將每個變量觀察的多行“轉換”為 R 中的單列

Question

假設我有一個看起來像這樣的數據框（實際上，它是一個更大的數據框，包含數百個變量和數千個觀察值）。

  VAR   Country  Year     Value
1 VAR 1 France   2014     1
2 VAR 1 France   2018     2
3 VAR 1 UK       2014     5
4 VAR 1 UK       2018     6
5 VAR 2 France   2014     2
6 VAR 2 France   2018     3
7 VAR 2 UK       2014     7
8 VAR 2 UK       2018     8

我需要將每個變量的多行“轉換”為列並匹配相應的值，所以它看起來像這樣：

  Country  Year   VAR 1   VAR 2
1 France   2014   1       2
2 France   2018   2       3
3 UK       2014   5       7
4 UK       2018   6       8

同樣，在實踐中假設有更多的變量和觀察結果。 任何人都知道一個簡單的代碼片段來使這變得容易？ 謝謝。

Answer 1

您可以使用data.table dcast ：

library(data.table)
rm(list = ls())

dt <- data.table(
  VAR = c('VAR 1', 'VAR 1', 'VAR 1','VAR 1', 'VAR 2', 'VAR 2', 'VAR 2', 'VAR 2'),
  Country = c('France', 'France', 'UK', 'UK', 'France', 'France', 'UK', 'UK'),
  Year = c(2014, 2018, 2014, 2018, 2014, 2018, 2014, 2018),
  Value = c(1, 2, 5, 6, 2, 3, 7, 8)
)

dt.wide <- dcast(
  dt,
  Country + Year ~ VAR, value.var = 'Value'
)


dt.wide

> dt.wide
   Country Year VAR 1 VAR 2
1:  France 2014     1     2
2:  France 2018     2     3
3:      UK 2014     5     7
4:      UK 2018     6     8

Answer 2

這將是pivot_wider一個選項

library(dplyr)
library(tidyr)
df1 %>%
     pivot_wider(names_from = VAR, values_from = Value)
# A tibble: 4 x 4
#  Country  Year `VAR 1` `VAR 2`
#  <chr>   <int>   <int>   <int>
#1 France   2014       1       2
#2 France   2018       2       3
#3 UK       2014       5       7
#4 UK       2018       6       8

數據

df1 <- structure(list(VAR = c("VAR 1", "VAR 1", "VAR 1", "VAR 1", "VAR 2", 
"VAR 2", "VAR 2", "VAR 2"), Country = c("France", "France", "UK", 
"UK", "France", "France", "UK", "UK"), Year = c(2014L, 2018L, 
2014L, 2018L, 2014L, 2018L, 2014L, 2018L), Value = c(1L, 2L, 
5L, 6L, 2L, 3L, 7L, 8L)), class = "data.frame", row.names = c("1", 
"2", "3", "4", "5", "6", "7", "8"))