The primary references for this lecture are

These notes contain some basic examples and the references are a more comprehensive overview.

Load data

library(tidyverse)
data<- read_csv('https://raw.githubusercontent.com/idc9/stor390/master/data/movies.csv')

Basic plot

ggplot(data=data) +
    geom_point(aes(x=imdb_rating, y=critics_score))

Labels

axis, title

ggplot will label the axes with the original variable names – let’s change that (and add a title). See r4ds 28.2 for more information.

# add labs()
ggplot(data=data) +
    geom_point(aes(x=imdb_rating, y=critics_score)) +
    labs(x='IMDB',
         y='Rotten Tomatoes (critics)',
         title='IMDB scores are predictive of Rotton Tomatoes')

# using ggtitle, xlab
# ggplot(data=data) +
#     geom_point(aes(x=imdb_rating, y=critics_score)) +
#     ggtitle('IMDB scores are predictive of Rotton Tomatoes') +
#     xlab('IMDB')

If you do a variable transformation in the geom_ ggplot will automatically add the transformation.

# add labs()
ggplot(data=data) +
    geom_point(aes(x=log10(imdb_rating), y=log10(critics_score)))

Annotation

Sometimes it’s useful to annotate a point.

# data frame for annotation
bad_movie <- data %>% 
    filter(critics_score== min(critics_score))


# add text label
ggplot(data=data) +
    geom_point(aes(x=imdb_rating, y=critics_score)) + 
    geom_text(data=bad_movie, aes(x=imdb_rating, y=critics_score, label=title), color='red')

The label goes off the plot – let’s nudge it back onto the plot.

ggplot(data=data) +
    geom_point(aes(x=imdb_rating, y=critics_score)) + 
    geom_text(data=bad_movie, aes(x=imdb_rating, y=critics_score, label=title),color='red', nudge_x= .3, nudge_y= 2)

Like everything in ggplot, geom_text takes a data frame as input. You can also use this to add text anywhere to the plot.

# data frame storing some text and the position the text should go on the plot
text_df <- tibble(x=4, y=75, text='some text goes here')

ggplot(data=data) +
    geom_point(aes(x=imdb_rating, y=critics_score)) +
    geom_text(data=text_df, aes(x=x, y=y, label=text))

axis limits

You can set the axis limits by adding lims().

ggplot(data=data) +
    geom_point(aes(x=imdb_rating, y=critics_score)) +
    lims(x=c(-5, 15), y=c(-10, 100))

ggplot(data=data) +
    geom_point(aes(x=imdb_rating, y=critics_score)) +
    scale_y_continuous(breaks=seq(from=0, to=100, by=10))

Theme

The theme controls the plot background and the legend. See the theme documentation for a long list of examples of what you can do. Below are a few important examples.

Change the background

By default ggplot uses a grey background – let’s make a blank background.

ggplot(data=data) +
    geom_point(aes(x=imdb_rating, y=critics_score)) +
    theme(panel.background = element_blank())

Legend

Adding additional aesthetic mappings like color or shape will automatically add a legend.

ggplot(data=data) +
    geom_point(aes(x=imdb_rating, y=critics_score, color=best_pic_nom, shape=best_pic_nom))

Change the legend position

ggplot(data=data) +
    geom_point(aes(x=imdb_rating, y=critics_score, color=best_pic_nom, shape=best_pic_nom)) +
    theme(legend.position = "bottom")

Or turn the legend off

ggplot(data=data) +
    geom_point(aes(x=imdb_rating, y=critics_score, color=best_pic_nom, shape=best_pic_nom)) +
    theme(legend.position = "none")

Customize colors

ggplot has default colors/shapes. You can customize the color with scale_colour_manual.

ggplot(data=data) +
    geom_point(aes(x=imdb_rating, y=critics_score, color=best_pic_nom, shape=best_pic_nom)) +
    scale_colour_manual(values = c(no = "red", yes = "blue"))

Each of the aesthetic mapping (color, shape, fill, etc) have their own scale_BLAH_manual. For example, we could also change the size and shape

ggplot(data=data) +
    geom_point(aes(x=imdb_rating, y=critics_score, color=best_pic_nom, shape=best_pic_nom, size=best_pic_nom)) +
    scale_colour_manual(values = c(no = "red", yes = "blue")) +
    scale_size_manual(values = c(no = 1, yes = 2)) +
    scale_shape_manual(values = c(no = 15, yes = 17))

```