個案研究導論

Tidyverse 的類別資料

Emily Robinson

Data Scientist

長條圖標題為「壓力艙裡，別人就是地獄」，附註為「在 874 名航空旅客受訪者中，認為某行為非常或有些失禮的比例」。最大一條是「明知小孩失控仍帶上機」82%，最小是「為了家人要求換位」17%。共 9 條。此調查來自 FiveThirtyEight，於 SurveyMonkey 進行。

原始資料集

# A tibble: 1,040 x 27
  RespondentID travel_amount   do_recline   
         <dbl> <chr>           <chr>        
1  3436139758. Once a year or… NA           
2  3434278696. Once a year or… About half t…
3  3434275578. Once a year or… Usually      
4  3434268208. Once a year or… Always       
# ... with 24 more variables: height <chr>,
#   children_sub_18 <chr>,
#   middle_arm_rest_three <chr>,
#   middle_arm_rest_two <chr>,
#   window_shade_control <chr>,
#   rude_move_seats <chr>, rude_talk <chr>,
#   times_get_up <chr>,
#   recliner_obligation <chr>,
#   rude_recline <chr>,
#   eliminate_recline <chr>,
#   rude_switch_seats_friend <chr>,

工具複習

wide_data

# A tibble: 2 x 3
  favorite_fruit favorite_vegetable disliked_dessert
  <chr>          <chr>              <chr>           
1 apple          carrot             cookie          
2 orange         cauliflower        cake

wide_data %>%
   mutate(across(where(is.character), as.factor))

# A tibble: 2 x 3
  favorite_fruit favorite_vegetable disliked_dessert
  <fct>          <fct>              <fct>           
1 apple          carrot             cookie          
2 orange         cauliflower        cake

tidyr pivot_longer()

wide_data %>%
   pivot_longer(everything(), names_to = "column", values_to = "value")

# A tibble: 6 x 2
  column             value      
  <chr>              <chr>      
1 favorite_fruit     apple      
2 favorite_fruit     orange     
3 favorite_vegetable carrot     
4 favorite_vegetable cauliflower
5 disliked_dessert   cookie     
6 disliked_dessert   cake

Select 輔助函式

wide_data %>%
   select(contains("favorite"))

# A tibble: 2 x 2
  favorite_fruit favorite_vegetable
  <chr>          <chr>             
1 apple          carrot            
2 orange         cauliflower

一起來練習吧！

Tidyverse 的類別資料