如何使用dplyr生成频率表 [英] How to use dplyr to generate a frequency table

问题描述

我喜欢创建一个在数据框中具有几列频率的表格。我正在复制下面我的数据框的一部分。

I like to create a table that has the frequency of several columns in my data frame. I am copying part of my data frame below.

该表应该具有颜色中的红色的频率（n和％），性别中具有F的频率。

The table is supposed to have frequency (both n and %) of "red" in Color and "F" in Gender.

我认为dplyr软件包可以做到这一点，但我无法理解。

I think that the dplyr package could do this but I cannot figure it out.

谢谢 -

    RespondentID    Color        Gender   
1     1503          Red           F      
2     1653          NA            M   
3     1982          Red           F   
4     4862          Red           NA   
15    4880          Blue          M  

推荐答案

library(dplyr)

df %>%
  count(Color, Gender) %>%
  mutate(prop = prop.table(n))

# Source: local data frame [4 x 4]
# Groups: Color [3]
# 
#    Color Gender     n      prop
#   (fctr) (fctr) (int)     (dbl)
# 1   Blue      M     1 1.0000000
# 2    Red      F     2 0.6666667
# 3    Red     NA     1 0.3333333
# 4     NA      M     1 1.0000000

每个注释更新 - 如果要单独查看每个变量，您将需要首先重新排列数据框。您可以使用 tidyr 完成此操作：

Updating per comment -- if you want to look at each variable separately, you will need to rearrange the dataframe first. You can accomplish this with tidyr:

library(tidyr)
library(dplyr)

gather(df, "var", "value", -RespondentID) %>%
  count(var, value) %>%
  mutate(prop = prop.table(n))

# Source: local data frame [6 x 4]
# Groups: var [2]
# 
#      var value     n  prop
#   (fctr) (chr) (int) (dbl)
# 1  Color  Blue     1   0.2
# 2  Color   Red     3   0.6
# 3  Color    NA     1   0.2
# 4 Gender     F     2   0.4
# 5 Gender     M     2   0.4
# 6 Gender    NA     1   0.2

这篇关于如何使用dplyr生成频率表的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！