Factors, Amounts, & Proportions

]

# Week 7: .fancy[Factors, Amounts, & Proportions]

### <svg aria-hidden="true" role="img" viewBox="0 0 512 512" style="height:1em;width:1em;vertical-align:-0.125em;margin-left:auto;margin-right:auto;font-size:inherit;fill:white;overflow:visible;position:relative;"><path d="M243.4 2.6l-224 96c-14 6-21.8 21-18.7 35.8S16.8 160 32 160v8c0 13.3 10.7 24 24 24H456c13.3 0 24-10.7 24-24v-8c15.2 0 28.3-10.7 31.3-25.6s-4.8-29.9-18.7-35.8l-224-96c-8-3.4-17.2-3.4-25.2 0zM128 224H64V420.3c-.6 .3-1.2 .7-1.8 1.1l-48 32c-11.7 7.8-17 22.4-12.9 35.9S17.9 512 32 512H480c14.1 0 26.5-9.2 30.6-22.7s-1.1-28.1-12.9-35.9l-48-32c-.6-.4-1.2-.7-1.8-1.1V224H384V416H344V224H280V416H232V224H168V416H128V224zM256 64a32 32 0 1 1 0 64 32 32 0 1 1 0-64z"/></svg> EMSE 4572 / 6572: Exploratory Data Analysis
### <svg aria-hidden="true" role="img" viewBox="0 0 448 512" style="height:1em;width:0.88em;vertical-align:-0.125em;margin-left:auto;margin-right:auto;font-size:inherit;fill:white;overflow:visible;position:relative;"><path d="M304 128a80 80 0 1 0 -160 0 80 80 0 1 0 160 0zM96 128a128 128 0 1 1 256 0A128 128 0 1 1 96 128zM49.3 464H398.7c-8.9-63.3-63.3-112-129-112H178.3c-65.7 0-120.1 48.7-129 112zM0 482.3C0 383.8 79.8 304 178.3 304h91.4C368.2 304 448 383.8 448 482.3c0 16.4-13.3 29.7-29.7 29.7H29.7C13.3 512 0 498.7 0 482.3z"/></svg> John Paul Helveston
### <svg aria-hidden="true" role="img" viewBox="0 0 448 512" style="height:1em;width:0.88em;vertical-align:-0.125em;margin-left:auto;margin-right:auto;font-size:inherit;fill:white;overflow:visible;position:relative;"><path d="M152 24c0-13.3-10.7-24-24-24s-24 10.7-24 24V64H64C28.7 64 0 92.7 0 128v16 48V448c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V192 144 128c0-35.3-28.7-64-64-64H344V24c0-13.3-10.7-24-24-24s-24 10.7-24 24V64H152V24zM48 192H400V448c0 8.8-7.2 16-16 16H64c-8.8 0-16-7.2-16-16V192z"/></svg> October 09, 2024

]

---

# Next projects due:

## - [Mini project 2](https://eda.seas.gwu.edu/2024-Fall/mini/2-exploring-data.html): Exploring Data (Due 10/15)
## - [Project Progress Report](https://eda.seas.gwu.edu/2024-Fall/project/2-progress-report.html) (Due 10/27)

---

## Today's data

``` r
avengers         <- read_csv(here('data', 'avengers.csv'))
bears            <- read_csv(here('data', 'north_america_bear_killings.csv'))
federal_spending <- read_csv(here('data', 'fed_spend_long.csv'))
gapminder        <- read_csv(here('data', 'gapminder.csv'))
lotr_words       <- read_csv(here('data', 'lotr_words.csv'))
milk_production  <- read_csv(here('data', 'milk_production.csv'))
wildlife_impacts <- read_csv(here('data', 'wildlife_impacts.csv'))
```

## New packages

The [{waffle} package](https://github.com/hrbrmstr/waffle)

``` r
install.packages("waffle")
```

---

# Week 7: .fancy[Factors, Amounts, & Proportions]

## 1. Manipulating factors
## 2. Graphing amounts

## BREAK

## 3. Graphing proportions

---

# Week 7: .fancy[Factors, Amounts, & Proportions]

## 1. .orange[Manipulating factors]
## 2. Graphing amounts

## BREAK

## 3. Graphing proportions

---

# Sorting in ggplot is done by reordering factors

]

