R: Removing NA values from a data frame

Question

A bit of a newbie question: I have a data frame with 7,000 observations of 15 variables and 800+ NA values.

I have figured out how to identify the rows with 4 or more NA values: DF[rowSums(is.na(DF)) >= 4, ] , but I'd like to remove the records with 4 or more NA values from the DF . Can someone let me know where to put the na.rm = T if this is the best way?

Many thanks, I'm new to R and have looked and looked...

Answer 1

Your condition identifies the rows you want to omit. Negate it to ID the rows you want to keep, then store the result in the variable (or another one.):

DF <- DF[!(rowSums(is.na(DF)) >= 4), ]