r - 如何只选择 R 中每个组的第一个非 NA 值？

Question

我有一个像这样的数据框

mydata <- data.frame(Id=c(01,01,01,01,01,01,02,02,02,02),
                     VISIT=c("Screeing","Baseline","Baseline","Baseline","Week 9","Week 9","Baseline","Week 2",
                             "Week 2","Week 2"),
                    Score=c(1,2,4,5,78,9,5,NA,3,4))

> mydata
   Id    VISIT Score
1   1 Screeing     1
2   1 Baseline     2
3   1 Baseline     4
4   1 Baseline     5
5   1   Week 9    78
6   1   Week 9     9
7   2 Baseline     5
8   2   Week 2     NA
9   2   Week 2     3
10  2   Week 2     4

我想做的是按 Id 和 VISIT 分组，然后选择每个组的第一个非 NA 值作为

> mydata

      Id VISIT    Score 
   <dbl> <fct>    <dbl> 
 1     1 Screeing     1     
 2     1 Baseline     2     
 5     1 Week 9      78    
 7     2 Baseline     5     
 9     2 Week 2       3     

这是我想到的

mydata<-mydata %>%
 group_by(Id,VISIT) %>% 
 mutate(first = dplyr::first(na.omit(Score)))

但它不会删除其他行，它只是创建一个新列，其中包含每个组的第一个非 NA 的重复值。

Answer 1

如果 base R 没问题，试试这个。默认情况下省略 NA。

aggregate( Score ~ Id + VISIT, mydata, function(x) x[1] )

  Id    VISIT Score
1  1 Baseline     2
2  2 Baseline     5
3  1 Screeing     1
4  2   Week 2     3
5  1   Week 9    78