The exercises are part of the Fundamentals of R course. For more, see the R for the Rest of Us website.

Load Packages

Let’s load the packages we need. These include tidyverse (especially the dplyr package) and janitor.

library(tidyverse)
## ── Attaching packages ───────────────────────────────────────── tidyverse 1.2.1 ──
## ✔ ggplot2 3.1.1          ✔ purrr   0.3.2     
## ✔ tibble  2.1.1          ✔ dplyr   0.8.0.1   
## ✔ tidyr   0.8.3.9000     ✔ stringr 1.4.0     
## ✔ readr   1.3.1          ✔ forcats 0.4.0
## ── Conflicts ──────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag()    masks stats::lag()
library(janitor)
## 
## Attaching package: 'janitor'
## The following objects are masked from 'package:stats':
## 
##     chisq.test, fisher.test

Import NHANES Data

Import your data into a data frame called NHANES. Use clean_names to make your variable names easy to work with.

nhanes <- read_csv("data/nhanes.csv") %>%
  clean_names()
## Parsed with column specification:
## cols(
##   .default = col_character(),
##   ID = col_double(),
##   Age = col_double(),
##   Weight = col_double(),
##   Height = col_double(),
##   BMI = col_double(),
##   DaysPhysHlthBad = col_double(),
##   DaysMentHlthBad = col_double(),
##   SleepHrsNight = col_double(),
##   PhysActiveDays = col_double(),
##   TVHrsDay = col_logical()
## )
## See spec(...) for full column specifications.
## Warning: 4859 parsing failures.
##  row      col           expected    actual              file
## 5001 TVHrsDay 1/0/T/F/TRUE/FALSE 2_hr      'data/nhanes.csv'
## 5002 TVHrsDay 1/0/T/F/TRUE/FALSE More_4_hr 'data/nhanes.csv'
## 5003 TVHrsDay 1/0/T/F/TRUE/FALSE 4_hr      'data/nhanes.csv'
## 5004 TVHrsDay 1/0/T/F/TRUE/FALSE 4_hr      'data/nhanes.csv'
## 5005 TVHrsDay 1/0/T/F/TRUE/FALSE 1_hr      'data/nhanes.csv'
## .... ........ .................. ......... .................
## See problems(...) for more details.

Scatterplot

Make a scatterplot that shows weight on the x axis and height on the y axis.

ggplot(data = nhanes,
       mapping = aes(x = weight,
                     y = height)) +
  geom_point()
## Warning: Removed 366 rows containing missing values (geom_point).

Histogram

Make a histogram that shows the distribution of the weight variable.

ggplot(data = nhanes, 
       mapping = aes(x = weight)) +
  geom_histogram()
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
## Warning: Removed 78 rows containing non-finite values (stat_bin).

Copy your code from above, but adjust it so that there are 50 bins.

ggplot(data = nhanes, 
       mapping = aes(x = weight)) +
  geom_histogram(bins = 50) 
## Warning: Removed 78 rows containing non-finite values (stat_bin).

Bar Chart

Bar Chart v1

Use the v1 approach to make a bar chart that shows a count of the number of people who say they smoke. Include NA responses.

ggplot(data = nhanes, 
       mapping = aes(x = smoke_now)) +
  geom_bar()

Bar Chart v2

Create a new data frame called sleep_by_gender that shows the average amount of sleep per night that males and females report getting. Drop any NA (or NaN) responses from this data frame.

sleep_by_gender <- nhanes %>% 
  group_by(gender) %>% 
  summarize(avg_sleep = mean(sleep_hrs_night, na.rm = TRUE))

Plot the average amount of sleep per night for males and females.

ggplot(data = sleep_by_gender, 
       mapping = aes(x = gender, 
                     y = avg_sleep)) +
  geom_bar(stat = "identity")

Make the same graph as above, but use geom_col instead of geom_bar.

ggplot(data = sleep_by_gender, 
       mapping = aes(x = gender, 
                     y = avg_sleep)) +
  geom_col()

color and fill

Take your graph from above (the one with geom_col) and make the inside of each bar a different color.

ggplot(data = sleep_by_gender, 
       mapping = aes(x = gender, 
                     y = avg_sleep,
                     fill = gender)) +
  geom_col()

Make your scatterplot from before that shows weight on the x axis and height on the y axis again, but make the dots show up in different colors based on the phys_active variable.

ggplot(data = nhanes,
       mapping = aes(x = weight,
                     y = height,
                     color = phys_active)) +
  geom_point()
## Warning: Removed 366 rows containing missing values (geom_point).

Scales

color

Take your scatterplot that you just made and add a scale using scale_color_brewer. Take a look at the help docs and choose a palette other than the default (hint: look at the palette argument).

ggplot(data = nhanes,
       mapping = aes(x = weight,
                     y = height,
                     color = phys_active)) +
  geom_point() +
  scale_color_brewer(palette = "Dark2")
## Warning: Removed 1750 rows containing missing values (geom_point).

