adding percent calculation to dataframe-CodePudding

I would like to create a table with frequency counts and percentages. I have the frequency counts and calculated the percentages. When I try to add it to the dataframe, it's doesn't add the calculations to the dataframe. The code:

fpanel <- rename(count(surveydata, panel), Freq = n)
fpanel$proportion <- fpanel$Freq / dim(surveydata)[1]
fresponder <- rename(count(surveydata, panel, responder), Freq = n)
fresponder$percent <- c(fresponder[1,3] / fpanel[1,2], 
                        fresponder[2,3] / fpanel[1,2],
                        fresponder[3,3] / fpanel[2,2],
                        fresponder[4,3] / fpanel[2,2],
                        fresponder[5,3] / fpanel[3,2],
                        fresponder[6,3] / fpanel[3,2],
                        fresponder[7,3] / fpanel[4,2],
                        fresponder[8,3] / fpanel[4,2])

And the results:

The calculations are correct without adding to the dataframe but when I add it, I get "<dbl 1>". What kind of error is this and how can I fix it?

CodePudding user response：

fresponder (and fpanel) is a tibble, and unlike data.frames from which they inherit, fresponder[1,2] is still a tibble. As a hasty demonstration, as.data.frame(fresponder)[1,2] should return a numeric instead of a tibble.

class(mtcars)
# [1] "data.frame"
mtcars[1,2]
# [1] 6
tibble(mtcars)[1,2]
# # A tibble: 1 x 1
#     cyl
#   <dbl>
# 1     6

Untested, but try this:

fr <- as.data.frame(fresponder)
fp <- as.data.frame(fpanel)
fresponder$percent <- c(fr[1,3] / fp[1,2], 
                        fr[2,3] / fp[1,2],
                        fr[3,3] / fp[2,2],
                        fr[4,3] / fp[2,2],
                        fr[5,3] / fp[3,2],
                        fr[6,3] / fp[3,2],
                        fr[7,3] / fp[4,2],
                        fr[8,3] / fp[4,2])
# rm(fp,fr)

(But ultimately, calculating $percent here in this way seems very fragile and not resilient/robust.)