从长转换为宽，在 R 中的两列上使用 pivot_wide()

Question

我想通过两列中的值将我的数据从长格式转换为宽格式。我如何使用 tidyverse 来做到这一点？

已更新dput

structure(list(Country = c("Algeria", "Benin", "Ghana", "Algeria", 
                       "Benin", "Ghana", "Algeria", "Benin", "Ghana"
), Indicator = c("Indicator 1", 
             "Indicator 1", 
             "Indicator 1", 
             "Indicator 2", 
             "Indicator 2", 
             "Indicator 2",
             "Indicator 3", 
             "Indicator 3", 
             "Indicator 3"
), Status = c("Actual", "Forecast", "Target", "Actual", "Forecast", 
          "Target", "Actual", "Forecast", "Target"), Value = c(34, 15, 5, 
                                                               28, 5, 2, 43, 5, 
                                                               1)), row.names 
= c(NA, -9L), class = c("tbl_df", "tbl", "data.frame"))


    Country Indicator   Status   Value
    <chr>   <chr>       <chr>    <dbl>
1 Algeria Indicator 1 Actual      34
2 Benin   Indicator 1 Forecast    15
3 Ghana   Indicator 1 Target       5
4 Algeria Indicator 2 Actual      28
5 Benin   Indicator 2 Forecast     5
6 Ghana   Indicator 2 Target       2
7 Algeria Indicator 3 Actual      43
8 Benin   Indicator 3 Forecast     5
9 Ghana   Indicator 3 Target       1

预期输出

Country Indicator1_Actual Indicator1_Forecast Indicator1_Target Indicator2_Actual

Algeria       34                    15                 5           28

等等

感谢任何提示！

foo <- data %>% pivot_wider(names_from = c("Indicator","Status"), values_from = "Value")

完美运行！

Answer 1

我认为您的 pivot_wider() 命令有误 data %>% pivot_wider(names_from = Indicator, values_from = c(Indicator, Status))

我打赌你不能对名称和值使用同一列。

试试这个代码

data %>% pivot_wider(names_from = c(Indicator, Status), values_from = Value))

说明：由于您希望 列名称 成为 指标 1_Actual，因此您需要指标和状态都在进行中进入你的 names_from

如果您提供示例数据和预期输出，将会很有帮助。但我在我的虚拟数据上对此进行了测试，它给出了预期的输出 -

数据：

# A tibble: 4 x 4
     a1    a2 a3       a4
  <int> <int> <chr> <dbl>
1     1     5 s        10
2     2     4 s        20
3     3     3 n        30
4     4     2 n        40

通话：a %>% pivot_wider(names_from = c(a2, a3), values_from = a4)

输出：

# A tibble: 4 x 5
     a1 `5_s` `4_s` `3_n` `2_n`
  <int> <dbl> <dbl> <dbl> <dbl>
1     1    10    NA    NA    NA
2     2    NA    20    NA    NA
3     3    NA    NA    30    NA
4     4    NA    NA    NA    40

如果要复现，这里有数据

structure(list(a1 = 1:4, a2 = 5:2, a3 = c("s", "s", "n", "n"), 
    a4 = c(10, 20, 30, 40)), row.names = c(NA, -4L), class = c("tbl_df", 
"tbl", "data.frame"))

编辑：对于尝试正确的 pivot_wider() 命令后编辑的问题 - 看起来您的数据实际上可能有重复项，在这种情况下，您看到的输出是有意义的 - 我建议您尝试使用 filter(Country == .., Indicator == .., Status == ..)

来确定您的数据是否真的有重复项

Answer 2

这可以通过在 pivot_wider().

的 names_from 参数中调用你的两列以更宽的旋转来实现

data %>% 
  pivot_wider(names_from = c("Indicator","Status"),
              values_from = "Value")

结果

  Country `Indicator 1_Ac… `Indicator 1_Fo… `Indicator 1_Ta… `Indicator 2_Ac… `Indicator 2_Fo…
  <chr>              <dbl>            <dbl>            <dbl>            <dbl>            <dbl>
1 Algeria               34               15                5               28                5

从长转换为宽，在 R 中的两列上使用 pivot_wide()

Converting from long to wide, using pivot_wide() on two columns in R

r

tidyr

reshape2