]

---

## .center[Two ways to sort]

**Method 1**: Use `reorder()` inside aesthetic mapping

``` r
# Format the data frame
federal_spending %>%
  group_by(department) %>%
  summarise(
    rd_budget_bil = sum(rd_budget_mil) / 10^3) %>%

# Make the chart
  ggplot() +
  geom_col(
    aes(
      x = rd_budget_bil,
*     y = reorder(department, rd_budget_bil)
    ), 
    width = 0.7, alpha = 0.8,
    fill = "steelblue"
  ) +
  scale_x_continuous(
    expand = expansion(mult = c(0, 0.05))) +
  theme_minimal_vgrid()
```

]]

]

---

## .center[Two ways to sort]

**Method 2**: Use `fct_reorder()` when formatting the data frame

``` r
# Format the data frame
federal_spending %>%
  group_by(department) %>%
  summarise(
    rd_budget_bil = sum(rd_budget_mil) / 10^3) %>%
  mutate(
*   department = fct_reorder(department, rd_budget_bil)
  ) %>%
# Make the chart
  ggplot() +
  geom_col(
    aes(x = rd_budget_bil, y = department),
    width = 0.7, alpha = 0.8,
    fill = "steelblue"
  ) +
  scale_x_continuous(
    expand = expansion(mult = c(0, 0.05))) +
  theme_minimal_vgrid()
```

]]

]

---

## Reorder & modify factors with<br>the **forcats** library

Loaded with `library(tidyverse)`

]

]

---

## Common situations for modifying / reording factors:

### 1. Reorder factors based on another numerical variable
### 2. Reorder factors manually
### 3. Modify factors manually
### 4. What if there are too many factor levels?

---

### 1. Reorder factors based on another **numerical variable**

Use `fct_reorder()`

]]

]

---

### 2. Reorder factors **manually**

``` r
# Format the data frame
lotr_words %>%
  pivot_longer(
      names_to = 'gender', 
      values_to = 'wordCount',
      cols = Female:Male) %>%

# Make the chart
  ggplot() +
  geom_col(
    aes(x = wordCount, y = Film),
    width = 0.7, alpha = 0.8
  ) +
  scale_x_continuous(
    expand = expansion(mult = c(0, 0.05))) +
  theme_minimal_vgrid()
```

]]

]

---

### 2. Reorder factors **manually** with `fct_relevel()`

]]

]

---

### 3. Modify factors manually

<br>

``` r
# Format the data frame
lotr_words %>%
  pivot_longer(
      names_to = 'gender', 
      values_to = 'wordCount',
      cols = Female:Male) %>%

]]

]

---

### 3. Modify factors manually with `fct_recode()`

`"new label" = "old label"`

``` r
# Format the data frame
lotr_words %>%
  pivot_longer(
      names_to = 'gender', 
      values_to = 'wordCount',
      cols = Female:Male) %>%
* mutate(
*   Film = fct_recode(Film,
*     'The Fellowship\nof the Ring' = 'The Fellowship Of The Ring',
*     'The Return\nof the King' = 'The Return Of The King')) %>%

]]

]

---

### 2 & 3. Modify and reorder factors manually

``` r
# Format the data frame
lotr_words %>%
  pivot_longer(
      names_to = 'gender', 
      values_to = 'wordCount',
      cols = Female:Male) %>% 
* mutate(
*   Film = fct_relevel(Film, levels = c(
*     'The Fellowship Of The Ring',
*     'The Two Towers',
*     'The Return Of The King')),
*   Film = fct_recode(Film,
*     'The Fellowship\nof the Ring' = 'The Fellowship Of The Ring',
*     'The Return\nof the King' = 'The Return Of The King')) %>%

# Make the chart
  ggplot() +
  geom_col(
      aes(x = wordCount, y = Film),
      width = 0.7, alpha = 0.8
  ) +
  scale_x_continuous(
    expand = expansion(mult = c(0, 0.05))) +
  theme_minimal_vgrid()
```

]]

]

---

### 4. What if there are too many factor levels?

]]

]

