如何在R中将百分比分布输出为二维表？

Question

我正在使用RStudio，我有一个dataframe（df1）。 df1包含几列，但我对这3列感兴趣：compname，dept，losYRS。

我想按compname得到每个dept的百分比分布。 我的R代码如下：

library(lubridate)
library(tidyverse)

df2<-(subset(df1,compName %in% c("A")))

df3<-df2 %>%
     group_by(dept) %>%
     summarise(count = n() / nrow(.) * 100 )
df3

# A tibble: 11 x 2
   dept              count
   <chr>             <dbl>
 1 F&B (Kitchen)     18.6 
 2 F&B (Restaurant)  20.3 
 3 FINANCE            5.08
 4 FRONT OFFICE      10.2 
 5 HOUSEKEEPING      22.0 
 6 HR                 1.69
 7 LEISURE AND SPORT  3.39
 8 MAINTENANCE        8.47
 9 RESERVATION        1.69
10 SPA                5.08
11 STEWARDING         3.39

不是每次都为每个dept进行子设置，而是有一种方法来获取以下结果（有或没有Grand Total）：

我还希望将数字格式化为小数点后零位并附加％符号（如上所示）。

样本数据（使用dput）：

structure(list(compName2 = c("A", "A", "C", 
"B", "C", "A", "A", "B", "B", 
"A", "C", "C", "A","B", "B", "A", "C", "C", 
"A", "B"), dept = c("MAINTENANCE", "OPERATIONS", 
"F&B (Kitchen)", "F&B (Kitchen)", "HOUSEKEEPING", "F&B (Restaurant)", 
"RESERVATION", "F&B (Restaurant)", "HOUSEKEEPING", "MAINTENANCE", 
"FRONT OFFICE", "HOUSEKEEPING", "MAINTENANCE", "HOUSEKEEPING", 
"MAINTENANCE", "F&B (Restaurant)", "HOUSEKEEPING", "F&B (Restaurant)", 
"F&B (Restaurant)", "MAINTENANCE"), losYRS = c(31, 30, 29, 28, 
28, 28, 28, 27, 27, 27, 27, 27, 27, 26, 26, 26, 26, 26, 26, 25
)), .Names = c("compName", "dept", "losYRS"), row.names = c(NA, 
20L), class = "data.frame")

Answer 1

这是我为您解决的tidyverse问题：

library(tidyverse)

###Calculate percentages
df2 <- df %>% group_by(compName) %>% group_split(keep=T) %>% 
  map(group_by,dept) %>% map(function(x){summarize(x,perc = n()/nrow(x)*100,compName=x$compName[1])}) %>% bind_rows %>%
  spread(compName,perc)

###Added with base::split() for compatibility reasons:

df2 <- df %>% split(f=df$compName) %>%  map(group_by,dept) %>% map(function(x){summarize(x,perc = n()/nrow(x)*100,compName=x$compName[1])}) %>% bind_rows %>%
  spread(compName,perc)

##Create summary df for printing
print_df <- df2 %>% mutate_at(c("A","B","C"),round) %>% 
  bind_rows(df2 %>% replace(is.na(.),0) %>% summarize_at(c("A","B","C"),"sum") %>% mutate_all(round) %>%
              bind_cols(enframe("Grand Total",value="dept",name=NULL))) %>% mutate_at(c("A","B","C"),function(x){paste(as.character(x),"%")})

#print as tab-separated .txt-file
write.table(print_df,file="Test.txt",row.names=F)