需要帮助在 R 中重塑数据集

Question

我有一个如下所示的数据集：

Name. Position number.student? Married?

Bob. 0001. YES. NO.
Susan. 0002. YES. YES.
Mark. 0003. NO.NO
Becky.0004.NO.YES
Billy.0005.YES.YES

我需要它看起来像这样：

Bob. 0001. YES. NO.
Susan. 0002. YES. NO.
Susan.0002.NO.YES
Mark. 0003. NO.NO
Becky.0004.NO.YES
Billy.0005.YES.NO
Billy.0005.NO.YES

换句话说，我有多个以“是”和“否”作为值的列标题。我需要为每个“是”分配一行，每个人都有必要的行数，具体取决于他们有多少“是”值。

如何在 R 中完成此操作？

Answer 1

一种方法是获取长格式数据，仅保留"YES" 行，为每个Name 创建一个序列列，并获取宽格式数据，用"NO" 填充空值。

library(dplyr)
library(tidyr)

df %>%
  pivot_longer(cols = student:Married) %>%
  filter(value == 'YES') %>%
  group_by(Name) %>%
  mutate(row = row_number()) %>%
  pivot_wider(values_fill = "NO") %>%
  select(-row)


#  Name  Positionnumber student Married
#  <chr>          <int> <chr>   <chr>  
#1 Bob                1 YES     NO     
#2 Susan              2 YES     NO     
#3 Susan              2 NO      YES    
#4 Becky              4 NO      YES    
#5 Billy              5 YES     NO     
#6 Billy              5 NO      YES    

然而，这会删除Name 具有两个"NO"s 的值，这是@Walker Harrison 建议的一个细微变化。

df %>%   
  pivot_longer(cols = student:Married) %>%   
  arrange(Name, name) %>%   
  group_by(Name) %>%   
  filter(value == 'YES' | (name == "student" & 
         value == 'NO' & lag(value) == 'NO')) %>%   
  mutate(row = row_number()) %>%   
  pivot_wider(values_fill = "NO") %>%   
  select(-row)


# Name  Positionnumber Married student
#  <chr>          <int> <chr>   <chr>  
#1 Becky              4 YES     NO     
#2 Billy              5 YES     NO     
#3 Billy              5 NO      YES    
#4 Bob                1 NO      YES    
#5 Mark               3 NO      NO     
#6 Susan              2 YES     NO     
#7 Susan              2 NO      YES    

数据

df <- structure(list(Name = c("Bob", "Susan", "Mark", "Becky", "Billy"
), Positionnumber = 1:5, student = c("YES", "YES", "NO", "NO", 
"YES"), Married = c("NO", "YES", "NO", "YES", "YES")), class = 
"data.frame", row.names = c(NA, -5L))