filter(.data, ...)

# Without %>%:
filter(mtcars, disp > 350 & mpg < 11)

# with %>%
mtcars %>%
  filter(disp > 350 & mpg < 11)

Note: You don’t have to provide the name of a data object within the dplyr function!

not: filter(mtcars, mtcars$disp > 350 & mtcars$mpg < 11)

but: filter(mtcars, disp > 350 & mpg < 11)

• Take the starwars database.
• Filter all cases with brown hair color and female gender (Hint: variables hair_color and sex)

library(tidyverse)
starwars %>% 
  filter(hair_color == "brown" & sex == "female")

select(.data, ...)

# Either column numbers or column names
mtcars %>% select(1, 3, carb)
# Use the `:` symbol for ranges
mtcars %>% select(1:3, am:carb)
# use `-` to drop variables and keep the others
mtcars %>% select(-am, -(1:3), -(disp:drat))

starts_with() and ends_with() can be used to select variables by parts of their names:

mtcars %>% select(starts_with("d"))
mtcars %>% select(-ends_with("b"))

• Take the starwars database.
• Filter for Human and select the variables name, height, and mass (Hint: variable species)

starwars %>% 
  filter(species == "Human") %>% 
  select(name:mass)

The everything() function helps to rearrange the columns of a data frame

starwars %>% 
  select(name, birth_year, sex, everything())

mutate(.data, ...)

mtcars %>%
  mutate(
    efficiency = mpg / disp,
    effect = round(hp / mpg * 100)
  )

• Take the starwars database.
• Create a new variable with the body mass index bmi. Hint: bmi = kg / m²
• Filter all humans with an bmi >= 26.
• Select name, bmi, height, and mass

starwars %>%
  mutate(bmi = mass/ (height/100)^2) %>%
  filter(bmi >= 26 & species == "Human") %>%
  select(name, bmi, height, mass)

mtcars %>%
  summarise(
    mean_mpg = mean(mpg),
    sd_mpg = sd(mpg),
    n = n()
  )

starwars %>%
  filter(species == "Human") %>%
  summarise(
    mean_height = mean(height, na.rm = TRUE),
    median_height = median(height, na.rm = TRUE),
    mean_mass = mean(mass, na.rm = TRUE),
    median_mass = median(mass, na.rm = TRUE)
  )

mtcars %>% group_by(cyl) %>%
  summarise(
    mean_mpg = mean(mpg),
    sd_mpg = sd(mpg),
    n = n()
  )

mtcars %>% group_by(cyl, am) %>%
  summarise(
    mean_mpg = mean(mpg),
    sd_mpg = sd(mpg),
    n = n()
  )

starwars %>%
  group_by(sex) %>%
  summarise(
    median_height = median(height, na.rm = TRUE),
    median_mass = median(mass, na.rm = TRUE),
    n = n()
  )

arrange(.data, ...)

dat <- data.frame(age = c(5,5,6,7,6), sen = c(0, 1, 1, 0, 0), points = c(34, 55, 22, 11, 9))
dat %>%
  arrange(sen, age) 

dat %>%
  arrange(sen, desc(points))

• Take the starwars database.
• Calculate the bmi.
• Summarize the median of the bmi grouped by species.
• round the bmi to one decimal.
• Also calculate the n for each group.
• Only show cases with n > 2.
• Arrange the resulting table. Sort by n in descending order.

starwars %>% 
  mutate(bmi = mass / (height / 100)^2) %>% 
  group_by(species) %>% 
  summarise(
    mean_bmi = median(bmi, na.rm = TRUE) %>% round(1), 
    n = n()
  ) %>%
  filter(n >2) %>%
  arrange(desc(n))