如何使用 dplyr 将跨因子级别的分组计数保存到新变量中？

Question

我尝试将各种因素水平的分组计数保存到一个新变量中：

假设我的数据如下所示：

a <- c(1,1,1,1,1,2,2,2,2,2,3,3,3,3,3,4,4,4,4,4)
b <- c("acc", "rej", "con", "acc", "rej", "con", "acc", "rej", "con", "acc", "rej", "con", "acc", "rej", "con","acc", "rej", "con", "acc", "rej")

df <- data.frame(a,b) 

生成的数据框应如下所示：

a <- c(1,2,3,4)
number_acc <- c(2,2,1,2)
number_rej <- c(2,1,2,2)
number_con <- c(1,2,2,1)

我尝试通过以下方式解决问题：

df2 <- df %>%
  group_by(a) %>% 
  mutate(number_acc = count(b == 'acc'), 
         number_rej = count(b == 'rej'),
         number_con = count(b == 'con'))

但是，我收到一条错误消息，指出方法“count”不能应用于“logical”类的对象。

感谢您的帮助！

Answer 1

使用 janitor 包中的 tabyl 函数：

您的数据：

a <- c(1,1,1,1,1,2,2,2,2,2,3,3,3,3,3,4,4,4,4,4)
b <- c("acc", "rej", "con", "acc", "rej", "con", "acc", "rej", "con", "acc", "rej", "con", "acc", "rej", "con","acc", "rej", "con", "acc", "rej")

df <- data.frame(a,b)

按计数汇总分组：

library(janitor)
data_summary <- tabyl(df, a, b)
data_summary

# a acc con rej
# 1   2   1   2
# 2   2   2   1
# 3   1   2   2
# 4   2   1   2

Answer 2

要使现有代码正常工作，我们需要 summarise 而不是 mutate，以及 sum 而不是 count em>：

df %>%
  group_by(a) %>% 
  summarise(number_acc = sum(b == 'acc'), 
            number_rej = sum(b == 'rej'),
            number_con = sum(b == 'con'))

# # A tibble: 4 x 4
#       a number_acc number_rej number_con
#   <dbl>      <int>      <int>      <int>
# 1     1          2          2          1
# 2     2          2          1          2
# 3     3          1          2          2
# 4     4          2          2          1

但是有更好的方法可以做到这一点，例如查看答案：

• Frequency count of two column in R
• dplyr : how to get two-way tables with marginal proportions?
• Count multiple columns and group by in R

Answer 3

这是另一种方法： 我们可以在count 之后使用pivot_wider 和names_glue：

library(tidyr)
library(dplyr)

df %>% 
  count(a,b) %>% 
  pivot_wider(
    names_from = b,
    values_from = n,
    names_glue = "{b}_{'number'}"
  )

      a acc_number con_number rej_number
  <dbl>      <int>      <int>      <int>
1     1          2          1          2
2     2          2          2          1
3     3          1          2          2
4     4          2          1          2