0

I have a dataframe that includes some "no" responses - this is analogous to "I don't know", and was used to allow participants to skip this question if they didn't know the answer.

I'm trying to filter out all of these "no" values across the entire df (816 columns). These are legitimate responses, so I don't just want to convert them to NA, but I'd like to filter them out for one particular analysis.

I've tried the following, but this also filters out partial matches, when "no" is found within another word (e.g. "nose" becomes "se", "cannon" becomes "cann"). How can I filter out "no" responses only when the entire value is a match, and skip any partial matches within other values?

naming_data %>%
   filter(!grepl('no', names))

Any help appreciated! Jamie

adajam
  • 193
  • 1
  • 10
  • 2
    See here if it helps, [How to use grep()/gsub() to find exact match](https://stackoverflow.com/questions/26813667/how-to-use-grep-gsub-to-find-exact-match). If that doesn't help, give a reproducible example. – deepseefan Oct 16 '19 at 09:21
  • Thanks for the guidance! This worked for me: `naming_data %>% filter(!grepl('^no$', names))`, using ^ to assign the start of the string, and $ to assign the end. – adajam Oct 16 '19 at 09:27

0 Answers0