In R , a question about change variable type to 'factor'-CodePudding

I use the code below to change the variable 'period' from 'character' to 'factor' (and factor levels sequence to match the order contained in 'period' text).

Below is the current code. Is there an available function for it ?

library(tidyverse)

raw_data <- 
  data.frame(
    period=c('Q2','Q1','8','7','40','41'),
    amount=c(1:6)
  ) 


arranged_data <- raw_data %>% arrange(match(parse_number(period),c(1:41))) %>% 
  mutate(period=fct_inorder(period))

CodePudding user response：

lvls <- unique(raw_data$period)
raw_data$period <- factor(raw_data$period, levels = lvls[order(parse_number(lvls))])

Resulting in:

levels(raw_data$period )
[1] "Q1" "Q2" "7"  "8"  "40" "41"

CodePudding user response：

> raw_data %>% mutate(period = factor(period)) %>% as_tibble()
# A tibble: 6 × 2
  period amount
  <fct>   <int>
1 Q2          1
2 Q1          2
3 8           3
4 7           4
5 40          5
6 41          6

As you can see, you just have to call the factor() function on a vector to convert it into a factor.

CodePudding user response：

I believe this should work! The issue is that fct_inorder doesn't change column type, so you should turn the column into a factor first:

library(tidyverse)

raw_data <- 
   data.frame(
      period=c('Q2','Q1','8','7','40','41'),
      amount=c(1:6)) 
    
arranged_data <- raw_data %>% arrange(match(parse_number(period), c(1:41))) %>%
   mutate(period = as.factor(period)) %>%
   mutate(period=fct_inorder(period))