Question

I’ve been searching for a command in R that will allow me to group by just a portion of what is in a field, rather than the entire field. I came up with a work around that works but it is a little cumbersome and clumsy. Here is a test data frame

name.list = data.frame(Name=c("jeff banks", "phil lender", "jeff brooks",
"barbara holcomb", "danny jefferson"),Age=c(27,34,25,45,32))
name.list

this is the output

             Name Age
1      jeff banks  27
2     phil lender  34
3     jeff brooks  25
4 barbara holcomb  45
5 danny jefferson  32

I would like to identify all the Name entires that have “jeff” in them
so I can use that as a group or assign a dummy variable. In other words, append
to my data frame something like this:

             Name Age   Jeff.field
1      jeff banks  27       1
2     phil lender  34       0
3     jeff brooks  25       1
4 barbara holcomb  45       0
5 danny jefferson  32       1

I came up with this solution but it is not very elegant

name.list2=name.list[grep("jeff",name.list$Name),]
name.list2$jeff.field=rep(1,dim(name.list2)[1])

name.list3=name.list[-grep("jeff",name.list$Name),]
name.list3$jeff.field=rep(0,dim(name.list3)[1])

name.list4=rbind(name.list2,name.list3)
name.list4

This gets me this data frame

             Name Age jeff.field
1      jeff banks  27          1
3     jeff brooks  25          1
5 danny jefferson  32          1
2     phil lender  34          0
4 barbara holcomb  45          0

Does anyone know of a more basic approach?

Answer 1

Here you go:

name.list$jeff.field <- grepl("jeff", name.list$Name)
name.list[order(name.list$jeff.field), ]

             Name Age jeff.field
2     phil lender  34      FALSE
4 barbara holcomb  45      FALSE
1      jeff banks  27       TRUE
3     jeff brooks  25       TRUE
5 danny jefferson  32       TRUE