CODEWITHSUNDEEP
r
amathew
amathew

Reputation: 366

Count of Row Frequency in R

Let's say that I have the following data frame with three columns.

data = data.frame(id=c(1:10), interest_1=c("food","","","drugs","beer","soda","","","drugs","sports"),
                  interest_2=c("fruits","car","jeans","","","","soda","shoes","","drugs"),
                  interest_3=c("","","","","soda","sports","","","",""))

data

I want to get a count of each row.

The following incident, where food is interest_1, fruits is interest_2, and nothing is interest_3 occurs only once.

   id interest_1 interest_2 interest_3
1   1       food     fruits

The following incident, where drugs are interest_1 and nothing is interest_2 or interest_3 occurs twice.

 id interest_1 interest_2 interest_3
  4      drugs                      
  9      drugs

I want to get a count the number of times that each incidence occurs. How would I go about doing this?

Output should look like:

 interest_1 interest_2 interest_3   count
     food     fruits                 1
                        car          1  
              jeans                  1
     drugs                           2

Upvotes: 0

Views: 2802

Answers (2)

stanekam
stanekam

Reputation: 4030

require(plyr)
ddply(data, .(interest_1, interest_2, interest_3), c("nrow"))

Upvotes: 3

Julián Urbano
Julián Urbano

Reputation: 8488

> aggregate(id~.,data,length)
  interest_1 interest_2 interest_3 id
1      drugs                        2
2                   car             1
3     sports      drugs             1
4       food     fruits             1
5                 jeans             1
6                 shoes             1
7                  soda             1
8       beer                  soda  1
9       soda                sports  1

Basically, this means: apply function length to to the vector made up of id values for each combination of the other columns.

Upvotes: 6

Related Questions