dplyr::mutatelibrary(gapminder)
library(dplyr)
Attaching package: 'dplyr'The following objects are masked from 'package:stats':
filter, lagThe following objects are masked from 'package:base':
intersect, setdiff, setequal, unionlibrary(magrittr)
americas_2002 <- gapminder %>% filter(continent == "Americas", year == 2002)
gapminder_2002 <- gapminder %>% filter(year == 2002) %>% select(!c(year))mutateworks in two contexts:
adds a new column
edits an already existing column
can add/edit several columns at once
uses its dedicated helper functions
across(), c_across()one more grouping function: rowwise()
mutateamericas_2002 %>%
mutate(GDP_millions = pop * gdpPercap/1000000 ) %>%
slice(1:10)# A tibble: 10 × 7
country continent year lifeExp pop gdpPercap GDP_millions
<fct> <fct> <int> <dbl> <int> <dbl> <dbl>
1 Argentina Americas 2002 74.3 38331121 8798. 337223.
2 Bolivia Americas 2002 63.9 8445134 3413. 28825.
3 Brazil Americas 2002 71.0 179914212 8131. 1462921.
4 Canada Americas 2002 79.8 31902268 33329. 1063270.
5 Chile Americas 2002 77.9 15497046 10779. 167039.
6 Colombia Americas 2002 71.7 41008227 5755. 236013.
7 Costa Rica Americas 2002 78.1 3834934 7723. 29619.
8 Cuba Americas 2002 77.2 11226999 6341. 71186.
9 Dominican Republic Americas 2002 70.8 8650322 4564. 39478.
10 Ecuador Americas 2002 74.2 12921234 5773. 74595.americas_2002 %>%
mutate(richest = max(gdpPercap),
perc_richest = round(gdpPercap/richest, 2)) # A tibble: 25 × 8
country continent year lifeExp pop gdpPercap richest perc_richest
<fct> <fct> <int> <dbl> <int> <dbl> <dbl> <dbl>
1 Argentina Americas 2002 74.3 3.83e7 8798. 39097. 0.23
2 Bolivia Americas 2002 63.9 8.45e6 3413. 39097. 0.09
3 Brazil Americas 2002 71.0 1.80e8 8131. 39097. 0.21
4 Canada Americas 2002 79.8 3.19e7 33329. 39097. 0.85
5 Chile Americas 2002 77.9 1.55e7 10779. 39097. 0.28
6 Colombia Americas 2002 71.7 4.10e7 5755. 39097. 0.15
7 Costa Rica Americas 2002 78.1 3.83e6 7723. 39097. 0.2
8 Cuba Americas 2002 77.2 1.12e7 6341. 39097. 0.16
9 Dominican Repu… Americas 2002 70.8 8.65e6 4564. 39097. 0.12
10 Ecuador Americas 2002 74.2 1.29e7 5773. 39097. 0.15
# ℹ 15 more rowsNotice how you can concatenate new columns in one mutate action and that the second uses the result of the first within the same command. But I am not sure whether this is always granted.
gapminder_2002 %>%
group_by(continent) %>%
mutate(richest = max(gdpPercap),
perc_richest = round(gdpPercap/richest, 2)) %>%
ungroup()# A tibble: 142 × 7
country continent lifeExp pop gdpPercap richest perc_richest
<fct> <fct> <dbl> <int> <dbl> <dbl> <dbl>
1 Afghanistan Asia 42.1 25268405 727. 36023. 0.02
2 Albania Europe 75.7 3508512 4604. 44684. 0.1
3 Algeria Africa 71.0 31287142 5288. 12522. 0.42
4 Angola Africa 41.0 10866106 2773. 12522. 0.22
5 Argentina Americas 74.3 38331121 8798. 39097. 0.23
6 Australia Oceania 80.4 19546792 30688. 30688. 1
7 Austria Europe 79.0 8148312 32418. 44684. 0.73
8 Bahrain Asia 74.8 656397 23404. 36023. 0.65
9 Bangladesh Asia 62.0 135656790 1136. 36023. 0.03
10 Belgium Europe 78.3 10311970 30486. 44684. 0.68
# ℹ 132 more rowsHere we perform the same operation as with Americas, but this time for each continent. The richest column has identical values for countries on one continent.
Note the difference in the behavior of max() and division (/). The max function always looks at the entire group whereas division operates on each row independently.
