如何在计算 R 中的另一列时计算总数和百分比？

Question

所有， 提前致谢。我有这个学校数据集。每个类别（在类别列中）都有一定范围的学生人数（例如，从 30 到 60 名学生），所以我需要计算：

1. 属于每个类别（从类别 1 到类别 4）的教室总数，以及
2. 属于该类别的教室的百分比。 例如，有多少教室（NumOfClassrooms 列）属于 Category_4，这些教室占教室总数的百分比是多少？这是我的问题的一个说明性示例：

ID = 1:1050
District = rep(c("AR", "CO", "AL", "KS", "IN", "ME", "KY", "ME", "MN", "NJ"), times = c(80, 120, 100, 110, 120, 100, 100, 120, 100, 100))
schoolName = randomNames::randomNames(1050, ethnicity = 5 ,which.names = "last") 
Grade = rep(c("First", "Second", "Third", "Fourth"), times = c(400, 300, 200, 150))
NumOfClassrooms =  sample(1:6)
StudentNumber = sample(1:90, 5)

AverageNumOfStudents = StudentNumber/NumOfClassrooms

Category = ifelse(AverageNumOfStudents >  0 & AverageNumOfStudents < 10, "category_1", 
                  ifelse(AverageNumOfStudents >=10  & AverageNumOfStudents < 30, "category_2",
                   ifelse(AverageNumOfStudents >=30  & AverageNumOfStudents <= 60, "category_3",
                  ifelse(AverageNumOfStudents > 60 , "category_4", "NA"))))
                  
 dat = data.frame(ID, schoolName, Grade, NumOfClassrooms, StudentNumber, AverageNumOfStudents, Category)

最后，我需要使用以下代码将基于“地区”列的结果划分为单独的 excel 文件（一旦完成上述两个步骤，它应该可以正常工作）。

Final_Divide = Final_df %>%
  dplyr::group_by(District) %>%
  dplyr::ungroup()


list_data <- split(Final_Divide,
                   Final_Divide$District)
options(digits=3)
Map(openxlsx::write.xlsx, list_data, paste0(names(list_data), '.xlsx'))

非常感谢您。

Answer 1

在您的代码之前设置随机种子以实现可重复性：

set.seed(42)
# Your code creating dat
Table1 <- xtabs(NumOfClassrooms~Category, dat)
Table1
# Category
# category_1 category_2 category_4 
#       1925       1575        175 

Table2 <- prop.table(Table1)
round(Table2, 4)   # Proportions
# Category
# category_1 category_2 category_4 
#     0.5238     0.4286     0.0476 
round(Table2 * 100, 2)   # Percent
# Category
# category_1 category_2 category_4 
#      52.38      42.86       4.76 

如果我们在dat 中包含District：

dat <- data.frame(ID, District, schoolName, Grade, NumOfClassrooms, StudentNumber, AverageNumOfStudents, Category)
Table3 <- xtabs(NumOfClassrooms~District+Category, dat)
addmargins(Table3)
#         Category
# District category_1 category_2 category_4  Sum
#      AL         187        149         16  352
#      AR         143        121         14  278
#      CO         220        180         20  420
#      IN         220        180         20  420
#      KS         198        166         19  383
#      KY         187        148         17  352
#      ME         407        329         36  772
#      MN         176        153         17  346
#      NJ         187        149         16  352
#      Sum       1925       1575        175 3675

按地区划分的行百分比：

round(prop.table(Table3, 1) * 100, 2)
#         Category
# District category_1 category_2 category_4
#       AL      53.12      42.33       4.55
#       AR      51.44      43.53       5.04
#       CO      52.38      42.86       4.76
#       IN      52.38      42.86       4.76
#       KS      51.70      43.34       4.96
#       KY      53.12      42.05       4.83
#       ME      52.72      42.62       4.66
#       MN      50.87      44.22       4.91
#       NJ      53.12      42.33       4.55

Answer 2

这是使用tidyverse的可能解决方案

dat %>% 
  mutate("Total Classrooms" = n()) %>% 
  group_by(Category) %>% 
  mutate("Number of Classrooms in Category" = n(),
         "Category Percentage" = `Number of Classrooms in Category`/`Total Classrooms` * 100) 

这会给我们：

# Groups:   Category [3]
      ID District schoolName  Grade NumOfClassrooms StudentNumber AverageNumOfStude~ Category  `Total Classroom~ `Number of Classrooms in~ `Category Percent~
   <int> <chr>    <chr>       <chr>           <int>         <int>              <dbl> <chr>                 <int>                     <int>              <dbl>
 1     1 AR       Svyatetskiy First               5            87              17.4  category~              1050                       525               50  
 2     2 AR       Booco       First               1            79              79    category~              1050                       175               16.7
 3     3 AR       Jones       First               6            49               8.17 category~              1050                       350               33.3
 4     4 AR       Sapkin      First               3             5               1.67 category~              1050                       350               33.3
 5     5 AR       Fosse       First               2            35              17.5  category~              1050                       525               50  
 6     6 AR       Vanwagenen  First               4            87              21.8  category~              1050                       525               50  
 7     7 AR       Orth        First               5            79              17.4  category~              1050                       525               50  
 8     8 AR       Moline      First               1            49              79    category~              1050                       175               16.7
 9     9 AR       Bradford    First               6             5               8.17 category~              1050                       350               33.3
10    10 AR       Wollman     First               3            35               1.67 category~              1050                       350               33.3
# ... with 1,040 more rows

如果您需要仅包含类别/# 个教室/百分比数据的单独表格：

dat %>% 
  mutate("Total Classrooms" = n()) %>% 
  group_by(Category) %>% 
  mutate("Number of Classrooms in Category" = n(),
         "Category Percentage" = `Number of Classrooms in Category`/`Total Classrooms` * 100) %>% 
  select(Category, "Number of Classrooms in Category", "Category Percentage") %>% 
  unique()

这给了我们：

# A tibble: 3 x 3
# Groups:   Category [3]
  Category   `Number of Classrooms in Category` `Category Percentage`
  <chr>                                   <int>                 <dbl>
1 category_2                                525                  50  
2 category_4                                175                  16.7
3 category_1                                350                  33.3

请注意，在您的帖子中，这段代码有点多余：

Final_Divide = Final_df %>%
  dplyr::group_by(District) %>%
  dplyr::ungroup()

如果您先group，然后立即ungroup，您实际上只是在这样做：

Final_Divide <- Final_df

您还可以考虑添加 split(.$District) 以将您的数据转换为一个列表，所有这些都包含在一段代码中：

dat %>% 
  mutate("Total Classrooms" = n()) %>% 
  group_by(Category) %>% 
  mutate("Number of Classrooms in Category" = n(),
         "Category Percentage" = `Number of Classrooms in Category`/`Total Classrooms` * 100) %>% 
  split(.$District)