Invalid factor level with rbind to data frame

When you do

c("one","lala",1)

this creates a vector of strings. The 1 is converted to character type, so that all elements in the vector are the same type.

Then rbind(a,b) will try to combine a which is a data frame and b which is a character vector and this is not what you want.

The way to do this is using rbind with data frame objects.

a <- NULL
b <- data.frame(A="one", B="lala", C=1)
d <- data.frame(A="two", B="lele", C=2)

a <- rbind(a, b)
a <- rbind(a, d)

Now we can see that the columns in data frame a are the proper type.

> lapply(a, class)
$A
[1] "factor"

$B
[1] "factor"

$C
[1] "numeric"

>

Notice that you must name the columns when you create the different data frame, otherwise rbind will fail. If you do

b <- data.frame("one", "lala", 1)
d <- data.frame("two", "lele", 2)

then

> rbind(b, d)
Error in match.names(clabs, names(xi)) : 
  names do not match previous names

Related