R-根据一列中跨不同列的公共值，将data.frame格式化为另一个“组合” data.frame

Question

I'm starting with a data frame that consists of three columns. Column#1 contains ids that indicate 3 different time periods when the weight (column#3) of some persons (column#2) has been measured in kg.

All persons have been measured irregularly, which means, that some persons are measured multiple times or just once within a time period but not across all time periods.

   id       person_name person_weight
    1          Carol         51
    1          Mike          76
    1          Mike          81
    1          Dave          66
    1          Carol         59
    2          James         78
    2          Simone        55
    2          Simone        49
    2          David         85
    3          Mike          93
    3          Dave          110
    3          Dave          98 

Actually, the whole thing here is just a simplified example.. so dont bother if this kind of data collections makes no sense.

Now, I want to calculate the average (mean) weight for each person within a time period and then end up with a combined data frame that looks like the following one:

group_id    Carol   Mike    Dave    James   Simone  David
   1         55     78.5     66      NA       NA     NA
   2         NA      NA      NA      78       52     85
   3         NA      93      104     NA       NA     NA

I tried some basic R functions (table, apply etc.) but couldn't deal with the dependence across the columns.

Thanks in advance for any help that brings me closer to the second/'combined' dataframe.

Answer 1

Seems like a simple dcast :

library(reshape2)
dcast(dat,id ~person_name,
      fun.aggregate = mean,
      value.var = "person_weight",fill = NA_real_)
  id Carol Dave David James Mike Simone
1  1    55   66    NA    NA 78.5     NA
2  2    NA   NA    85    78   NA     52
3  3    NA  104    NA    NA 93.0     NA