question_type does not populate count, prop or rnp dataframes.

The question_type function correctly populates the raw dataframe classifying sentences according to the Interrogative, word but accompanying dataframes summarising the results are filled with NAs. This is the case for examples I have run and also for the example given in the reference manual.

e.g.
text <- c("Did you go to the shop?", "How do I swim?", "When is dinner ready?")
types <- question_type(text)