---

### 4. What if there are too many factor levels?

**Strategy**: Merge smaller factors into "Other" with `fct_other()`

``` r
# Format the data frame
federal_spending %>%
* mutate(
*   department = fct_other(department,
*     keep = c('DOD', 'HHS', 'NIH', 'NASA', 'DOE'))) %>%
  group_by(department) %>%
  summarise(
    rd_budget_bil = sum(rd_budget_mil) / 10^3) %>%
  mutate(
    department = fct_reorder(department, rd_budget_bil)) %>%

# Make the chart
  ggplot() +
  geom_col(
    aes(x = rd_budget_bil, y = department),
    width = 0.7, alpha = 0.8,
    fill = "steelblue"
  ) +
  scale_x_continuous(
    expand = expansion(mult = c(0, 0.05))) +
  theme_minimal_vgrid()
```

]]

]

---

### 4. What if there are _really_ too many factor levels?

``` r
# Format the data frame
avengers %>%
  mutate(
*   name_alias = fct_reorder(name_alias, appearances)) %>%

# Make the chart
  ggplot() +
  geom_col(
    aes(x = appearances,y = name_alias),
    width = 0.7, alpha = 0.8
  ) +
  scale_x_continuous(
    expand = expansion(mult = c(0, 0.05))) +
  theme_minimal_vgrid()
```

]]

]

---

### 4. What if there are _really_ too many factor levels?

**Strategy**: Keep top N, drop the rest with `slice()`

``` r
# Format the data frame
avengers %>%
  mutate(
    name_alias = fct_reorder(name_alias, appearances)) %>%
* arrange(desc(appearances)) %>%
* slice(1:10) %>%

# Make the chart
  ggplot() +
  geom_col(
    aes(x = appearances, y = name_alias),
    width = 0.7, alpha = 0.8
  ) +
  scale_x_continuous(
    expand = expansion(mult = c(0, 0.05))) +
  theme_minimal_vgrid()
```

]]

]

---

### 4. What if there are _really_ too many factor levels?

`slice()` works with grouping too!

``` r
# Format the data frame
avengers %>%
  mutate(
    name_alias = fct_reorder(name_alias, appearances)) %>%
* arrange(desc(appearances)) %>%
* group_by(gender) %>%
* slice(1:10) %>%

# Make the chart
  ggplot() +
  geom_col(
    aes(
      x = appearances, 
      y = name_alias,
      fill = gender
    ),
    width = 0.7, alpha = 0.8
  ) +
  scale_x_continuous(
    expand = expansion(mult = c(0, 0.05))) +
  theme_minimal_vgrid()
```

]]

]

---

## Your turn - practice manipulating factors

Use the `wildlife_impacts` data to create the following plot

---

# Week 7: .fancy[Factors, Amounts, & Proportions]

## 1. Manipulating factors
## 2. .orange[Graphing amounts]

## BREAK

## 3. Graphing proportions

---

# Show amounts with:

]

]

]

---

## Bar chart

]

## Dot chart

]

## Lollipop chart

]

---

## Bars are good for highlighting specific categories

---

## Use lollipops when:

### - The bars are overwhelming<br> - You're not highlighting categories

]

]

---

## Or use dots and don't set axis to 0

]

]

---

## How to make a **Bar chart**

``` r
# Summarize the data
federal_spending %>%
  group_by(department) %>%
  summarise(rd_budget_bil = sum(rd_budget_mil) / 10^3) %>%
* mutate(department = fct_reorder(department, rd_budget_bil)) %>%

# Make chart
  ggplot() +
* geom_col(
*   aes(x = rd_budget_bil, y = department),
    width = 0.7, alpha = 0.8,
    fill = 'steelblue') +
  scale_x_continuous(
    expand = expansion(mult = c(0, 0.05))) +
  theme_minimal_vgrid()
```

]]

]

---

## Filling the bars with color

``` r
# Summarize the data
federal_spending %>%
  group_by(department) %>%
  summarise(rd_budget_bil = sum(rd_budget_mil) / 10^3) %>%
  mutate(
    department = fct_reorder(department, rd_budget_bil),
*   is_dod = if_else(
*     department == 'DOD', TRUE, FALSE)) %>%

# Make the chart
  ggplot() +
  geom_col(
    aes(x = rd_budget_bil, y = department,
*       fill = is_dod),
    width = 0.7, alpha = 0.8) +
  scale_x_continuous(
    expand = expansion(mult = c(0, 0.05))) +
  theme_minimal_vgrid() +
  theme(legend.position = 'none')
```

