r_ggplot

Question 1

What does the “gg” in “ggplot” stand for?

Answer 1

• Grammar Graphics

Question 2

How do you import ggplot

Answer 2

• library(ggplot2)

Question 3

• What is one limitation of ggplot

Answer 3

• Works exclusively with data tables
• In these data tables:
  
  • rows have to be observations
  • columns have to be variables

Question 4

What happens if you do the following?

• murders %>% ggplot()

Answer 4

• This renders a blank plot, since no geometry has been defined.

Question 5

What is Median Absolute Deviation

Answer 5

• mad(x)
• robust measure of central tendency
• not sensitive to the presence of outliers
  
  • unlike standard mean and standard deviation

Question 6

How can you use exploratory data analysis to detect that an error was made?

Answer 6

• A boxplot, histogram, or qq-plot would reveal a clear outlier

Question 7

Given the following:

• x <- Galton$child

Do the following:

• Write a function called error_avg that takes a value k and returns the average of the vector x after the first entry changed to k.
• Show the results for k=10000 and k=-10000.

Answer 7

• error_avg <- function(k){
  x[1] <- k
  return(mean(x))
  }

error_avg(10000)
error_avg(-10000)

Question 8

What are layers in ggplot?

Answer 8

• In ggplot, graphs are created by adding layers
• They are added component by component
• Layers can:
  
  • define geometries
  • compute summary statistics
  • define what scales to use
  • even change styles
• To add layers, we use a symbol plus

Question 9

Create a ggplot scatter plot with the following:

• “murders” dataset
• x-axis: population/10^6
• y-axis: total

Answer 9

murders %>% ggplot() +

geom_point( aes( x = population/10^6, y = total) )

Question 10

What are the functions to add labels to the x and y axis’, and add a title to the plot?

Answer 10

• x-axis label: xlab(“<label>”)</label>
• y-axis label: ylab(“<label>”)</label>
• plot title: ggtitle(“<title>")</title>

Question 11

How do you add a color to a category of a variable, such as region? Example

Answer 11

• We have to use a mapping
• To map each point to a color, we need to use aes
• geom_point( aes( col=region), size = 3)

Question 12

How would you add a line to the plot with the following characteristics:

• Dashed
• Goes through log10(r)
• Darkgrey color

Answer 12

• geom_abline( intercept = log10(r), lty = 2, color = “darkgrey”)

Question 13

How do you do the following:

• Change the legend label from “region” to “Region”

Answer 13

• scale_color_discrete(name = “Region”)

Question 14

How would you do the following:

• change the plot them to “theme_economist”

Answer 14

• library(ggthemes)
• theme_economist()

Question 15

Create a plot of the following:

• data: male heights from the heights dataset
• bin width: 1
• color: bars blue with black border
• label: (x-axis) “Male heights in inches”
• title: (plot) “Histogram”

Answer 15

p <- heights %>%

+ filter(sex==”Male”) %>%

+ ggplot(aes (x = height) )

+ geom_histogram(binwidth = 1, fill = “blue”, col = “black”)

+ xlab(“Male heights in inches”)

+ ggtitle(“Histogram”)

Question 16

Take the previously created plot ‘p’ and make the following change:

• change to smooth density plot with blue color

Answer 16

• p + geom_density(fill = “blue”)

Question 17

What is the default standard deviation and mean of a qqplot

Answer 17

• mean = 0
• standard deviation = 1

Question 18

How would you adjust an existing qqplot“p” to change the default mean and sd to the mean and sd of the “height” variable

Answer 18

• Create a new object “params” with mean(height) and sd(height)
• add the new object “params” to the dparams function of geom_qq

params <- heights %>%

+ filter(sex = “Male”) %>%

+ summarize(mean = mean(height), sd = sd(height) )

p + geom_qq(dparams = params)

Question 19

What is the class of p <- ggplot(murders)

Answer 19

• ggplot

Question 20

Using the pipe %>%, create an object p associated with the heights dataset instead of with the murders dataset as in previous exercises.

Answer 20

• p <- murders %>% ggplot()

Question 21

Create a scatter plot from the murders with the following:

• “total” on the x-axis
• “population” on the y-axis
• label the points with “abb”
• color the labels blue

Answer 21

murders %>% ggplot(aes(population, total,label= abb)) +
geom_label(color=”blue”)

Question 22

Create a scatter plot from the murders with the following:

• “total” on the x-axis
• “population” on the y-axis
• label the points with “abb”
• color the labels by region

Answer 22

murders %>% ggplot(aes(population, total, label = abb, color=region)) +
geom_label()

Question 23

Make the following change to the existing ggplot ‘p’:

• Change both axes to be in the log scale. Make sure you do not redefine p - just add the appropriate layers.
• Add a title to the plot “Gun murder data”

Answer 23

p + scale_x_log10() +

scale_y_log10() +

ggtitle(“Gun murder data”)

Question 24

Create a ggplot object called p using the pipe to assign the heights data to a ggplot object.

Assign height to the x values through the aes function.

Answer 24

• p <- heights %>% ggplot(aes(x = height))

Question 25

Create a ggplot from the “heights” dataset with separate density plots for males and females by defining group by sex

• 2 ways to do this

Answer 25

heights %>%
ggplot(aes(height, group = sex)) +
geom_density()

heights %>%
ggplot(aes(height, color = sex)) +
geom_density()