基于包含 tidyverse 类别的其他两列创建新变量

Question

我正在尝试操作数据并希望基于两个具有分类级别的列创建新变量。我试图通过 tidyverse 以 3 种方式实现这一点（需要坚持使用这个库）。

这是我尝试过的。

data_indices <- df_subs %>%
   mutate(ifelse(sex == "Female" & patient.class == "Not_Admitted", "female_not_admitted",
          ifelse(sex == "Female" & patient.class == "ORDINARY ADMISSION", "female_admitted",
          ifelse(sex == "Male" & patient.class == "Not_Admitted", "male_not_admitted",
          ifelse(sex == "Male" & patient.class == "ORDINARY ADMISSION",  "male_admitted")))))

然而，我得到了一个错误：

Error in mutate_impl(.data, dots) : 
Evaluation error: argument "no" is missing, with no default.

其次，我也尝试过 tidyverse 中的 unite 功能，但没有成功。

最后，第三，我在同一个库中搜索并找到了组索引。这就是我所做的，但惨遭失败。 :)

data_indices <- ds_sum_age_sex %>%
  bind_cols(Group_indic = group_indices("patient.class", "sex"))

出现错误：

 Error in UseMethod("group_indices_") : 
 no applicable method for 'group_indices_' applied to an object of class 
 "character"

有人可以帮忙吗？

Answer 1

有点难以理解你需要这些函数来做什么。但这是一个猜测:)

df_subs <- data.frame(sex = c("Male", 
                              "Female", 
                              "Female"),
                      patient.class = c("Not_Admitted", 
                                        "ORDINARY ADMISSION", 
                                        "ORDINARY ADMISSION"))

# Question 1
df_subs %>% 
  mutate(sex_patient_class = case_when(
    sex == "Female" & patient.class == "Not_Admitted"          ~ "female_not_admitted",
    sex == "Female" & patient.class == "ORDINARY ADMISSION"    ~ "female_admitted",
    sex == "Male"   & patient.class == "Not_Admitted"          ~ "male_not_admitted",
    sex == "Male"   & patient.class == "ORDINARY ADMISSION"    ~ "male_admitted"
  ))

# Question 2
df_subs %>% 
  unite(sex_patient_class, sex, patient.class, sep = "_")

# Question 3
df_subs %>% 
  mutate(group_indices = group_indices(., patient.class, sex))