如何在R中创建DAC列：多条件判断及特殊列名处理

问题 #

需处理包含多种弱势社区（DAC）认定来源的公开数据集，创建名为DAC的新列：只要任意一个认定列符合条件，该列就显示"Yes"，否则显示"No"。相关认定列及取值规则如下：

• Disadvantaged_Community_OEHHA：值为全大写"YES"或"NO"
• Disadvantaged_Community_SB535：值为首字母大写"Yes"或"No"
• DAC(CARB)_LIC(CARB)：值为"Low Income and Disadvantaged"或"NA"
• Disadvantaged_Community_CE53：值为全大写"YES"或"NO"
• Disadvantaged_Community_CE54：值为全大写"YES"或"NO"

同时疑问：带括号的列名DAC(CARB)_LIC(CARB)是否会有问题？已查到用反引号`可解决，但尝试的代码未成功，最后尝试的代码如下：

clean_transport_2022 <-transform(clean_transport_2022, DAC = ifelse(Disadvantaged_Community_OEHHA == "YES" | Disadvantaged_Community_SB535 == "Yes", `DAC(CARB)_LIC(CARB)` == "Low Income and Disadvantaged" | Disadvantaged_Community_CE53 = "YES" | Disadvantaged_Community_CE54 = "YES", "Yes", "No"))

数据集预览：

> head(clean_transport_2022)
# A tibble: 6 × 22
  Recipient_Contractor    Project_Title Project_Amount Project_Type
  <chr>                   <chr>                  <dbl> <chr>       
1 ABAG                    Bay Area EV …         14533. Electric Ve…
2 CA EV Alliance          Bay Area Cha…         12474. Electric Ve…
3 RTC Fuels, LLC dba Pea… Pearson Fuel…         71053  E85 Fueling…
4 Blink Acquisition       Nissan Elect…          6849. Electric Ve…
5 Redwood Coast Energy A… North Coast …         70000  Electric Ve…
6 ABAG                    Bay Area EV …          7266. Electric Ve…
# ℹ 18 more variables: Fuel_Type <chr>, Supply_Chain_Phase <chr>,
#   RDD_D_Phase <chr>, Status <chr>, Project_City <chr>,
#   Project_Zip <chr>, County <chr>, Air_District <chr>,
#   Electric_Service_Area <chr>,
#   Disadvantaged_Community_OEHHA <chr>,
#   Disadvantaged_Community_SB535 <chr>, Low_Income_SB535 <chr>,
#   Project_Location <chr>, LowIncome_SB1000 <chr>, …

解决方法 #

原代码错误点 #

1. ifelse参数顺序错误：正确格式为ifelse(条件, 条件为真时的值, 条件为假时的值)，你将多个OR条件拆分为参数，导致逻辑混乱。
2. 条件判断误用赋值符号=，应使用等于符号==（如Disadvantaged_Community_CE53 = "YES"为错误写法）。
3. 多OR条件未合理包裹，逻辑优先级不明确。

正确实现代码 #

方案1：Base R 写法

clean_transport_2022 <- transform(clean_transport_2022, 
                                  DAC = ifelse(
                                    Disadvantaged_Community_OEHHA == "YES" |
                                      Disadvantaged_Community_SB535 == "Yes" |
                                      `DAC(CARB)_LIC(CARB)` == "Low Income and Disadvantaged" |
                                      Disadvantaged_Community_CE53 == "YES" |
                                      Disadvantaged_Community_CE54 == "YES",
                                    "Yes", 
                                    "No"
                                  ))

方案2：dplyr 写法（逻辑更清晰）

若习惯使用tidyverse工具，推荐用mutate实现：

library(dplyr)

clean_transport_2022 <- clean_transport_2022 %>%
  mutate(DAC = case_when(
    Disadvantaged_Community_OEHHA == "YES" |
      Disadvantaged_Community_SB535 == "Yes" |
      `DAC(CARB)_LIC(CARB)` == "Low Income and Disadvantaged" |
      Disadvantaged_Community_CE53 == "YES" |
      Disadvantaged_Community_CE54 == "YES" ~ "Yes",
    TRUE ~ "No"
  ))

带括号列名的处理 #

你的判断正确：包含特殊字符（括号、空格等）的列名，在R中需用反引号`包裹才能正常调用，上述代码已正确处理DAC(CARB)_LIC(CARB)列。

另外，若DAC(CARB)_LIC(CARB)列的"NA"是缺失值（而非字符串"NA"），可将对应条件改为!is.na(DAC(CARB)_LIC(CARB))，需根据实际数据情况调整。

内容的提问来源于stack exchange，提问作者Julia Caron