Warning: Mutate_impl(.data, dots): Unequal factor levels: coercing to character

Question

The Titanic Dataset can be downloaded from kaggle: kaggle.com/c/titanic/data. Please use the train.csv or install the package 'titanic' and use the dataset titanic_train.

This works

library(dplyr)
library(stringr)

titanic <- titanic %>%
    mutate(Cabin_Letter = ifelse(!is.na(Cabin), str_extract(Cabin, "[A-Z]+"), 'Unknown'))

This does not work entirely

titanic <- titanic %>%
    mutate(Cabin_Letter = factor(ifelse(!is.na(Cabin), str_extract(Cabin, "[A-Z]+"), 'Unknown')))

Warning:

Warning messages: 1: In mutate_impl(.data, dots) : Unequal factor levels: coercing to character 2: In mutate_impl(.data, dots) : binding character and factor vector, coercing into character vector 3: In mutate_impl(.data, dots) : binding character and factor vector, coercing into character vector 4: In mutate_impl(.data, dots) : binding character and factor vector, coercing into character vector 5: In mutate_impl(.data, dots) : binding character and factor vector, coercing into character vector 6: In mutate_impl(.data, dots) : binding character and factor vector, coercing into character vector 7: In mutate_impl(.data, dots) : binding character and factor vector, coercing into character vector

How could I resolve this issue? I don't want to take the extra line:

titanic$Cabin_letter <- factor(titanic$Cabin_letter)

Answer 1

This issue can happen if the data is grouped (grouped_df) using the group_by() function. I just ran into it. My solution was to ungroup() the data frame and then convert to factor using as.factor().