CODEWITHSUNDEEP
rdataframe
andrewj
andrewj

Reputation: 3035

Adding a row with incomplete observations to a data.frame

Suppose I have the following data.frame:

    foo <- data.frame(ocean=c('Atlantic', 'Pacific'), 
        city=c('Boston', 'San Francisco'), 
        state=c('MA', 'CA'), stringsAsFactors=F)

I would like to add a third row, but in this case, only specify the city and leave the other fields blank.

Some naive ways, which do not work include:

foo$city[3] <- 'Providence'
# generates an error message about replacement has 3 rows, data has 2

This also generates an error message

rbind(foo, data.frame(city='Providence'))
# generates an error message, number of columns of arguments do not match

I'm interested in how other users would approach this.

Upvotes: 4

Views: 4688

Answers (2)

Wojciech Sobala
Wojciech Sobala

Reputation: 7551

foo[3,] <- NA
foo$city[3] <- 'Providence'

OR

foo[3,] <- c(NA,'Providence',NA)

Upvotes: 3

Btibert3
Btibert3

Reputation: 40126

Another approach might be to use rbind.fill in the plyr package. 

library(plyr)
bar <- data.frame(city="Providence", stringsAsFactors=F)
str(bar)

foobar <- rbind.fill(foo, bar)
str(foobar)

head(foobar)

Upvotes: 3

Related Questions