如何跨多个列应用 ifelse function 并在 R 中创建新列

Question

我想在我的数据集的多个列中应用 ifelse function 并创建新的“重新评分”列。 这是一个示例数据集：

data = data.frame(year = "2021",
                  month = sample(x = c(1:12), size = 10, replace = TRUE),
                  C1 = sample(x = c('Off', 'Yes'), size = 10, replace = TRUE),
                  C2 = sample(x = c('Off', 'Yes'), size = 10, replace = TRUE),
                  C3 = sample(x = c('Off', 'Yes'), size = 10, replace = TRUE),
                  C4 = sample(x = c('Off', 'Yes'), size = 10, replace = TRUE),
                  C5 = sample(x = c('Off', 'Yes'), size = 10, replace = TRUE),
                  C6 = sample(x = c('Off', 'Yes'), size = 10, replace = TRUE),
                  C7 = sample(x = c('Off', 'Yes'), size = 10, replace = TRUE),
                  C8 = sample(x = c('Off', 'Yes'), size = 10, replace = TRUE),
                  C9 = sample(x = c('Off', 'Yes'), size = 10, replace = TRUE),
                  C10 = sample(x = c('Off', 'Yes'), size = 10, replace = TRUE))

我想在以 C 开头的所有行中应用这样的 function：

rescored = data %>%
  mutate(T1 = ifelse(C1 == "Off", 1,
                     ifelse(C1 == "Yes", 0, NA)))

我的真实数据集有 50 行或更多行需要应用此 function。 有没有一种简单的方法可以做到这一点？ 我已经尝试在 dplyr 中使用“交叉”的变体，如下所示，但没有成功。 我敢肯定还有一个“应用”选项。

rescored = data %>%
  mutate(across(C1:C50, ifelse(~ .x == "Off", 1,
                               ifelse(~.x == "Yes", 0, NA))))

Answer 1

以下选项似乎有效。 我不确定是否有更优雅的方式来做到这一点。 在第二步中重命名变量似乎并不理想。

rescore <- function(x, na.rm = FALSE) (ifelse(x == "Off", 1, ifelse(x == "Yes", 0, NA)))
                                       
data %>% 
  mutate_at(c(as.vector(paste0("C", 1:50))), funs(scr = rescore)) %>%
  rename_at(vars(ends_with("_scr")), funs(paste("scr", 1:50, sep = "_")))

Answer 2

只需这样做（您必须在 function 语句的开头使用twiddle ~ ，而不是在每个参数之前。）

data %>%
  mutate(across(starts_with('C'), ~ifelse( .x == "Off", 1,
                               ifelse(.x == "Yes", 0, NA))))

   year month C1 C2 C3 C4 C5 C6 C7 C8 C9 C10
1  2021     1  1  0  0  1  1  0  0  1  1   1
2  2021    12  1  1  0  0  1  1  1  0  1   0
3  2021    10  1  0  1  0  0  1  0  0  1   1
4  2021     3  0  1  1  1  0  1  0  0  0   1
5  2021    11  1  0  1  1  1  0  1  0  0   0
6  2021    12  1  0  0  1  1  1  0  0  1   0
7  2021     4  0  0  0  1  1  0  1  0  1   0
8  2021     2  0  0  0  1  0  0  0  0  1   0
9  2021     3  0  0  1  0  0  1  0  0  1   0
10 2021     9  1  0  0  0  0  0  1  0  0   0

或者这个，如果你想保留原始列

data %>%
  mutate(across(starts_with('C'), ~ifelse( .x == "Off", 1, 0), .names = 'scr_{sub("C", "", .col)}'))
#>    year month  C1  C2  C3  C4  C5  C6  C7  C8  C9 C10 scr_1 scr_2 scr_3 scr_4
#> 1  2021     7 Yes Yes Yes Off Yes Off Off Yes Yes Yes     0     0     0     1
#> 2  2021    11 Off Yes Yes Yes Yes Yes Off Yes Yes Yes     1     0     0     0
#> 3  2021     1 Yes Yes Off Off Yes Yes Yes Off Yes Yes     0     0     1     1
#> 4  2021     5 Yes Off Off Yes Yes Yes Yes Off Yes Yes     0     1     1     0
#> 5  2021     6 Off Off Yes Yes Off Off Off Yes Off Yes     1     1     0     0
#> 6  2021    12 Yes Yes Yes Off Off Yes Yes Yes Off Yes     0     0     0     1
#> 7  2021     1 Off Off Off Off Yes Off Off Off Yes Yes     1     1     1     1
#> 8  2021     1 Yes Yes Yes Off Off Yes Yes Off Off Yes     0     0     0     1
#> 9  2021     8 Off Yes Off Yes Off Off Yes Yes Yes Yes     1     0     1     0
#> 10 2021    10 Off Yes Off Yes Yes Off Off Yes Off Off     1     0     1     0
#>    scr_5 scr_6 scr_7 scr_8 scr_9 scr_10
#> 1      0     1     1     0     0      0
#> 2      0     0     1     0     0      0
#> 3      0     0     0     1     0      0
#> 4      0     0     0     1     0      0
#> 5      1     1     1     0     1      0
#> 6      1     0     0     0     1      0
#> 7      0     1     1     1     0      0
#> 8      1     0     0     1     1      0
#> 9      1     1     0     0     0      0
#> 10     0     1     1     0     1      1

^{由代表 package (v2.0.0) 于 2021 年 5 月 15 日创建}