Do nearly the same thing to change the color of the last bar chart you made (the one about sleep and gender).

ggplot(data = sleep_by_gender, 
       mapping = aes(x = gender, 
                     y = avg_sleep,
                     fill = gender)) +
  geom_col() +
  scale_fill_brewer(palette = "Dark2")

x and y axes

Copy the graph you just made and change the y axis so it goes from 0 to 8.

ggplot(data = sleep_by_gender, 
       mapping = aes(x = gender, 
                     y = avg_sleep,
                     fill = gender)) +
  geom_col() +
  scale_fill_brewer(palette = "Dark2") +
  scale_y_continuous(limits = c(0, 8))

Copy the last code chunk. Then adjust the breaks on the y axis so that it shows 0 to 8 by 1 (i.e. 0, 1, 2, etc).

ggplot(data = sleep_by_gender, 
       mapping = aes(x = gender, 
                     y = avg_sleep,
                     fill = gender)) +
  geom_col() +
  scale_fill_brewer(palette = "Dark2") +
  scale_y_continuous(limits = c(0, 8),
                     breaks = c(0, 1, 2, 3, 4, 5, 6, 7, 8))

Text and Labels

Copy your last code chunk. Then do the following:

  1. Add text labels to each bar.
  2. Use the round argument to just show one decimal place in each label.
  3. Use the vjust argument to have them show up at the inner edge of the bars.
  4. Make the labels all white.
ggplot(data = sleep_by_gender, 
       mapping = aes(x = gender, 
                     y = avg_sleep,
                     fill = gender)) +
  geom_col() +
  geom_text(aes(label = round(avg_sleep, 1)),
            vjust = 1.5,
            color = "white") +
  scale_fill_brewer(palette = "Dark2",
                    na.value = "blue") +
  scale_y_continuous(limits = c(0, 8),
                     breaks = c(0, 1, 2, 3, 4, 5, 6, 7, 8))

Do the same thing as above, but use geom_label instead of geom_text. Also, do the following:

  1. Use the vjust argument to have them show up at the outer edge of the bars.
  2. Don’t show the legend (hint: you’ll have to add code in two places to make this happen).
ggplot(data = sleep_by_gender, 
       mapping = aes(x = gender, 
                     y = avg_sleep,
                     fill = gender)) +
  geom_col(show.legend = FALSE) +
  geom_label(aes(label = round(avg_sleep, 1)),
            vjust = -1.1,
            show.legend = FALSE,
            color = "white") +
  scale_fill_brewer(palette = "Dark2",
                    na.value = "blue") +
  scale_y_continuous(limits = c(0, 8),
                     breaks = c(0, 1, 2, 3, 4, 5, 6, 7, 8))

Plot Labels

Use the code chunk from above with geom_text (not the last one with geom_label). Do the following:

  1. Add a title
  2. Add a better y axis label
  3. Remove the x axis label
  4. Remove the legend (hint: use the show.legend argument again)
ggplot(data = sleep_by_gender, 
       mapping = aes(x = gender, 
                     y = avg_sleep,
                     fill = gender)) +
  geom_col(show.legend = FALSE) +
  geom_text(aes(label = round(avg_sleep, 1)),
            vjust = 1.5,
            color = "white") +
  scale_fill_brewer(palette = "Dark2",
                    na.value = "blue") +
  scale_y_continuous(limits = c(0, 8),
                     breaks = c(0, 1, 2, 3, 4, 5, 6, 7, 8)) +
  labs(title = "Women sleep slightly more than men on average",
       y = "Hours of sleep per night",
       x = "")

Themes

Install and load the hrbrthemes package. It’s a package that provides some great default themes.

It’s not available on CRAN, where we normally install packages from, which means you have to install it slightly differently. You’ll use the devtools package and then use this to install the hrbrthemes package. Use the code below.

# install.packages("devtools")
# devtools::install_github("hrbrmstr/hrbrthemes")
library(hrbrthemes)

Then add the theme_ipsum to your plot.

ggplot(data = sleep_by_gender, 
       mapping = aes(x = gender, 
                     y = avg_sleep,
                     fill = gender)) +
  geom_col(show.legend = FALSE) +
  geom_text(aes(label = round(avg_sleep, 1)),
            vjust = 1.5,
            color = "white") +
  scale_fill_brewer(palette = "Dark2",
                    na.value = "blue") +
  scale_y_continuous(limits = c(0, 8),
                     breaks = c(0, 1, 2, 3, 4, 5, 6, 7, 8)) +
  labs(title = "Women sleep slightly more than men on average",
       y = "Hours of sleep per night",
       x = "") +
  theme_ipsum()

Facets

I’ve created a data frame called sleep_by_gender_by_age for you. Run the code chunk below to load the data frame.

sleep_by_gender_by_age <- nhanes %>% 
  group_by(gender, age_decade) %>% 
  summarize(avg_sleep = mean(sleep_hrs_night, na.rm = TRUE)) %>% 
  drop_na()

Let’s take a look at sleep_by_gender_by_age.

sleep_by_gender_by_age

Now, see if you can recreate this plot. Much of the code will be the same from your previous plots using the sleep_by_gender data frame so just make some small changes.

ggplot(data = sleep_by_gender_by_age, 
       aes(x = age_decade,
           y = avg_sleep,
           fill = gender)) +
  geom_col(show.legend = FALSE) +
  geom_text(aes(label = round(avg_sleep, 1)),
            vjust = 1.5,
            color = "white") +
  scale_y_continuous(limits = c(0, 8),
                     breaks = seq(0, 8, by = 1)) +
  scale_fill_brewer(palette = "Dark2") +
  facet_wrap(~gender) +
  theme_ipsum() +
  labs(title = "Sleep by gender and age",
       y = "Hours of sleep per night",
       x = "")

Save Plots

Save your last plot to a PNG that is 8 inches wide and 5 inches high. Put it in the plots directory and call it “my-sleep-plot.png”

ggsave(filename = "plots/my-sleep-plot.png",
       height = 5,
       width = 8,
       unit = "in")