在函数中使用 dplyr slice_min 来获取与列最小值关联的行

Question

我正在尝试创建一个函数，该函数返回特定条件下死亡率最低的医院。该函数的输入是状态和条件。我还应该指出我是 R 新手。我非常喜欢 dplyr 包中的函数，但似乎在使用 slice_min 函数并将变量传递给它时遇到了一些问题。如果有人可以建议什么是错误的以及如何纠正它，将不胜感激。我将在下面展示我尝试过的两个似乎不起作用的选项。

函数调用示例为：

best_hospital("CA","HeartFailure")

功能是

best_hospital <- function(state, disease_type){
    library(dplyr)
    library(janitor)

    #Loads data
    data_path <- c('D:/Data Science/CourseraR/Assign3 Info/outcome-of-care-measures.csv')
    data_df <- read.csv(data_path, colClasses = "character")

    #Replaces capitals, spaces and special characters in names
    data_df <- data_df %>% 
            clean_names() %>%
            select(state,hospital_name,
               hospital_30_day_death_mortality_rates_from_heart_attack,
               hospital_30_day_death_mortality_rates_from_heart_failure,
               hospital_30_day_death_mortality_rates_from_pneumonia)

    colnames(data_df) <- c("State", "HospitalName","HeartAttack", "HeartFailure", "Pneumonia")
    
    data_df <- data_df %>%
            filter(State == {{state}} ) %>%
            select(State, HospitalName, {{disease_type}} ) 
    
    # This DOES NOT work   
    # data_df <- data_df %>% slice_min(order_by = {{disease_type}}, with_ties = TRUE)
    
    #This DOES NOT work
    data_df <- data_df %>% slice_min(order_by = HeartFailure, with_ties = TRUE)
    
    print(head(data_df))

}

我上面展示的两种方法都选择并返回数据。但不是与感兴趣列的最小值相关联的行（在本例中为心力衰竭）。欢迎任何帮助！谢谢

Answer 1

尝试以下方法：

library(dplyr)
library(janitor)

best_hospital <- function(data_df, state, disease_type){
 
  data_df <- data_df %>%
               filter(State == state) %>%
               select(State, HospitalName, disease_type) 
  
  data_df <- data_df %>% 
              slice_min(order_by = .data[[disease_type]], with_ties = TRUE)
  
  return(data_df)
}


#Loads data
data_path <- c('D:/Data Science/CourseraR/Assign3 Info/outcome-of-care-measures.csv')
data_df <- read.csv(data_path, colClasses = "character")

#Replaces capitals, spaces and special characters in names
data_df <- data_df %>% 
  clean_names() %>%
  select(state,hospital_name,
         hospital_30_day_death_mortality_rates_from_heart_attack,
         hospital_30_day_death_mortality_rates_from_heart_failure,
         hospital_30_day_death_mortality_rates_from_pneumonia)

colnames(data_df) <- c("State", "HospitalName","HeartAttack", "HeartFailure", "Pneumonia")

result <- best_hospital(data_df, "CA","HeartFailure")
result