Comparisons

.leftcol30[
<center>
<img src="https://raw.githubusercontent.com/emse-eda-gwu/2021-Spring/master/images/eda_hex_sticker.png" width=250>
</center>
]

### <svg style="height:0.8em;top:.04em;position:relative;fill:white;" viewBox="0 0 512 512"><path d="M496 128v16a8 8 0 0 1-8 8h-24v12c0 6.627-5.373 12-12 12H60c-6.627 0-12-5.373-12-12v-12H24a8 8 0 0 1-8-8v-16a8 8 0 0 1 4.941-7.392l232-88a7.996 7.996 0 0 1 6.118 0l232 88A8 8 0 0 1 496 128zm-24 304H40c-13.255 0-24 10.745-24 24v16a8 8 0 0 0 8 8h464a8 8 0 0 0 8-8v-16c0-13.255-10.745-24-24-24zM96 192v192H60c-6.627 0-12 5.373-12 12v20h416v-20c0-6.627-5.373-12-12-12h-36V192h-64v192h-64V192h-64v192h-64V192H96z"/></svg> EMSE 4575: Exploratory Data Analysis
### <svg style="height:0.8em;top:.04em;position:relative;fill:white;" viewBox="0 0 448 512"><path d="M224 256c70.7 0 128-57.3 128-128S294.7 0 224 0 96 57.3 96 128s57.3 128 128 128zm89.6 32h-16.7c-22.2 10.2-46.9 16-72.9 16s-50.6-5.8-72.9-16h-16.7C60.2 288 0 348.2 0 422.4V464c0 26.5 21.5 48 48 48h352c26.5 0 48-21.5 48-48v-41.6c0-74.2-60.2-134.4-134.4-134.4z"/></svg> John Paul Helveston
### <svg style="height:0.8em;top:.04em;position:relative;fill:white;" viewBox="0 0 448 512"><path d="M0 464c0 26.5 21.5 48 48 48h352c26.5 0 48-21.5 48-48V192H0v272zm320-196c0-6.6 5.4-12 12-12h40c6.6 0 12 5.4 12 12v40c0 6.6-5.4 12-12 12h-40c-6.6 0-12-5.4-12-12v-40zm0 128c0-6.6 5.4-12 12-12h40c6.6 0 12 5.4 12 12v40c0 6.6-5.4 12-12 12h-40c-6.6 0-12-5.4-12-12v-40zM192 268c0-6.6 5.4-12 12-12h40c6.6 0 12 5.4 12 12v40c0 6.6-5.4 12-12 12h-40c-6.6 0-12-5.4-12-12v-40zm0 128c0-6.6 5.4-12 12-12h40c6.6 0 12 5.4 12 12v40c0 6.6-5.4 12-12 12h-40c-6.6 0-12-5.4-12-12v-40zM64 268c0-6.6 5.4-12 12-12h40c6.6 0 12 5.4 12 12v40c0 6.6-5.4 12-12 12H76c-6.6 0-12-5.4-12-12v-40zm0 128c0-6.6 5.4-12 12-12h40c6.6 0 12 5.4 12 12v40c0 6.6-5.4 12-12 12H76c-6.6 0-12-5.4-12-12v-40zM400 64h-48V16c0-8.8-7.2-16-16-16h-32c-8.8 0-16 7.2-16 16v48H160V16c0-8.8-7.2-16-16-16h-32c-8.8 0-16 7.2-16 16v48H48C21.5 64 0 85.5 0 112v48h448v-48c0-26.5-21.5-48-48-48z"/></svg> February 24, 2021
]

---
class: inverse

# Quiz 2

---
class: center, middle, inverse

# .fancy[.blue[Tip of the week]]

# Shortcut keys

---
class: middle, inverse

## 1) Quick shortcuts

- **Windows**: `ALT` + `-`
- **Mac**: `OPTION` + `-`
]