]]

The DOD's R&D budget is nearly the same as all other departments combined

]]

---

## Filling the bars with color

``` r
# Summarize the data
federal_spending %>%
  group_by(department) %>%
  summarise(rd_budget_bil = sum(rd_budget_mil) / 10^3) %>%
  mutate(
    department = fct_reorder(department, rd_budget_bil),
    is_dod = if_else(
      department == 'DOD', TRUE, FALSE)) %>%

# Make the chart
  ggplot() +
  geom_col(
    aes(x = rd_budget_bil, y = department, 
        fill = is_dod),
    width = 0.7, alpha = 0.8) +
  scale_x_continuous(
    expand = expansion(mult = c(0, 0.05))) +
* scale_fill_manual(values = c('grey', 'steelblue')) +
  theme_minimal_vgrid() +
  theme(legend.position = 'none')
```

]]

The DOD's R&D budget is nearly the same as all other departments combined

]]

---

## How to make a **Dot chart**

Summarize data frame:

``` r
# Summarize the data
federal_spending %>%
  group_by(department) %>%
  summarise(
    rd_budget_bil = sum(rd_budget_mil) / 10^3) %>%
  mutate(
    department = fct_reorder(department, rd_budget_bil)) %>%

# Make the chart
  ggplot() +
* geom_point(
*   aes(x = rd_budget_bil, y = department),
    size = 2.5, color = 'steelblue') +
  theme_minimal_vgrid()
```

]]

**Dot chart** of federal R&D spending by department

]]

---

## How to make a **Lollipop chart**

Summarize data frame:

``` r
# Summarize the data
federal_spending %>%
  group_by(department) %>%
  summarise(
    rd_budget_bil = sum(rd_budget_mil) / 10^3) %>%
  mutate(
    department = fct_reorder(department, rd_budget_bil)) %>%

# Make the chart
  ggplot() +
* geom_segment(
*   aes(x = 0, xend = rd_budget_bil,
*       y = department, yend = department),
    color = 'grey') +
* geom_point(
*   aes(x = rd_budget_bil, y = department),
    size = 2.5, color = 'steelblue') +
  theme_minimal_vgrid()
```

]]

**Lollipop chart** of federal R&D spending by department

]]

---

## Your turn - practice plotting amounts

Create the following charts:

Data: `bears`
<center>
<img src="figs/bear_bars.png" width=500>
</center>

]

Data: `milk_production`
<center>
<img src="figs/milk_dots_top10-1.png">
</center>

]

---

# Break!

## Stand up, Move around, Stretch!

---

# Week 7: .fancy[Factors, Amounts, & Proportions]

## 1. Manipulating factors
## 2. Graphing amounts

## BREAK

## 3. .orange[Graphing proportions]

---

# Show proportions with:

]

]

]

---

## Bar charts

]

## Pie charts

]

## Waffle charts

]

---

## Stacked bars

# Make the chart
  ggplot() +
* geom_col(
*   aes(x = "", y = milk_produced, fill = state),
    width = 0.7, alpha = 0.8) +
  scale_y_continuous(
    expand = expansion(mult = c(0, 0.05))) +
  theme_minimal_hgrid() +
  labs(x = NULL,
       y = 'Milk produced (lbs)',
       fill = 'State',
       title = '2017 Milk Production\nby State')
```

]]

]

---

## Stacked bars - Rotated also looks good

# Make the chart
  ggplot() +
* geom_col(
*   aes(x = milk_produced, y = "", fill = state),
    width = 0.7, alpha = 0.8) +
  scale_x_continuous(
    expand = expansion(mult = c(0, 0.05))) +
  theme_minimal_hgrid() +
  labs(y = NULL,
       x = 'Milk produced (lbs)',
       fill = 'State',
       title = '2017 Milk Production by State')
```

]]

]

