From wide to long data

Reshaping Data with tidyr

Jeroen Boeye

Head of Machine Learning, Faktion

 

separate()

Column with multiple variables

One column per variable

 

separate_rows()

Cell with multiple values

One value per cell

Reshaping Data with tidyr

Values in column headers

nuke_df

# A tibble: 2 x 6
  country            `1945` `1946` `1948` `1949` `1951`
  <chr>               <int>  <int>  <int>  <int>  <int>
1 United States           3      2      3     NA     16
2 Russian Federation     NA     NA     NA      1      2
Reshaping Data with tidyr

Values in column headers

Variables as headers

Variables as headers tidy

Reshaping Data with tidyr

The pivot_longer() function

nuke_df %>% 
  pivot_longer(`1945`:`1951`)

# A tibble: 10 x 3
   country            name  value
   <chr>              <chr> <int>
 1 United States      1945      3
 2 United States      1946      2
 3 United States      1948      3
 4 United States      1949     NA
 5 United States      1951     16
 6 Russian Federation 1945     NA
 # ... with 4 more rows
Reshaping Data with tidyr

The pivot_longer() function

nuke_df %>% 
  pivot_longer(c(`1945`, `1946`, `1948`, `1949`, `1951`))

# A tibble: 10 x 3
   country            name  value
   <chr>              <chr> <int>
 1 United States      1945      3
 2 United States      1946      2
 3 United States      1948      3
 4 United States      1949     NA
 5 United States      1951     16
 6 Russian Federation 1945     NA
 # ... with 4 more rows
Reshaping Data with tidyr

The pivot_longer() function

nuke_df %>% 
  pivot_longer(-country)

# A tibble: 10 x 3
   country            name  value
   <chr>              <chr> <int>
 1 United States      1945      3
 2 United States      1946      2
 3 United States      1948      3
 4 United States      1949     NA
 5 United States      1951     16
 6 Russian Federation 1945     NA
 # ... with 4 more rows
Reshaping Data with tidyr

pivot_longer() arguments

nuke_df %>% 
  pivot_longer(-country, names_to = "year", values_to = "n_bombs")

# A tibble: 10 x 3
   country            year  n_bombs
   <chr>              <chr> <int>
 1 United States      1945      3
 2 United States      1946      2
 3 United States      1948      3
 4 United States      1949     NA
 5 United States      1951     16
 6 Russian Federation 1945     NA
 # ... with 4 more rows
Reshaping Data with tidyr

pivot_longer() arguments

nuke_df %>% 
  pivot_longer(
    -country, 
    names_to = "year", 
    values_to = "n_bombs", 
    values_drop_na = TRUE
  )

# A tibble: 6 x 3
  country            year  n_bombs
  <chr>              <chr>   <int>
1 United States      1945        3
2 United States      1946        2
3 United States      1948        3
4 United States      1951       16
5 Russian Federation 1949        1
6 Russian Federation 1951        2
Reshaping Data with tidyr

pivot_longer() arguments

nuke_df %>% 
  pivot_longer(
    -country, 
    names_to = "year", 
    values_to = "n_bombs", 
    values_drop_na = TRUE,
    names_transform = list(year = as.integer)
  )

# A tibble: 6 x 3
  country             year n_bombs
  <chr>              <int>   <int>
1 United States       1945       3
2 United States       1946       2
3 United States       1948       3
4 United States       1951      16
5 Russian Federation  1949       1
6 Russian Federation  1951       2
Reshaping Data with tidyr

Let's practice!

Reshaping Data with tidyr

Preparing Video For Download...