Loop through one column with unique values, and group by another to calculate the variance-CodePudding

Can someone please help me with this. I have a data frame as the one below:

 Scores    ID     Value
 M1        A       5.67
 M1        A       9.99
 M2        A       10.96
 M1        A       7.89
 M1        B       9.36
 M3        A       4.56
 M2        A       5.55
 M1        B       8.97

In this data set, the scores are cell types and the IDs repeat for each cell type (i.e M1 has 3 As) I want to loop through each score (cell types) and then calculate the variance for that score for each individual repeating ID (in this case A). So essentially I am measuring the variability for each score within each ID (not between).

Below is the code I thought of but, all the values in the .csv file have the same output.

 for (i in df1$scores) {

     T1 <- aggregate(value ~ ID, df1, function(x) c(Var = var(x), Count = length(x)))

     T1

     write.csv(T1,file=paste0(i,"_withinID.csv"))

 }

Thanks in advance!

CodePudding user response：

Here we need to loop over the unique values of the 'scores' and the aggregate should be based on the subset of the data

for(i in unique(df1$scores)) {
    T1 <- aggregate(value ~ ID, subset(df1, scores == i), 
         function(x) c(Var = var(x), Count = length(x)))
   write.csv(T1, file = paste0(i, "_withinID.csv"))
}