---

## Stacked bars - not great for more than a few categories

``` r
# Format the data
milk_production %>%
  filter(year == 2017) %>%
  mutate(state = fct_other(state,
*   keep = c('California', 'Wisconsin',
*            'New York', 'Idaho'))) %>%
  group_by(state) %>%
  summarise(milk_produced = sum(milk_produced))

# Make the chart
  ggplot() +
  geom_col(
    aes(x = "", y = milk_produced, fill = state),
    width = 0.7, alpha = 0.8) +
  scale_y_continuous(
    expand = expansion(mult = c(0, 0.05))) +
  theme_minimal_vgrid() +
  labs(x = NULL,
       y = 'Milk produced (lbs)',
       fill = 'State',
       title = '2017 Milk Production\nby State')
```

]]

]

---

## Dodged bars

Better for **part-to-whole comparison**

``` r
# Format the data
milk_production %>%
  filter(year == 2017) %>%
  mutate(state = fct_other(state,
    keep = c('California', 'Wisconsin'))) %>%
  group_by(state) %>%
  summarise(milk_produced = sum(milk_produced)) %>%
  mutate(state = fct_reorder(state, milk_produced)) %>%

# Make the chart
  ggplot() +
  geom_col(
    aes(x = milk_produced, y = state),
    width = 0.7, alpha = 0.8) +
  scale_x_continuous(
    expand = expansion(mult = c(0, 0.05))) +
  theme_minimal_vgrid() +
  labs(x = 'Milk produced (lbs)',
       y = 'State',
       title = '2017 Milk Production by State')
```

]]

Okay:
<center>
<img src="figs/milk_2017_bars_stacked_rotated.png">
</center>

Better:
<center>
<img src="figs/milk_2017_bars_dodged.png">
</center>

]

---

## Dodged bars

``` r
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)) %>%

# Make the chart
  ggplot() +
  geom_col(
    aes(x = milk_produced, 
        y = as.factor(year),
        fill = state),
*   position = 'dodge',
    width = 0.7, alpha = 0.8) +
  scale_x_continuous(
    expand = expansion(mult = c(0, 0.05))) +
  theme_minimal_vgrid() +
  labs(x = 'Milk produced (lbs)',
       y = 'Year',
       fill = 'State',
       title = '1970 & 2017 Milk Production by State')
```

]]

Better for comparing **total**:

Better for comparing **parts**:

]

---

## .center[Where stacking is useful]

.font70[https://fivethirtyeight.com/features/the-dollar-and-cents-case-against-hollywoods-exclusion-of-women/]

]

### - **2 to 3 groups**
### - Proportions over time

]

---

## .center[Where stacking is useful]

]

### - 2 to 3 groups
### - **Proportions over time**

]

---

## The Notorious P-I-E

Start with a bar chart

# Make the chart
  ggplot() +
* geom_col(
*   aes(x = "", y = milk_produced, fill = state),
    width = 0.7, alpha = 0.8) +
  theme_minimal_hgrid() +
  labs(x = NULL,
       y = 'Milk produced (lbs)',
       fill = 'State',
       title = '2017 Milk Production\nby State')
```

]]

]

---

## The Notorious P-I-E

Convert bar to pie with `coord_polar()`

# Make the chart
  ggplot() +
  geom_col(
    aes(x = "", y = milk_produced, fill = state),
    width = 0.7, alpha = 0.8) +
* coord_polar(theta = "y") +
  theme_minimal_hgrid() +
  labs(x = NULL,
       y = 'Milk produced (lbs)',
       fill = 'State',
       title = '2017 Milk Production by State')
```

]]

]

---

``` r
# Format the data
milk_production %>%
  filter(year == 2017) %>%
  mutate(state = fct_other(state,
    keep = c('California', 'Wisconsin'))) %>%
  group_by(state) %>%
  summarise(milk_produced = sum(milk_produced)) %>%
* arrange(desc(state)) %>%
* mutate(p = 100*(milk_produced / sum(milk_produced)),
*        label = str_c(round(p), '%')) %>%

# Make the chart
  ggplot() +
  geom_col(
    aes(x = "", y = milk_produced, fill = state),
    width = 0.7, alpha = 0.8) +
* geom_text(
*   aes(x = "", y = milk_produced, label = label),
*   color = "white", size = 6,
*   position = position_stack(vjust = 0.5)) +
  coord_polar(theta = "y") +
* theme_map() +
  labs(x = NULL,
       y = NULL,
       fill = 'State',
       title = '2017 Milk Production by State')
```

