One Variable

class: middle, center, title-slide

# One Variable
## STAE04: Data Visualization
### Johan Larsson
### The Department of Statistics, Lund University

---

## Visualizing a Single Variable

best place to start: continuous or discrete one-dimensional data

often need **statistical transformation** or **position adjustment**

Dotplots are great for small and discrete data.

```r
library(tidyverse)

ggplot(faithful, aes(waiting)) +
* geom_dotplot(binwidth = 1)
```

<div class="figure" style="text-align: center">
<img src="05-one-variable_files/figure-html/unnamed-chunk-1-1.png" alt="Duration between eruptions of the Old Faithful geyser." width="504" />
<p class="caption">Duration between eruptions of the Old Faithful geyser.</p>
</div>

---

## Histograms

.pull-left[
For larger datasets it is often better to
use a statistical transformation.

Histograms accomplish this by **binning** and **summarizing** (counting).

**caution:** choice of bins may introduce bias
]

.pull-right[

```r
faithful %>%
  ggplot(aes(waiting)) +
* geom_histogram(bins = 15)
```

<div class="figure" style="text-align: center">
<img src="05-one-variable_files/figure-html/unnamed-chunk-2-1.png" alt="Histogram of Old Faithful data." width="360" />
<p class="caption">Histogram of Old Faithful data.</p>
</div>

]

---

## Density Plots

attractive for variables that are continuous

but sensitive to settings (type of kernel, bandwidth, and more)

.pull-left[

```r
faithful %>%
  ggplot(aes(waiting)) +
    # tighter bandwidth
*   geom_density(bw = 5)
```

<img src="05-one-variable_files/figure-html/unnamed-chunk-3-1.png" width="360" style="display: block; margin: auto;" />
]

.pull-right[

```r
faithful %>%
  ggplot(aes(waiting)) +
    # longer bandwidth
*   geom_density(bw = 10)
```

<img src="05-one-variable_files/figure-html/unnamed-chunk-4-1.png" width="360" style="display: block; margin: auto;" />
]

often good idea to add a **rug** layer (`geom_rug()`) to density plots

---

## Boxplots

.pull-left[
most common type:

- **median** as middle bar, 
  - 1st and 3rd **quartiles** as edges of box,
  - whiskers to last observation within 1.5 times the
    inter-quartile range, and
  - points (or stars) beyond whiskers

useful when visualizing categories or many variables at once

more abstraction than histogram or densityplot

not suitable for data with multiple modes
]

.pull-right[

```r
faithful %>%
  ggplot(aes(waiting)) +
*   geom_boxplot()
```

<div class="figure" style="text-align: center">
<img src="05-one-variable_files/figure-html/unnamed-chunk-5-1.png" alt="Boxplot of the Old Faithful data, which fails completely in accurately describing the distribution." width="360" />
<p class="caption">Boxplot of the Old Faithful data, which fails completely in accurately describing the distribution.</p>
</div>
]

---

## Violin Plots

a type of density plot, often used instead of boxplot

same pitfalls as with other density plots

```r
faithful %>%
  ggplot(aes(waiting, y = 1)) +
*   geom_violin()
```

<div class="figure" style="text-align: center">
<img src="05-one-variable_files/figure-html/unnamed-chunk-6-1.png" alt="geom_violin() does not work with a single variable, so we need to use the trick y = 1 here." width="468" />
<p class="caption">geom_violin() does not work with a single variable, so we need to use the trick y = 1 here.</p>
</div>

---

## Several Variables at Once

boxplots and violin plots are useful for visualizing multiple
variables at once

need to first transform data into **long** format

```r
*pivot_longer(mpg, c(displ, cty, hwy)) %>%
  ggplot(aes(name, value)) +
    geom_boxplot()
```

---

### Facets

mostly used to visualize **small multiples** by splitting the plot into facets on a 
categorical variable

but can also be used to visualize multiple variables at once

for this use case, it is normally good to set `scales = "free_x"` (or sometimes `"free"`)

```r
pivot_longer(mpg, c(displ, cty, hwy)) %>%
  ggplot(aes(value)) +
    geom_histogram(bins = 10) +
*   facet_wrap("name", scales = "free_x")
```