使用 tidyverse 中不同数据集的信息更新列

Question

我正在尝试让 R (

tidyverse

) 检查国家/地区（缩写位于两列中，两列都需要更新）包含国家/地区主列表的列，并将其替换为完整的国家/地区名称。我尝试过

ifelse

声明，但得到了奇怪的结果。数据集可以在这里找到。任何建议都会非常有帮助。

Answer 1

您可以通过国家代码

join

这两个数据框。

library(tidyverse)
library(readxl)

df <- read_xlsx('~Data SU23 Enroll R AY22-23 2023-08-23 2 Stack Overflow.xlsx')
countrydata <- read_xlsx('~TBL Country codes.xlsx')

glimpse(df)
#> Rows: 542
#> Columns: 3
#> $ ID          <chr> "F31769765", "E23531197", "Q07441087", "Y92280507", "F2688…
#> $ `MA Nation` <chr> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA…
#> $ `PR Nation` <chr> NA, "PK", "BG", NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, "M…
glimpse(countrydata)
#> Rows: 37
#> Columns: 2
#> $ CountryCode <chr> "BF", "BG", "BM", "BR", "CA", "CE", "CH", "GH", "GM", "HA"…
#> $ CountryName <chr> "BAHAMAS, THE", "BANGLADESH", "MYANMAR", "BRAZIL", "CANADA…

df %>%
  # Put all columns with country codes in a long layout
  pivot_longer(-ID) %>%
  filter(!is.na(value)) %>%
  # Join with the country code table
  left_join(countrydata,
            by = join_by(value == CountryCode)) %>%
  # Drop the country code column
  select(-value) %>%
  # Return to the two country columns layout
  pivot_wider(names_from = name,
              values_from = CountryName) %>%
  # Append the rows for IDs without country data
  bind_rows(df %>% filter(is.na(`PR Nation`) & is.na(`MA Nation`)))
#> # A tibble: 542 × 3
#>    ID        `PR Nation`    `MA Nation`
#>    <chr>     <chr>          <chr>      
#>  1 E23531197 PAKISTAN       <NA>       
#>  2 Q07441087 BANGLADESH     <NA>       
#>  3 U79148472 MEXICO         <NA>       
#>  4 Y43292349 PAKISTAN       <NA>       
#>  5 A40257720 INDIA          <NA>       
#>  6 Y64624318 CHINA          <NA>       
#>  7 B97628594 JAPAN          <NA>       
#>  8 T06694322 IRELAND        <NA>       
#>  9 J67643839 UNITED KINGDOM <NA>       
#> 10 B11219391 CHINA          <NA>       
#> # ℹ 532 more rows

^{创建于 2023-08-24，使用 reprex v2.0.2}