Adding character in matrix in R

Question

I import csv by.

data<-read.csv('test.csv')

and then data has value

> data
  name a b c d
1    a 1 2 1 1
2    b 2 2 2 4

now i wish to add row at the bottom of data so,I add

results <- c('aa',2,3,4,5)
data[nrow(data)+1,]<-results

But it cause me warring and character value is not inserted.

Warning message:
In `[<-.factor`(`*tmp*`, iseq, value = "aa") :
  invalid factor level, NA generated

and data has value =

> data
  name a b c d
1    a 1 2 1 1
2    b 2 2 2 4
3 <NA> 2 3 4 5

but i want to have

> data
  name a b c d
1    a 1 2 1 1
2    b 2 2 2 4
3    aa 2 3 4 5

I am new in R and have tried for the ans for many hours so any help will be appreciated.Thank you.

Answer 1

Could you please try following:(I am putting data in here you could use read.csv )

dat <- read.table(text = "name a b c d
1    a 1 2 1 1
2    b 2 2 2 4",
header = TRUE,stringsAsFactors = FALSE)
dat
results <- c('aa',2,3,4,5)
rbind(dat,results)

Output(without using stringAsFactors = FALSE ):

  name a b c d
1    a 1 2 1 1
2    b 2 2 2 4
3 <NA> 2 3 4 5
Warning message:
In `[<-.factor`(`*tmp*`, ri, value = "aa") :
  invalid factor level, NA generated

Output without warning message:

  name a b c d
1    a 1 2 1 1
2    b 2 2 2 4
3   aa 2 3 4 5

Answer 2

There are several problems here:

1. The first column is a factor so you can only add levels that that factor has.
2. If c(...) is used with a mix of characters and numerics the result is character.
3. data is a data.frame, not a matrix -- the question incorrectly states it is a matrix. (If it were a matrix then it would be a character matrix since you can't mix character and numeric in a matrix.)

To fix these problems use as.is = TRUE in read.csv and use list instead of c :

data <- read.csv("test.csv", as.is = TRUE)
data[nrow(data) + 1, ] <- list('aa',2,3,4,5)

giving:

> data
  name a b c d
1    a 1 2 1 1
2    b 2 2 2 4
3   aa 2 3 4 5

Note

The file in reproducible form is generated from:

Lines <- "name a b c d
1    a 1 2 1 1
2    b 2 2 2 4
"
temp <- read.table(text = Lines, as.is = TRUE)
write.csv(temp, "test.csv", row.names = FALSE, quote = FALSE)

Answer 3

This should work.

data <- read.csv('test.csv', stringsAsFactors = FALSE)

In a short explanation, read.csv treat every character column as a factor by default. Factor are a special class of data that can only take a few values determined at its creation.

Answer 4

Include 'aa' as one of the levels and it should work with the OP's factor column. Also, as list can hold different type instead of a vector (formed with c ), assign the next row with a list as input

levels(data$name) <- c(levels(data$name), 'aa')
data[nrow(data) +1,] <- list('aa', 2, 3, 4, 5)
data
#  name a b c d
#1    a 1 2 1 1
#2    b 2 2 2 4
#3   aa 2 3 4 5

data

data <- structure(list(name = structure(1:2, .Label = c("a", "b"),
 class = "factor"), 
a = 1:2, b = c(2L, 2L), c = 1:2, d = c(1L, 4L)), .Names = c("name", 
 "a", "b", "c", "d"), row.names = c("1", "2"), class = "data.frame")