Summarise and group_by not working with factor variables-CodePudding

I'm currently using the tidyverse package version 1.3.1, and when I run the following code:

data <- data.frame(gender = c(1,2,1,2,2,2,2,1,2,1), age = c(18,20,21,24,25,24,24,25,22,21))

data <- data%>%
  mutate(gender = factor(gender, levels = c("male", "female")))

data%>%
  group_by(gender)%>%
  summarise(mean = mean(age))

I get these results

   # A tibble: 1 × 2
  gender  mean
  <fct>  <dbl>
1 NA      22.4

CodePudding user response：

Yes, you should change the labels and not levels.

library(dplyr)

data%>%
  mutate(gender = factor(gender, labels = c("male", "female"))) %>%
  group_by(gender)%>%
  summarise(mean = mean(age))

#  gender  mean
#  <fct>  <dbl>
#1 male    21.2
#2 female  23.2

CodePudding user response：

We don't need to convert to factor for recoding. It can be directly done by using the 'gender' (numeric variable) as index for replacing the values

library(dplyr)
data %>%
    group_by(gender = c("male", "female")[gender]) %>%
    summarise(mean = mean(age, na.rm = TRUE))

-output

# A tibble: 2 × 2
  gender  mean
  <chr>  <dbl>
1 female  23.2
2 male    21.2

Or using fct_recode

library(forcats)
data %>%
   group_by(gender = fct_recode(as.character(gender), male = "1",
         female = "2")) %>% 
   summarise(mean = mean(age, na.rm = TRUE))
# A tibble: 2 × 2
  gender  mean
  <fct>  <dbl>
1 male    21.2
2 female  23.2