- **Windows**: `CTRL` + `SHIFT` + `M`
- **Mac**: `COMMAND` + `SHIFT` + `M`
]

---
class: middle, inverse

## 2) Edit multiple lines of code at once

1. Press and hold `ALT` (Windows) or `OPTION` (Mac)
2. Select multiple lines of code

https://twitter.com/i/status/995394452821721088

---
class: middle, inverse

.rightcol80[
## "At the heart of quantitative reasoning is a single question: Compared to what?"
## -- Edward Tufte
]

---

## Today's data

```r
college_all_ages <- read_csv(here('data', 'college_all_ages.csv'))
gapminder        <- read_csv(here('data', 'gapminder.csv'))
marathon         <- read_csv(here('data', 'marathon.csv'))
milk_production  <- read_csv(here('data', 'milk_production.csv'))
*internet_regions <- read_csv(here('data', 'internet_users_region.csv'))
```

## New packages

```r
install.packages("ggrepel")
install.packages("ggridges")
```

---

# Week 7: .fancy[Comparisons]

## 1. Comparing to a reference
## 2. Comparing variables

## BREAK

## 3. Comparing distributions

---

# Week 7: .fancy[Comparisons]

## 1. .orange[Comparing to a reference]
## 2. Comparing variables

## BREAK

## 3. Comparing distributions

<!--
Comparing things to a reference line:
- Add a simple line
- diverging bars / lollipops,

- In any of these plots, adding a benchmark can be really useful
- Another way is to compare things to a **computed** benchmark, like the mean - diverging bars / lollipops
-->

---
class: middle

## .center[For this section, we'll be using this data frame:]

```r
gapminder_americas <- gapminder %>%
  filter(continent == "Americas", year == 2007) %>%
  mutate(country = fct_reorder(country, lifeExp))
```
]

---
class: center

## Use reference lines to add context to chart

.rightcol[
<img src="figs/life-exp-dots-mean-1.png" width="432" style="display: block; margin: auto;" />
]

---
class: center

## Or make zero the reference line

.leftcol[
<img src="figs/life-exp-dots-diverging-1.png" width="432" style="display: block; margin: auto;" />
]

.rightcol[
<img src="figs/life-exp-bars-diverging-1.png" width="432" style="display: block; margin: auto;" />
]

---

## How to add a reference line

Add vertical line with `geom_vline()`

```r
ggplot(gapminder_americas) +
  geom_point(
    aes(x = lifeExp, y = country),
    color = 'steelblue', size = 2.5) +
* geom_vline(
*     xintercept = mean(gapminder_americas$lifeExp),
*     color = 'red', linetype = 'dashed') +
  theme_minimal_vgrid() +
  labs(x = 'Life expectancy (years)',
       y = 'Country')
```
]]

.rightcol40[
<img src="figs/unnamed-chunk-8-1.png" width="432" style="display: block; margin: auto;" />
]

---

## How to add a reference line

Add text with `annotate()`

```r
ggplot(gapminder_americas) +
  geom_point(
    aes(x = lifeExp, y = country),
    color = 'steelblue', size = 2.5) +
  geom_vline(
      xintercept = mean(gapminder_americas$lifeExp),
      color = 'red', linetype = 'dashed') +
  annotate(
*   'text', x = 73.2, y = 'Puerto Rico',
*   color = 'red', hjust = 1,
*   label = 'Mean\nLife\nExpectancy') +
  theme_minimal_vgrid() +
  labs(x = 'Life expectancy (years)',
       y = 'Country')
```
]]

.rightcol40[.center[
<img src="figs/unnamed-chunk-10-1.png" width="432" style="display: block; margin: auto;" />
]]

