将行与多列数据R合并为一行

如何解决将行与多列数据R合并为一行

我有一个测量体重前后状况的数据集。我试图将具有相似ID的行合并为一行，并创建另外两列以显示每个ID之前和之后的结果。

原始表

ID  Sex         Race        TIME    SH           PRE_WEIGHT POST_WEIGHT
1   FEMALE      White       POST    Poor         168            152
1   FEMALE      White       PRE     Good         168            152
2   MALE        Hispanic    POST    Very Poor    157            143
2   MALE        Hispanic    PRE     Poor         157            143
3   FEMALE      Black       POST    Good         129            118
3   FEMALE      Black       PRE     Satisfactory 129            118

结果应如下所示：

ID  Sex         Race        TIME    SH          PRE_WEIGHT  POST_WEIGHT TIME_PRE    TIME_POST   SH_PRE      SH_POST
1   FEMALE      White       POST    Poor        168         152         POST        PRE         Poor        Good
2   MALE        Hispanic    POST    Very Poor   157         143         POST        PRE         Very Poor   Poor
3   FEMALE      Black       POST    Good        129         118         POST        PRE         Good        Satisfactor

我试图用很慢的重塑来完成它，但是没有用。我也无法用提迪尔重塑桌子。

希望任何人都可以帮助我。

解决方法

带有data.table

的选项

library(data.table)
dcast(setDT(df1),ID + Sex + Race  + PRE_WEIGHT + 
           POST_WEIGHT ~ TIME,value.var = 'SH')

-输出

#   ID    Sex     Race PRE_WEIGHT POST_WEIGHT      POST          PRE
#1:  1 FEMALE    White        168         152      Poor         Good
2#:  2   MALE Hispanic        157         143 Very Poor         Poor
3#:  3 FEMALE    Black        129         118      Good Satisfactory

,

一个简单的基本R选项可能正在使用reshape，例如

reshape(
  df,direction = "wide",idvar = c("ID","Sex","Race"),timevar = "TIME"
)

给出

  ID    Sex     Race   SH.POST PRE_WEIGHT.POST POST_WEIGHT.POST       SH.PRE
1  1 FEMALE    White      Poor             168              152         Good
3  2   MALE Hispanic Very Poor             157              143         Poor
5  3 FEMALE    Black      Good             129              118 Satisfactory
  PRE_WEIGHT.PRE POST_WEIGHT.PRE
1            168             152
3            157             143
5            129             118

数据

> dput(df)
structure(list(ID = c(1L,1L,2L,3L,3L),Sex = c("FEMALE","FEMALE","MALE","FEMALE"),Race = c("White","White","Hispanic","Black","Black"),TIME = c("POST","PRE","POST","PRE"),SH = c("Poor","Good","Very Poor","Poor","Satisfactory"),PRE_WEIGHT = c(168L,168L,157L,129L,129L),POST_WEIGHT = c(152L,152L,143L,118L,118L)),class = "data.frame",row.names = c(NA,-6L
))

,

正如@Duck提到的，您可以做到

library(dplyr)
library(tidyr)

df3 %>% 
   group_by(ID) %>% 
   pivot_wider(names_from = TIME,values_from = SH,names_prefix = "SH_")

#> # A tibble: 3 x 7
#> # Groups:   ID [3]
#>      ID Sex    Race     PRE_WEIGHT POST_WEIGHT SH_POST   SH_PRE      
#>   <dbl> <chr>  <chr>         <dbl>       <dbl> <chr>     <chr>       
#> 1     1 FEMALE White           168         152 Poor      Good        
#> 2     2 MALE   Hispanic        157         143 Very Poor Poor        
#> 3     3 FEMALE Black           129         118 Good      Satisfactory

您的数据

df3 <- readr::read_table("
ID  Sex         Race        TIME    SH           PRE_WEIGHT POST_WEIGHT
1   FEMALE      White       POST    Poor         168            152
1   FEMALE      White       PRE     Good         168            152
2   MALE        Hispanic    POST    Very Poor    157            143
2   MALE        Hispanic    PRE     Poor         157            143
3   FEMALE      Black       POST    Good         129            118
3   FEMALE      Black       PRE     Satisfactory 129            118")