3

$y
[1]  TRUE FALSE  TRUE
```

It is usually better to give names to list elements...

- Note: a data frame is just a list where each element is:
  - An atomic vector
  - Of uniform length

---
# Lists? Vectors?

Many operations in R are _vectorized_ (i.e., operate on an entire vector all in one go):

Adding a "scalar" (i.e., a vector of length 1) to a vector:

```r
some.numbers <- c(1,2,3)
some.numbers + 1
```

```
[1] 2 3 4
```

Adding two vectors of identical length:

```r
some.numbers + c(2,2,2)
```

```
[1] 3 4 5
```

---

# Lists? Vectors?

What if there is a length mis-match?

```r
some.numbers + c(8,8,8,8)
```

```
[1]  9 10 11  9
```

What's going on?

The shorter vector is _recycled_ until it matches the length of the larger vector:

```r
c(1,2,3,1) + c(8,8,8,8)
```

```
[1]  9 10 11  9
```

(Note that you may get a _warning_ from R if this happens)

This is how adding a scalar to a vector works, under the hood!

---

# Lists? Vectors?

Vectorization is very useful:

```r
y_hat <- predict(some.model)
residuals <- y_hat - my.data$y
```

Many things that would use _loops_ in other languages can be done using vectorized operations in R.

Many functions operate on vectors:

- `mean()`, `sum()`, etc.

One way to think of them : factories to convert atomic vectors of length `n` to atomic vectors of length 1!

---

# Data wrangling with `dplyr`

.pull-left[
From previous:

- `group_by`

- `summarize`

]

.pull-right[
Adding onto your arsenal of...

- `filter`

- `arrange`

- `mutate`

- `glimpse`

- `distinct`

- `count`

- `tally`

- `pull`

- `top_n`
]

---
class: middle, center, inverse

# 😈

## More on `mutate`

---
# 3 ways to `mutate`

1. <font color="#ED1941">Create a new variable with a specific value</font>

1. Create a new variable based on other variables

1. Change an existing variable

```r
sounds %>% 
  mutate(form = "WS")
```

```
# A tibble: 33 × 5
     age sound          kids_produce kids_respond form 
   <dbl> <chr>                 <dbl>        <dbl> <chr>
 1     8 cockadoodledoo            1           35 WS   
 2     8 meow                      0           35 WS   
 3     8 woof woof                 3           35 WS   
 4     9 cockadoodledoo            0           91 WS   
 5     9 meow                      2           93 WS   
 6     9 woof woof                 2           93 WS   
 7    10 cockadoodledoo            0          139 WS   
 8    10 meow                      5          145 WS   
 9    10 woof woof                 4          143 WS   
10    11 cockadoodledoo            0           94 WS   
# … with 23 more rows
```

---
# 3 ways to `mutate`

1. Create a new variable with a specific value

1. <font color="#ED1941">Create a new variable based on other variables</font>

1. Change an existing variable

```r
sounds %>% 
  mutate(prop_produce = kids_produce / kids_respond)
```

```
# A tibble: 33 × 5
     age sound          kids_produce kids_respond prop_produce
   <dbl> <chr>                 <dbl>        <dbl>        <dbl>
 1     8 cockadoodledoo            1           35       0.0286
 2     8 meow                      0           35       0     
 3     8 woof woof                 3           35       0.0857
 4     9 cockadoodledoo            0           91       0     
 5     9 meow                      2           93       0.0215
 6     9 woof woof                 2           93       0.0215
 7    10 cockadoodledoo            0          139       0     
 8    10 meow                      5          145       0.0345
 9    10 woof woof                 4          143       0.0280
10    11 cockadoodledoo            0           94       0     
# … with 23 more rows
```

---
# 3 ways to `mutate`

1. Create a new variable with a specific value

1. Create a new variable based on other variables

1. <font color="#ED1941">Change an existing variable</font>

```r
sounds %>% 
  mutate(prop_produce = prop_produce * 100)
```

```
# A tibble: 33 × 5
     age sound          kids_produce kids_respond prop_produce
   <dbl> <chr>                 <dbl>        <dbl>        <dbl>
 1     8 cockadoodledoo            1           35         2.86
 2     8 meow                      0           35         0   
 3     8 woof woof                 3           35         8.57
 4     9 cockadoodledoo            0           91         0   
 5     9 meow                      2           93         2.15
 6     9 woof woof                 2           93         2.15
 7    10 cockadoodledoo            0          139         0   
 8    10 meow                      5          145         3.45
 9    10 woof woof                 4          143         2.80
10    11 cockadoodledoo            0           94         0   
# … with 23 more rows
```

---
class: middle, center, inverse

# ⌛️

## Let's review some helpful functions for `mutate` + `summarize`

---
class: inverse, bottom, center
background-image: url("../images/peapod.png")
background-size: 25%

## Remember:

## Base R + Tidyverse

---
class: middle, center, inverse

#💡

## First:

## Arithmetic

*especially useful for* `mutate`

See:

http://r4ds.had.co.nz/transform.html#mutate-funs

---

