在 R 中使用 NA 将多个变量从宽转为长

Question

我目前正在尝试使用 R 中的pivot_longer 函数转置（宽到长）多个变量。

这是我所拥有的

primary_id = c("A1", "A2")
id1 = c("B1", "B2")
score1 = c("good", "bad")
id2 = c("C1", "C2")
score2 = c(NA, "neutral")
id3 = c("D1", "D2")
score3 = c("bad", "neutral")
id4 = c("E1", "E2")
score4 = c(NA, "good")
have = data.frame(primary_id, id1, score1, id2, score2, id3, score3, id4, score4)

have
  primary_id id1 score1 id2  score2 id3  score3 id4 score4
1         A1  B1   good  C1    <NA>  D1     bad  E1   <NA>
2         A2  B2    bad  C2 neutral  D2 neutral  E2   good

请注意，我有 20 个 id 变量和 20 个 Score 变量（示例仅显示 4 个）

这就是我正在寻找的

  primary_id ids  scores
1         A1  B1    good
2         A1  C1    <NA>
3         A1  D1     bad
4         A1  E1    <NA>
5         A2  B2     bad
6         A2  C2 neutral
7         A2  D2 neutral
8         A2  E2    good

任何建议/帮助将不胜感激！请注意，分数变量到处都有 NA。

Answer 1

在基数 R 中：

\n

reshape(have,-1, dir='long', sep='', idvar = 'primary_id')\n\n     primary_id time id   score\nA1.1         A1    1 B1    good\nA2.1         A2    1 B2     bad\nA1.2         A1    2 C1    <NA>\nA2.2         A2    2 C2 neutral\nA1.3         A1    3 D1     bad\nA2.3         A2    3 D2 neutral\nA1.4         A1    4 E1    <NA>\nA2.4         A2    4 E2    good\n

\n

在数据表中：

\n

library(data.table)\nmelt(setDT(have),1, patterns(id = '^id', score = '^score'))\n   primary_id variable id   score\n1:         A1        1 B1    good\n2:         A2        1 B2     bad\n3:         A1        2 C1    <NA>\n4:         A2        2 C2 neutral\n5:         A1        3 D1     bad\n6:         A2        3 D2 neutral\n7:         A1        4 E1    <NA>\n8:         A2        4 E2    good\n

\n

在整洁的宇宙中

\n

library(tidyverse)\npivot_longer(have, -1, names_to = '.value', names_pattern = '(\\\\D+)')\n\n# A tibble: 8 \xc3\x97 3\n  primary_id id    score  \n  <chr>      <chr> <chr>  \n1 A1         B1    good   \n2 A1         C1    NA     \n3 A1         D1    bad    \n4 A1         E1    NA     \n5 A2         B2    bad    \n6 A2         C2    neutral\n7 A2         D2    neutral\n8 A2         E2    good   \n

\n