---

## How to make zero the reference point

```r
gapminder_diverging <- gapminder_americas %>%
    mutate(
        # Subtract the mean
*       lifeExp = lifeExp - mean(lifeExp),
        # Define the fill color
*       color = ifelse(lifeExp > 0, 'Above', 'Below'))
```

```r
ggplot(gapminder_diverging) +
  geom_col(
*   aes(x = lifeExp, y = country, fill = color),
    width = 0.7, alpha = 0.8) +
  scale_fill_manual(
*   values = c('steelblue', 'red')) +
  theme_minimal_vgrid() +
  theme(legend.position = 'none') +
  labs(
    x = 'Country',
    y = 'Difference from mean life expectancy (years)')
```
]]

.rightcol40[
<img src="figs/unnamed-chunk-13-1.png" width="432" style="display: block; margin: auto;" />
]

---
class: inverse

### Your turn - comparing to a reference

Use the `milk_production.csv` data to create the following charts<br>showing differences from the mean state milk production in 2017.

.leftcol[
<img src="figs/milk-lollipop-mean-1.png" width="60%" style="display: block; margin: auto;" />
]

.rightcol[
<img src="figs/milk-bars-diverging-1.png" width="60%" style="display: block; margin: auto;" />
]

---

# Week 7: .fancy[Comparisons]

## 1. Comparing to a reference
## 2. .orange[Comparing variables]

## BREAK

## 3. Comparing distributions

<!--
Comparing categories with facets

Comparing two things (dodged bars, slope chart, dumbbell chart)

- dodged comparisons are fine, but really no more than 2 things.
- Finally, overlapping bars are great when you want to show when something exceeds a threshold. E.g. going over your budget.
- Using facets to break up 3-4 groups of 2 is okay.
- A better approach for multiple categories:
    - slope charts
    - dumbbell charts
-->

---
class: center, middle

## Neither of these charts are great

---
class: center

## "Parallel Coordinates" plots work well

```r
diamonds %>%
  count(clarity, cut) %>%
  ggplot(
    aes(x = clarity, y = n,
*       color = cut, group = cut)) +
  geom_line() +
  geom_point() +
  scale_y_continuous(limits = c(0, 5100)) +
  theme_half_open(font_size = 18) +
  labs(y = "Count")
```
]]

---

## Consider facets for **comparing across categories**

```r
diamonds %>%
  count(clarity, cut) %>%
  ggplot() +
  geom_col(aes(x = clarity, y = n),
           width = 0.7) +
* facet_wrap(vars(cut), nrow = 1) +
  scale_y_continuous(
    expand = expansion(mult = c(0, 0.05))) +
  theme_minimal_hgrid(font_size = 16)
```
]

---

## Consider facets for **comparing across categories**

```r
diamonds %>%
  count(clarity, cut) %>%
  mutate(n = n / 1000) %>%
  ggplot() +
  geom_col(aes(x = clarity, y = n),
           width = 0.7) +
* facet_wrap(vars(cut), ncol = 2) +
* coord_flip() +
  scale_y_continuous(
    expand = expansion(mult = c(0, 0.05))) +
* theme_minimal_vgrid(font_size = 16) +
  labs(y = "Count (thousands)")
```
]]

.rightcol[
<img src="figs/unnamed-chunk-17-1.png" width="576" style="display: block; margin: auto;" />
]

---
background-image: url("images/ft-coronavirus.jpg")
background-size: contain

.right[<br><br><br><br><br><br><br><br><br><br><br><br><br><br><br><br><br> From [Financial Times](https://www.ft.com/coronavirus-latest)]

---
class: inverse, center

## When comparing across multiple categories, consider:

<img src="figs/unnamed-chunk-18-1.png" width="504" />
]

<img src="figs/unnamed-chunk-19-1.png" width="70%" style="display: block; margin: auto;" />
]

