Avoid using assign() in R

Question

In many answers here in SO, posters advise to avoid using assign() to create new variables inside a loop, like this code reproduced here from this question:

myDf <- mtcars
splitVar <- factor(myDf$gear)
levelsVar <- levels(splitVar)
splitDataFrame <- split(myDf, splitVar)
for (i in 1:length(levelsVar)) {
  assign(paste0("newDataFrameGear", levelsVar[i]), data.frame(splitDataFrame[i]))
}
ls(pattern = "^newData")

This post explains why it is considered bad practice, but what other options are available to prevent this?

Did you [Google](http://stackoverflow.com/questions/17559390/why-is-using-assign-bad)? — David Arenburg, Jan 24 '17 at 00:09
The duplicate might only cover the "bad" part. Could it be that the workaround portion of his question still be left unanswered? — Tim Biegeleisen, Jan 24 '17 at 00:14
I had 2 questions. The question marked as a duplicate does not show good examples of how to prevent this. — Paulo MiraMor, Jan 24 '17 at 00:15
For "how to prevent this", see also [How to make a list of data frames in R?](http://stackoverflow.com/a/24376207/903061). — Gregor Thomas, Jan 24 '17 at 01:00

score 1 · Answer 1 · answered Jan 24 '17 at 00:05

1

Why not just keep using the original list, but assign the names you want:

names(splitDataFrame) <- paste0("newDataFrameGear",
                                as.character(levels(splitVar)))

answered Jan 24 '17 at 00:05

Tim Biegeleisen

502,043
27
286
360

Thanks for your answer, but I was meaning a more general solution, not only for this case in specific. – Paulo MiraMor Jan 24 '17 at 11:49

Avoid using assign() in R

1 Answers1