2-16_ggplot2_practicals_answers.qmd

## ggplot: Answers {.unnumbered}

::: callout-warning
Make sure that you try the exercises yourself first before looking at the answers
:::

::: panel-tabset
### Question 1

Explore the data set using dim(), str() and help(), which variables are continuous, which variables are discrete? Is this data set ready for plotting with ggplot?

### Answer 1

Explore the data set using dim(), str() and help(), which variables are continuous, which variables are discrete? Is this data set ready for plotting with ggplot?

```{r, eval=F}
dim(diamonds)
str(diamonds)
help(diamonds)
```
:::

::: panel-tabset
### Question 2

Use ggplot to plot a scatterplot of the relationship between the diamonds' carat and their price

### Answer 2

Use ggplot to plot a scatterplot of the relationship between the diamonds' carat and their price

```{r, eval=F}
ggplot() + geom_point(data=diamonds, aes(x=carat, y=price))

#or

ggplot(diamonds) + geom_point(aes(x=carat, y=price))

#or

ggplot(diamonds, aes(x=carat, y=price)) + geom_point()
```

Cut, color, and clarity are factors, and therefore discrete. The others are numeric continuous variables.
:::

::: panel-tabset
### Question 3

Make all dots darkblue and set the alpha value to 0.1

### Answer 3

Make all dots darkblue and set the alpha value to 0.1

```{r, eval=F}
ggplot(diamonds, aes(x=carat, y=price)) + geom_point(color="darkblue", alpha=0.1)
```
:::

::: panel-tabset
### Question 4

Visualize the influence of the color of a diamond on its price by mapping the diamond color to the color aesthetic

### Answer 4

Visualize the influence of the color of a diamond on its price by mapping the diamond color to the color aesthetic

```{r, eval=F}
#The color of the dots will be overwritten if we specify it statically  
#in the geom_point function itself
ggplot(diamonds, aes(x=carat, y=price, color=color)) + geom_point(alpha=0.1)
```
:::

::: panel-tabset
### Question 5

Use a ggplot barplot to visualize diamond clarity depending on color, map diamond color to x and diamond clarity to fill

### Answer 5

Use a ggplot barplot to visualize diamond clarity depending on color, map diamond color to x and diamond clarity to fill

```{r, eval=F}
ggplot(diamonds, aes(x=color, fill=clarity)) + geom_bar()
```
:::

::: panel-tabset
### Question 6

Create a boxplot of the carat of a diamond based on its clarity and add whiskers using stat_boxplot

### Answer 6

Create a boxplot of the carat of a diamond based on its clarity and add whiskers using stat_boxplot

```{r, eval=F}
ggplot(diamonds, aes(x=clarity, y=carat)) + 
	stat_boxplot(geom="errorbar", width=0.5) + 
	geom_boxplot()
```
:::

::: panel-tabset
### Question 7

Add a geom_point layer to the previous plot mapping the diamonds price to the color

### Answer 7

Add a geom_point layer to the previous plot mapping the diamonds price to the color

```{r, eval=F}
ggplot(diamonds, aes(x=clarity, y=carat)) + 
	stat_boxplot(geom="errorbar", width=0.5) + 
	geom_boxplot() + 
	geom_point(aes(color=price))
```
:::

::: panel-tabset
### Question 8

Create a histogram of the price of the diamonds and separate the histograms into facets using diamond color, choose a good binwith or number of bins

### Answer 8

Create a histogram of the price of the diamonds and separate the histograms into facets using diamond color, choose a good binwith or number of bins

```{r, eval=F}
ggplot(diamonds, aes(x=price)) + 
	geom_histogram(binwidth = 100) + 
	facet_grid(color ~ .)
```
:::

::: panel-tabset
### Question 9

Create a grid of facets of the same histogram by comparing both color and cut

### Answer 9

Create a grid of facets of the same histogram by comparing both color and cut

```{r, eval=F}
ggplot(diamonds, aes(x=price)) + 
	geom_histogram(binwidth = 100) + 
	facet_grid(color ~ cut)
```
:::

::: panel-tabset
### Question 10

Use `aggregate(diamonds, by = list(cut = diamonds$cut, color = diamonds$color), mean)` to calculate the mean of all variables by cut and color. Create a heatmap of the mean prices by cut and color using geom_tile

### Answer 10

Use `aggregate(diamonds, by = list(cut = diamonds$cut, color = diamonds$color), mean)` to calculate the mean of all variables by cut and color. Create a heatmap of the mean prices by cut and color using geom_tile

```{r, eval=F}
#Aggregate uses a function (in this case mean) to aggregate all variables 
#in a given data.frame by a list of variables given in "by"
mean.price <- aggregate(diamonds, by = list(cut = diamonds$cut, color = diamonds$color), mean)

ggplot(mean.price, aes(x=cut, y=color, fill=price)) + 
	geom_tile()
```
:::

::: panel-tabset
### Question 11

Change the title of the heatmap to "Average prices"

### Answer 11

Change the title of the heatmap to "Average prices"

```{r, eval=F}
ggplot(mean.price, aes(x=cut, y=color, fill=price)) + 
	geom_tile() + 
	labs(title="Average prices")
```
:::

::: panel-tabset
### Question 12

Change the gradient of the fill scale using 'scale_fill_gradient2'. Have it go from darkblue to white to darkred, set the midpoint to 4500

### Answer 12

Change the gradient of the fill scale using 'scale_fill_gradient2'. Have it go from darkblue to white to darkred, set the midpoint to 4500

```{r, eval=F}
ggplot(mean.price, aes(x=cut, y=color, fill=price)) + 
	geom_tile() + 
	labs(title="Average prices") + 
	scale_fill_gradient2(low="darkblue", mid="white", high="darkred", midpoint = 4500)
```
:::

::: panel-tabset
### Question 13

Choose and add a theme to the heatmap, or create a theme yourself using the options listed at <http://ggplot2.tidyverse.org/reference/theme.html>

### Answer 13

Choose and add a theme to the heatmap, or create a theme yourself using the options listed at <http://ggplot2.tidyverse.org/reference/theme.html>

```{r, eval=F}
ggplot(mean.price, aes(x=cut, y=color, fill=price)) + 
	geom_tile() + 
	labs(title="Average prices") + 
	scale_fill_gradient2(low="darkblue", mid="white", high="darkred", midpoint = 4500) + 
	theme_minimal()
```
:::