---

## .center[When comparing **only 2** things,<br>dodged bars are a good starting point]

```r
milk_compare <- milk_production %>%
  filter(year %in% c(1970, 2017)) %>%
  mutate(state = fct_other(state,
    keep = c('California', 'Wisconsin'))) %>%
  group_by(year, state) %>%
  summarise(
    milk_produced = sum(milk_produced) / 10^9)
```

```
#> # A tibble: 6 x 3
#> # Groups:   year [2]
#>    year state      milk_produced
#>   <dbl> <fct>              <dbl>
#> 1  1970 California          9.46
#> 2  1970 Wisconsin          18.4 
#> 3  1970 Other              89.1 
#> 4  2017 California         39.8 
#> 5  2017 Wisconsin          30.3 
#> 6  2017 Other             145.
```

]]

.rightcol[
<img src="figs/unnamed-chunk-21-1.png" width="504" style="display: block; margin: auto;" />
]

---

## .center[When comparing **only 2** things,<br>dodged bars are a good starting point]

```r
ggplot(milk_compare) +
  geom_col(
    aes(x = milk_produced, y = state,
        fill = as.factor(year)),
    width = 0.7, alpha = 0.8,
*   position = 'dodge') +
  scale_fill_manual(
      values = c('grey', 'steelblue'),
      guide  = guide_legend(reverse = TRUE)) +
  scale_x_continuous(
    expand = expansion(mult = c(0, 0.05))) +
  theme_minimal_vgrid() +
  labs(
    x = 'Milk produced (billion lbs)',
    y = NULL,
    fill = 'Year')
```
]]

.rightcol[
<img src="figs/unnamed-chunk-23-1.png" width="504" style="display: block; margin: auto;" />
]

---
class: center

## Avoid putting >2 categories in legend (if possible)

<img src="figs/milk-compare-dodged-bad-1.png" width="504" style="display: block; margin: auto;" />
]

<img src="figs/unnamed-chunk-24-1.png" width="504" style="display: block; margin: auto;" />
]

---

## .center[Or use facets to get rid of the legend!]

<img src="figs/unnamed-chunk-25-1.png" width="504" style="display: block; margin: auto;" />
]

<img src="figs/unnamed-chunk-26-1.png" width="504" style="display: block; margin: auto;" />
]

---
class: center

## "Bullet" charts are also effective for comparing **2** things

---

## How to make a "bullet" chart

```r
milk_compare %>%
* spread(year, milk_produced) %>%
  ggplot() +
  geom_col(
    aes(x = `1970`, y = state, fill = '1970'),
*       width = 0.7) +
  geom_col(
    aes(x = `2017`, y = state, fill = '2017'),
*       width = 0.3) +
  scale_fill_manual(
    values = c('grey', 'black')) +
  scale_x_continuous(
      expand = expansion(mult = c(0, 0.05))) +
  theme_minimal_vgrid(font_size = 18) +
  labs(
    x = 'Milk produced (billion lbs)',
    y = NULL,
    fill = "Year")
```
]]

.rightcol[.center[
<img src="figs/unnamed-chunk-28-1.png" width="576" style="display: block; margin: auto;" />
]]

---
class: center

## With **more than 2** things, dodged bars can get confusing

Still comparing 2 time periods, but across **10** categories

---
class: center

### Strategies for comparing 2 things across **more than 2 categories**

<img src="figs/unnamed-chunk-29-1.png" width="504" style="display: block; margin: auto;" />
]

<img src="figs/milk-dumbbell-chart-1.png" width="396" style="display: block; margin: auto;" />
]

---
class: center

### Strategies for comparing 2 things across **more than 2 categories**

<img src="figs/unnamed-chunk-30-1.png" width="504" style="display: block; margin: auto;" />
]

<img src="figs/milk-slope-chart-1.png" width="432" style="display: block; margin: auto;" />
]

---

- Compare **magnitudes** across two periods / groups

<img src="figs/unnamed-chunk-31-1.png" width="432" style="display: block; margin: auto;" />
]

- _Change_ in **rankings**
- Highlight individual categories

<img src="figs/unnamed-chunk-32-1.png" width="432" style="display: block; margin: auto;" />
]

---