在 R 中的多列上转换从长到宽的数据库传播数据

Question

我正在努力将长数据框转换为宽数据框，但有些复杂。我有一列 ID，其中有重复的条目 - 每个条目都指的是个人的时间点。我还有其他列（即 visit、var1 和 var2），其中报告了有关每个时间点的数据。这是一个可重现的例子：

df <- data.frame(ID=c(1,1,1,1,2,2,2,3,3,3),
                 visit=c(1,4,5,7,1,3,4,2,5,6),
                 var1=c("AF","no","no","no","AG","AG","no","BA","BA","BA"),
                 var2=c("good","good","good","bad","good","good","bad","good","good","good"))

并且输出：

   ID visit var1 var2
1   1     1   AF good
2   1     4   no good
3   1     5   no good
4   1     7   no  bad
5   2     1   AG good
6   2     3   AG good
7   2     4   no  bad
8   3     2   BA good
9   3     5   BA good
10  3     6   BA good

我真的需要得到一个数据框，其中每个 ID 只包含一行，每个其他变量包含多个列，例如带有数字后缀（例如 visit_1 , visit_2, visit_3 等）。

我想到的输出是这样的：

  ID visit_1 visit_2 visit_3 visit_4 var1_1 var1_2 var1_3 var1_4 var2_1 var2_2 var2_3 var2_4
1  1       1       4       5       7     AF     no     no     no   good   good   good    bad
2  2       1       3       4      NA     AG     AG     no   <NA>   good   good    bad   <NA>
3  3       2       5       6      NA     BA     BA     BA   <NA>   good   good   good   <NA>.

其中 visit、var1 和 var2 列的每个条目基本上都放置在基于 ID 列的单独的顺序列中。

我试过 data.frame::dcast 和 tidyr::spread，还有 pivot_wider()，但看起来这些公式最终会根据实际的 值得到多列 而不是生成固定的列。例如，使用 pivot_wider():

df %>% pivot_wider(names_from = ID, values_from = c("visit","var1","var2"))

它 returns 我说了一个错误

Warning messages:
1: Values are not uniquely identified; output will contain list-cols.
* Use `values_fn = list` to suppress this warning.
* Use `values_fn = length` to identify where the duplicates arise
* Use `values_fn = {summary_fun}` to summarise duplicates 
2: Values are not uniquely identified; output will contain list-cols.
* Use `values_fn = list` to suppress this warning.
* Use `values_fn = length` to identify where the duplicates arise
* Use `values_fn = {summary_fun}` to summarise duplicates 
3: Values are not uniquely identified; output will contain list-cols.
* Use `values_fn = list` to suppress this warning.
* Use `values_fn = length` to identify where the duplicates arise
* Use `values_fn = {summary_fun}` to summarise duplicates

有人可以帮忙吗？

Answer 1

您必须为每个组添加一个顺序 ID：

library(tidyverse)
df %>% 
  group_by(ID) %>% 
  mutate(count = row_number()) %>% 
  pivot_wider(ID, names_from = count, values_from = c(visit, var1, var2))

# A tibble: 3 x 13
# Groups:   ID [3]
#     ID visit_1 visit_2 visit_3 visit_4 var1_1 var1_2 var1_3 var1_4 var2_1 var2_2 var2_3 var2_4
#  <dbl>   <dbl>   <dbl>   <dbl>   <dbl> <chr>  <chr>  <chr>  <chr>  <chr>  <chr>  <chr>  <chr> 
#1     1       1       4       5       7 AF     no     no     no     good   good   good   bad   
#2     2       1       3       4      NA AG     AG     no     NA     good   good   bad    NA    
#3     3       2       5       6      NA BA     BA     BA     NA     good   good   good   NA

在 R 中的多列上转换从长到宽的数据库传播数据

Transform long-to-wide database spreading data on multiple columns in R

r

dataframe

reshape