Structure R Dataframe from Long to Wide

Question

I have a dateframe as such:

long <- data.frame(subj = c(1,1,2,2,2), code = c(\"a\", \"b\", \"a\", \"d\", \"e\"))

   subj code
 1    1    a
 2    1    b
          


        

Answer 1

We create a sequence column for the grouping column 'subj' and then do dcast. We can use dcast from data.table. Convert the 'data.frame' to 'data.table' (setDT(long)), grouped by 'subj', create the sequence column 'new' and reshape from 'long' to 'wide' with dcast.

library(data.table)#v1.9.6+
setDT(long)[, new:=paste('code', 1:.N, sep='.'), by = subj]
dcast(long, subj~new, value.var='code')

---

Or this can be done with spread from tidyr after creating the sequence column using dplyr methods

library(dplyr)
library(tidyr)
long %>% 
    group_by(subj) %>% 
    mutate(new=paste('code', row_number(), sep='.')) %>% 
    spread(new, code)