Ignore NAs when counting frequency of values

user_18269823 0
Viewed 64

I am new to R and am trying to count how often a certain value occurs in a variable ('Q1'), and store that number in a new variable ('Freqnt'). I want to ignore NAs.

I have managed to get something to work:

library(dplyr)
test3 <- test3 %>% group_by(Q1) %>% mutate(Freqnt=n())

#     Q1      Freqnt
#1    1       6
#2    1       6
#3    1       6
#4    1       6
#5    1       6
#6    1       6
#7   NA       2
#8   NA       2

As you can see this works but this also counts NAs, which I do not want.

The following does not count the NAs, but it completely removes the rows with NAs, which I don't want either:

test3 <- test3 %>% filter(!is.na(Q1)) %>% group_by(Q1) %>% mutate(Freqnt=n())

#     Q1      Freqnt
#1    1       6
#2    1       6
#3    1       6
#4    1       6
#5    1       6
#6    1       6

How can I, instead of dropping rows with NAs, enter something like 'NA' or leave the cell empty for the Freqnt variable, if Q1 cannot be counted?

Thanks!

dplyr na count r
1 Answers

We could do

library(dplyr)
test3 %>%
  mutate(Freqnt = Q1) %>%
  group_by(Q1) %>% 
  mutate(Freqnt= if(all(is.na(Q1))) NA_real_ else sum(!is.na(Freqnt))) %>%
  ungroup
user_37322710
Related