R - 计算基于另一个数据框的组合指标答案

【问题标题】：R - Calculating a combined metric for a dataframe based on anotherR - 计算基于另一个数据框的组合指标
【发布时间】：2018-05-01 14:44:30
【问题描述】：

我的数据框如下所示：

df = data.frame(Region=c(rep("NORDICS",1100),rep("DACH",900),rep("MED",1800),rep("CEE",15000),
                     rep("FRANCE",2000),rep("UK&I",2500)),
            Score=c(sample(seq(from = 1, to = 4, by = 1), size = 1100, replace = TRUE,prob = c(0.6,0.2,0.1,0.1)),
                 sample(seq(from = 1, to = 4, by = 1), size = 900, replace = TRUE,prob = c(0.3,0.3,0.2,0.2)),
                 sample(seq(from = 1, to = 4, by = 1), size = 1800, replace = TRUE,prob = c(0.8,0.1,0.05,0.05)),
                 sample(seq(from = 1, to = 4, by = 1), size = 15000, replace = TRUE,prob = c(0.2,0.2,0.2,0.4)),
                 sample(seq(from = 1, to = 4, by = 1), size = 2000, replace = TRUE,prob = c(0.9,0.05,0.03,0.02)),
                 sample(seq(from = 1, to = 4, by = 1), size = 2500, replace = TRUE,prob = c(0.9,0.05,0.03,0.02))))

数据框是按地区划分的各个分数的集合，其中每个观察结果都是一个问题的单个分数（Score 列）。

问题是从 1 到 4。

基于此数据框，我从Score 列按区域计算 KPI。 KPI 是1 或2 响应的总和除以给定区域的响应总数。

下面我的代码按地区计算KPI：

library(dplyr)

KPI_by_Region=df %>% group_by(Region) %>%
summarise(KPI = sum(Score %in% c(1,2))/n())

我的问题

仅使用包含按地区划分的 KPI 分数的 KPI_by_Region 数据框 -

我可以找出所有地区组合的 KPI 分数吗？在整个数据帧 (df) 上运行我的代码？

【问题讨论】：

如果你还在summary中维护一个n列，你应该能够重构你需要的东西？
我该怎么做？你能示范一下吗

标签： r dplyr sample summarization

【解决方案1】：

这是否给出了您正在寻找的结果？

KPI_by_Region <- df %>%
  group_by(Region) %>%
  summarise(KPI = sum(Score %in% c(1,2))/n(), Count = n())

allRegionsKPI <- sum(KPI_by_Region$KPI * KPI_by_Region$Count) / sum(KPI_by_Region$Count)

【讨论】：