CODEWITHSUNDEEP
r
soosus
soosus

Reputation: 1217

R: Storing counts of occurences of strings from one column in a new column

Say I have a dataframe, df, that looks like this:

              timestamp           residence
2014/01/29 10:46:46 PM EST        Virginia, USA
2014/01/29 10:51:01 PM EST        Maryland, USA
2014/01/29 10:54:08 PM EST   Massachusetts, USA
2014/01/29 10:55:00 PM EST         Indiana, USA
2014/01/29 11:02:31 PM EST        Michigan, USA
2014/01/29 11:19:42 PM EST        Virginia, USA

Now I want to take this and create a new dataframe, df.count, which contains one column listing every string found under in df$residence uniquely (once) and a second column listing counts for the number of occurrences of each string in df$residence. This is similar to

table(df$residence)

but the output format would instead look like:

         residence     count
     Virginia, USA     2
     Maryland, USA     1
Massachusetts, USA     1
      Indiana, USA     1
     Michigan, USA     1

Upvotes: 0

Views: 105

Answers (3)

Sven Hohenstein
Sven Hohenstein

Reputation: 81693

Another solution with aggregate:

setNames(aggregate(seq(nrow(df)) ~ residence,df, length), c("residence","count"))

           residence count
1       Indiana, USA     1
2      Maryland, USA     1
3 Massachusetts, USA     1
4      Michigan, USA     1
5      Virginia, USA     2

Upvotes: 1

josliber
josliber

Reputation: 44320

I suppose you could use table to build this new data frame:

tab <- table(df$residence)
data.frame(residence=names(tab), count=as.vector(tab))
#            residence count
# 1       Indiana, USA     1
# 2      Maryland, USA     1
# 3 Massachusetts, USA     1
# 4      Michigan, USA     1
# 5      Virginia, USA     2

Upvotes: 2

BrodieG
BrodieG

Reputation: 52637

If you're okay with residence as the names only:

with(df, data.frame(count=tapply(residence, residence, length)))

If you want an actual column with residence:

with(df, {
  summ <-tapply(residence, residence, length)
  data.frame(residence=names(summ), count=summ)
} )

Upvotes: 1

Related Questions