]]

### The Notorious P-I-E

Final chart with labels & `theme_map()`

]

---

## Pies are still useful if the sum of components matters

]

]

]

---

## The best pies are **square pies**

---

### Waffle plots

``` r
*library(waffle)

# Format the data
milk_production %>%
  filter(year == 2017) %>%
  mutate(state = fct_other(state,
    keep = c('California', 'Wisconsin'))) %>%
  group_by(state) %>%
  summarise(milk_produced = sum(milk_produced)) %>%
* mutate(milk_produced = milk_produced / 10^9) %>%

# Make the chart
  ggplot() +
* geom_waffle(
*   aes(fill = state, values = milk_produced),
*   color = "white", size = 1, n_rows = 10) +
  scale_x_discrete(expand = c(0, 0)) +
  scale_y_discrete(expand = c(0, 0)) +
  theme_minimal() +
  labs(fill = 'State',
       x = NULL, y = NULL,
       title = '2017 Milk Production by State',
       subtitle = '(1 square = 1 billion lbs)')
```

]]

**Use values between 100 - 1,000**

```
#> # A tibble: 3 × 2
#>   state      milk_produced
#>   <fct>              <dbl>
#> 1 California          39.8
#> 2 Wisconsin           30.3
#> 3 Other              145.
```

]

---

### Waffle plots

``` r
*library(waffle)

# Make the chart
  ggplot() +
  geom_waffle(
    aes(fill = state, values = milk_produced),
    color = "white", size = 1, n_cols = 10,
*   flip = TRUE) +
  scale_x_discrete(expand = c(0, 0)) +
  scale_y_discrete(expand = c(0, 0)) +
  theme_minimal() +
  labs(fill = 'State',
       x = NULL, y = NULL,
       title = '2017 Milk Production by State',
       subtitle = '(1 square = 1 billion lbs)')
```

]]

Use `flip = TRUE` for vertical

```
#> # A tibble: 3 × 2
#>   state      milk_produced
#>   <fct>              <dbl>
#> 1 California          39.8
#> 2 Wisconsin           30.3
#> 3 Other              145.
```

]

---

``` r
library(waffle)

# Format the data
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)) %>%
  mutate(milk_produced = milk_produced / 10^9) %>%

# Make the chart
  ggplot() +
  geom_waffle(
    aes(fill = state, values = milk_produced),
    color = "white", size = 1, n_rows = 10,
    flip = TRUE) +
* facet_wrap(vars(year), strip.position = 'bottom') +
  scale_x_discrete(expand = c(0, 0)) +
  scale_y_discrete(expand = c(0, 0)) +
  theme_minimal() +
  labs(fill = 'State',
       x = NULL, y = NULL,
       title = '1970 & 2017 Milk Production by State',
       subtitle = '(1 square = 1 billion lbs)')
```

]]

Use `facet_wrap` for side-by-side waffles

```
#> # A tibble: 3 × 2
#>   state      milk_produced
#>   <fct>              <dbl>
#> 1 California          39.8
#> 2 Wisconsin           30.3
#> 3 Other              145.
```

]

---

Stacked bars

Dodged bars

]

Pie chart

Waffle chart

]

---

## Your turn

Using the `wildlife_impacts` data, create plots that shows the proportion of incidents that occur at each different time of day.

For this exercise, you can remove `NA` values.

Try to create the following plots:

- Stacked bars
- Dodged bars
- Pie chart
- Waffle chart

]

To get started, you'll need to first summarize the data:

``` r
wildlife_summary <- wildlife_impacts %>%
  filter(!is.na(time_of_day)) %>%
  count(time_of_day)

wildlife_summary
```

```
#> # A tibble: 4 × 2
#>   time_of_day     n
#>   <chr>       <int>
#> 1 Dawn         1270
#> 2 Day         25123
#> 3 Dusk         1717
#> 4 Night       12735
```

]