if_elsegapminder_2002 %>%
mutate(is_Asia = if_else(
condition = continent == "Asia",
true = "Asian country",
false = "Not in Asia"))# A tibble: 142 × 6
country continent lifeExp pop gdpPercap is_Asia
<fct> <fct> <dbl> <int> <dbl> <chr>
1 Afghanistan Asia 42.1 25268405 727. Asian country
2 Albania Europe 75.7 3508512 4604. Not in Asia
3 Algeria Africa 71.0 31287142 5288. Not in Asia
4 Angola Africa 41.0 10866106 2773. Not in Asia
5 Argentina Americas 74.3 38331121 8798. Not in Asia
6 Australia Oceania 80.4 19546792 30688. Not in Asia
7 Austria Europe 79.0 8148312 32418. Not in Asia
8 Bahrain Asia 74.8 656397 23404. Asian country
9 Bangladesh Asia 62.0 135656790 1136. Asian country
10 Belgium Europe 78.3 10311970 30486. Not in Asia
# ℹ 132 more rowscase_whengapminder %>% filter(year == 2002) %>%
mutate(is_Asia = case_when(
continent == "Asia" ~ "Asian country",
continent == "Africa" ~ "African country",
continent == "Europe" ~ "European country",
continent == "Americas" ~ "American country",
continent == "Oceania" ~ "Oceanian country",
.default = NA
))# A tibble: 142 × 7
country continent year lifeExp pop gdpPercap is_Asia
<fct> <fct> <int> <dbl> <int> <dbl> <chr>
1 Afghanistan Asia 2002 42.1 25268405 727. Asian country
2 Albania Europe 2002 75.7 3508512 4604. European country
3 Algeria Africa 2002 71.0 31287142 5288. African country
4 Angola Africa 2002 41.0 10866106 2773. African country
5 Argentina Americas 2002 74.3 38331121 8798. American country
6 Australia Oceania 2002 80.4 19546792 30688. Oceanian country
7 Austria Europe 2002 79.0 8148312 32418. European country
8 Bahrain Asia 2002 74.8 656397 23404. Asian country
9 Bangladesh Asia 2002 62.0 135656790 1136. Asian country
10 Belgium Europe 2002 78.3 10311970 30486. European country
# ℹ 132 more rowsNote the specific syntax of case_when!
mutate existing with acrossgapminder_2002 %>%
mutate(across(pop, ~ round(.x/1000000, 3)))# A tibble: 142 × 5
country continent lifeExp pop gdpPercap
<fct> <fct> <dbl> <dbl> <dbl>
1 Afghanistan Asia 42.1 25.3 727.
2 Albania Europe 75.7 3.51 4604.
3 Algeria Africa 71.0 31.3 5288.
4 Angola Africa 41.0 10.9 2773.
5 Argentina Americas 74.3 38.3 8798.
6 Australia Oceania 80.4 19.5 30688.
7 Austria Europe 79.0 8.15 32418.
8 Bahrain Asia 74.8 0.656 23404.
9 Bangladesh Asia 62.0 136. 1136.
10 Belgium Europe 78.3 10.3 30486.
# ℹ 132 more rowsacross lists the columns on which you want to perform the same transformation. Note its peculiar syntax: it is inside mutate and incorporates both the enumeration of the columns to transform and the transformation, which is written in the formula notation with tilde and .x for the first argument.
mutate existing on selected columnsgapminder_2002 %>% select(!c(continent)) %>%
mutate(across(
where(~ is.numeric(.x)),
~ round(.x, digits = 0)
) )# A tibble: 142 × 4
country lifeExp pop gdpPercap
<fct> <dbl> <dbl> <dbl>
1 Afghanistan 42 25268405 727
2 Albania 76 3508512 4604
3 Algeria 71 31287142 5288
4 Angola 41 10866106 2773
5 Argentina 74 38331121 8798
6 Australia 80 19546792 30688
7 Austria 79 8148312 32418
8 Bahrain 75 656397 23404
9 Bangladesh 62 135656790 1136
10 Belgium 78 10311970 30486
# ℹ 132 more rowsacross works well with all column-selecting helper functions, such as where or starts_with.
sum , mean and many other functions collapse all values in a column.
Override with rowwise() %>% mutate(c_across(….)
iris %>% slice(1:3) %>% rowwise() %>%
mutate(sum_leafmeasure = sum(c_across(where(~ is.numeric(.x))))) %>% ungroup()# A tibble: 3 × 6
Sepal.Length Sepal.Width Petal.Length Petal.Width Species sum_leafmeasure
<dbl> <dbl> <dbl> <dbl> <fct> <dbl>
1 5.1 3.5 1.4 0.2 setosa 10.2
2 4.9 3 1.4 0.2 setosa 9.5
3 4.7 3.2 1.3 0.2 setosa 9.4rowwise and c_across 😐iris %>% slice(1:3) %>%
mutate(sum_leafmeasure = sum(Sepal.Length, Sepal.Width, Petal.Length, Petal.Width)) Sepal.Length Sepal.Width Petal.Length Petal.Width Species sum_leafmeasure
1 5.1 3.5 1.4 0.2 setosa 29.1
2 4.9 3.0 1.4 0.2 setosa 29.1
3 4.7 3.2 1.3 0.2 setosa 29.1meaniris %>% slice(1:3) %>% rowwise() %>%
mutate(mean_leafmeasure = mean(c_across(!c(Species)))) %>% ungroup()# A tibble: 3 × 6
Sepal.Length Sepal.Width Petal.Length Petal.Width Species mean_leafmeasure
<dbl> <dbl> <dbl> <dbl> <fct> <dbl>
1 5.1 3.5 1.4 0.2 setosa 2.55
2 4.9 3 1.4 0.2 setosa 2.38
3 4.7 3.2 1.3 0.2 setosa 2.35rowwisesum() – sums values across columns
mean() – computes average across columns
sd() – standard deviation
var() – variance
min() / max() – minimum / maximum
median() – median value
IQR() – interquartile range