创建一个变量以计算列子集的每行中唯一值的数量

Posted 2023-02-14

【中文标题】创建一个变量以计算列子集的每行中唯一值的数量

【英文标题】：Create a variable to count the number of unique values in each row for a subset of columns

【发布时间】：2022-01-18 22:52:44

【问题描述】：

我想创建一个变量来计算列子集（即基线、wave1、wave2、wave3）每行中唯一值的数量。到目前为止，我有以下内容。我已经包含了一个带有变量“示例”的示例数据集，以显示我所追求的。我还包含了变量“change”，它显示了使用下面的代码创建的变量。

# Create example data
data <- structure(list(age = c("18", "19", NA, "40", "21", "33", "32", 
"34", "43", "22"), baseline = c("1", "1", NA, "4", "1", "3", 
"2", "4", "3", "2"), wave1 = c("1", "1", "2", "4", "4", "3", 
"2", "4", "3", "2"), wave2 = c("1", "1", "4", "4", NA, "3", 
"2", "4", "3", "2"), wave3 = c("1", "2", NA, "4", "4", "3", 
"2", "4", "3", "4"), example = c("1", "2", "2", "1", "2", "1", 
"1", "1", "1", "2"), change = c(6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 
6L, 6L)), row.names = c(NA, -10L), groups = structure(list(.rows = structure(list(
    1L, 2L, 3L, 4L, 5L, 6L, 7L, 8L, 9L, 10L), ptype = integer(0), class = c("vctrs_list_of", 
"vctrs_vctr", "list"))), row.names = c(NA, -10L), class = c("tbl_df", 
"tbl", "data.frame")), class = c("rowwise_df", "tbl_df", "tbl", 
"data.frame"))

library(dplyr)
# Create a var for change at any point (ignoring NAs)
data <- data %>% 
  rowwise() %>% #perform operation by row
  mutate(change = length(unique(na.omit(baseline,wave1,wave2,wave3))))

【问题讨论】：

我不认为有一个更有效的过程。 @akrun 刚刚建议使用n_distinct 函数来替换您的length(unique(.))，以及使用c_across，但是虽然它们增加了可读性（并且是dplyr-canonical），但我不知道你会发现很多更好。

data[,"change"] <- apply(data[,2:5],1,function(x) length(na.omit(unique(x))))

【参考方案1】：

我们可以使用n_distinct，我们可以使用na.rm 参数来删除NA 元素（尽管在OP 的数据中，它是"NA"）

library(dplyr)
data %>%
   type.convert(as.is = TRUE) %>%
   rowwise %>% 
   mutate(change = n_distinct(c_across(baseline:wave3), na.rm = TRUE)) %>%
   ungroup

-输出

# A tibble: 10 × 7
     age baseline wave1 wave2 wave3 example change
   <int>    <int> <int> <int> <int>   <int>  <int>
 1    18        1     1     1     1       1      1
 2    19        1     1     1     2       2      2
 3    NA       NA     2     4    NA       2      2
 4    40        4     4     4     4       1      1
 5    21        1     4    NA     4       2      2
 6    33        3     3     3     3       1      1
 7    32        2     2     2     2       1      1
 8    34        4     4     4     4       1      1
 9    43        3     3     3     3       1      1
10    22        2     2     2     4       2      2

---

或者dapply from collapse 的更快选项

library(collapse)
data$change <- dapply(slt(ungroup(data), baseline:wave3), 
      MARGIN = 1, FUN = fndistinct)