R: How to convert categories to columns?

Question

I have data like this:

Id Action timestamp
1   click    #########
1    view    #########
1    data    #########
2    click   #########
2   click    #########

I want to convert the categories into columns containing frequency(counts) like this:

id click view data
1    1    1     1
2    2    0     0

How can I do this? Thanks !

Quickest way is to use `dcast` from either "data.table" or "reshape2". Lots of answers on SO with such solutions. — A5C1D2H2I1M1N2O1R2T1, Jan 01 '16 at 16:44
`table(df$Id, df$Action)` from base package will do the same — DrDom, Jan 01 '16 at 17:07

score 7 · Accepted Answer · answered Jan 01 '16 at 17:28

7

Here is how you can do it.

# create the data frame
df <- data.frame(Id=c(1,1,1,2,2), Action=c("click", "view", "data", "click", "click"))
df
#>   Id Action
#> 1  1  click
#> 2  1   view
#> 3  1   data
#> 4  2  click
#> 5  2  click

# Use reshape2::dcast
library(reshape2)
dcast(df, Id ~ Action, fun.aggregate = length)
#>   Id click data view
#> 1  1     1    1    1
#> 2  2     2    0    0

answered Jan 01 '16 at 17:28

Selva

2,045
1
23
18

1

If you'd like to save keystrokes the function defaults to length `dcast(df, Id~Action)` – Pierre L Jan 01 '16 at 17:37

R: How to convert categories to columns?

1 Answers1