Group by and Count character values R

Question

I need to count how many orders is per date and how many "yes" and "no" for each date. How can I from the data frame like this:

  order <- c("order_1",  "order_2","order_3","order_4","order_5")
  check <- c("yes",  "no","no","yes","yes")
  df1 <-  data.frame(order,  date,check )
  df1


    order   date      check
1 order_1 2020-01-01   yes
2 order_2 2020-01-01    no
3 order_3 2020-01-01    no
4 order_4 2020-01-02   yes
5 order_5 2020-01-02   yes

Get table like this:

    date      number_orders  yes  no
1 2020-01-01       3          1    2
2 2020-01-02       2          0    0

´´´

Answer 1

Start by adding your yes and no columns to the dataframe. Let yes be true whenever check == 'yes' and false otherwise, and let no be true whenever check == 'no' and false otherwise.

df1 <- df1 %>% mutate(yes = (check == 'yes'), no = (check == 'no'))

Now group by date and use summarize() to count up the number of orders. You can take advantage of the fact that TRUE evaluates to 1 and FALSE evaluates to 0 to sum over the yes and no columns.

df1 <- df1 %>% 
  group_by(date) %>% 
  summarise(number_orders = n(),
         yes = sum(yes),
         no = sum(no))

The full code can be written as one line with dplyr's pipe operation.

df1 <- df1 %>% mutate(yes = (check == 'yes'), no = (check == 'no')) %>%
  group_by(date) %>% 
  summarise(number_orders = n(),
         yes = sum(yes),
         no = sum(no))