```r
?Arithmetic
```

<table>
 <thead>
  <tr>
   <th style="text-align:left;"> Operator </th>
   <th style="text-align:left;"> Description </th>
   <th style="text-align:left;"> Usage </th>
  </tr>
 </thead>
<tbody>
  <tr>
   <td style="text-align:left;"> + </td>
   <td style="text-align:left;"> addition </td>
   <td style="text-align:left;"> x + y </td>
  </tr>
  <tr>
   <td style="text-align:left;"> - </td>
   <td style="text-align:left;"> subtraction </td>
   <td style="text-align:left;"> x - y </td>
  </tr>
  <tr>
   <td style="text-align:left;"> * </td>
   <td style="text-align:left;"> multiplication </td>
   <td style="text-align:left;"> x * y </td>
  </tr>
  <tr>
   <td style="text-align:left;"> / </td>
   <td style="text-align:left;"> division </td>
   <td style="text-align:left;"> x / y </td>
  </tr>
  <tr>
   <td style="text-align:left;"> ^ </td>
   <td style="text-align:left;"> raised to the power of </td>
   <td style="text-align:left;"> x ^ y </td>
  </tr>
  <tr>
   <td style="text-align:left;"> abs </td>
   <td style="text-align:left;"> absolute value </td>
   <td style="text-align:left;"> abs(x) </td>
  </tr>
  <tr>
   <td style="text-align:left;"> %/% </td>
   <td style="text-align:left;"> integer division </td>
   <td style="text-align:left;"> x %/% y </td>
  </tr>
  <tr>
   <td style="text-align:left;"> %% </td>
   <td style="text-align:left;"> remainder after division </td>
   <td style="text-align:left;"> x %% y </td>
  </tr>
</tbody>
</table>

```r
5 %/% 2 # 2 goes into 5 two times with...
```

```
[1] 2
```

```r
5 %% 2 # 1 left over
```

```
[1] 1
```

---
class: middle, center, inverse

#💡

## Second:

## Summaries

*especially useful for* `summarize`

*even more useful after a* `group_by`

See:

http://r4ds.had.co.nz/transform.html#summarise-funs

---

<table>
 <thead>
  <tr>
   <th style="text-align:left;"> Description </th>
   <th style="text-align:left;"> Usage </th>
  </tr>
 </thead>
<tbody>
  <tr>
   <td style="text-align:left;"> sum </td>
   <td style="text-align:left;"> sum(x) </td>
  </tr>
  <tr>
   <td style="text-align:left;"> minimum </td>
   <td style="text-align:left;"> min(x) </td>
  </tr>
  <tr>
   <td style="text-align:left;"> maximum </td>
   <td style="text-align:left;"> max(x) </td>
  </tr>
  <tr>
   <td style="text-align:left;"> mean </td>
   <td style="text-align:left;"> mean(x) </td>
  </tr>
  <tr>
   <td style="text-align:left;"> median </td>
   <td style="text-align:left;"> median(x) </td>
  </tr>
  <tr>
   <td style="text-align:left;"> standard deviation </td>
   <td style="text-align:left;"> sd(x) </td>
  </tr>
  <tr>
   <td style="text-align:left;"> variance </td>
   <td style="text-align:left;"> var(x) </td>
  </tr>
  <tr>
   <td style="text-align:left;"> rank </td>
   <td style="text-align:left;"> rank(x) </td>
  </tr>
</tbody>
</table>

* All allow for `na.rm` argument to remove `NA` values before summarizing. The default setting for this argument is *always* `na.rm = FALSE`, so if there is one `NA` value the summary will be `NA`.

* See "Maths Functions" in the RStudio Base R Cheatsheet:
https://github.com/rstudio/cheatsheets/blob/master/base-r.pdf

* Any function that operates _on a vector_ can be used...

---
class: inverse, middle, center

![](../images/alicedata-lego-colors.jpg)
## <small>"Spent day pondering grayscale vs colourscale using `ggplot`"</small>

*photo and caption courtesy [@alice-data](https://twitter.com/alice_data)*

---
# Today's lab: COLORS

Specifically, discrete colors.

At the end of today's lab, you'll see an extra section on continuous colors.

---
## But first: `shape`

`Shape` works like any other `ggplot` aesthetic:

```r
penguins %>% 
  ggplot(aes(x=bill_length_mm, y=bill_depth_mm, 
             shape=species, color=species)) + 
  geom_point(size=2)
```

---
## But first: `shape`

---
## Shapes with `color = "hotpink"`

---
## Shapes with `fill = "gold"`

---
## Default shape for `geom_point`

🕵🏽 Requires spelunking into the dark corners of the `ggplot2` code on [GitHub](https://github.com/tidyverse/ggplot2/blob/master/R/geom-point.r):

```r
default_aes = aes(
    shape = 19, colour = "black", size = 1.5, fill = NA,
    alpha = NA, stroke = 0.5
  )
```

So, the default for `geom_point(shape = 19)`! This is important to remember: this shape only "understands" the *color* aesthetic, but not the *fill* aesthetic.

---
## Beyond `fill`/`color` aesthetics: "Scales"

For when we need to go beyond what `fill` and `color` can do for us...

* Like adding multiple `geom_*` objects, we can add several `scale_`

* e.g. `scale_fill_discrete()`, etc.
  
  * Time to jump to the lab worksheet!