修改R代码：将变量标签交叉表的X替换为对应变量名

原代码实现了跨dataframe的变量标签交叉对比，但仅用"X"标记存在性，无法体现标签相同但变量名不同的情况。以下是修改后的代码，将单元格中的"X"替换为实际变量名：

library(haven)
library(labelled)

# 测试数据准备：修改变量标签和变量名
var_label(GEM2001$weight) <- "Weight provided by data vendor"
var_label(GEM2002$weight) <- "Weight according to vendor"
names(GEM2001)[names(GEM2001) == "country"] <- "country_updtd"

df_list <- list(GEM2001 = GEM2001, GEM2002 = GEM2002)

# 生成标签-变量名映射表（每个dataframe单独处理）
df_label_maps <- lapply(df_list, function(df) {
  labels <- unlist(var_label(df))
  # 反转键值对：标签作为键，变量名作为值
  setNames(names(labels), labels)
})

# 收集所有唯一标签
all_labels <- unique(unlist(lapply(df_label_maps, names)))

# 构建最终交叉表
label_table <- data.frame(label = all_labels)
for (df_name in names(df_label_maps)) {
  # 匹配每个标签对应的变量名，无匹配则为空字符串
  label_table[[df_name]] <- sapply(all_labels, function(lab) {
    ifelse(lab %in% names(df_label_maps[[df_name]]), 
           df_label_maps[[df_name]][[lab]], 
           "")
  })
}

label_table

关键修改说明 #

• 原代码仅提取标签，修改后生成标签→变量名的映射表，保留了标签与变量名的对应关系
• 循环部分不再简单判断存在性，而是通过映射表匹配每个标签对应的变量名
• 若存在多变量对应同一标签的情况，可将代码修改为paste(df_label_maps[[df_name]][lab], collapse = ", ")，实现多个变量名的拼接显示

可复现测试数据 #

# 第一个数据集
GEM2001 <- structure(list(setid = structure(c(7700001, 7700002, 7700003, 
7700004, 7700005, 7700006), label = "Harmonization ID", format.spss = "F12.0", display_width = 14L), 
    setid_ne = structure(c(1000000007700001, 1000000007700002, 
    1000000007700003, 1000000007700004, 1000000007700005, 1000000007700006
    ), label = "Alternative ID variable to avoid duplicates across years", format.spss = "F15.0", display_width = 17L), 
    yrsurv = structure(c(2001, 2001, 2001, 2001, 2001, 2001), label = "Year survey was administered", format.spss = "F4.0"), 
    country_updtd = structure(c(7, 7, 7, 7, 7, 7), label = "Country", format.spss = "F4.0", display_width = 9L, labels = c(`United States` = 1, 
    Russia = 7, Egypt = 20, `South Africa` = 27, Greece = 30, 
    Netherlands = 31, Belgium = 32, France = 33, Spain = 34, 
    Hungary = 36, Italy = 39, Romania = 40, Switzerland = 41, 
    Austria = 43, `United Kingdom` = 44, Denmark = 45, Sweden = 46, 
    Norway = 47, Poland = 48, Germany = 49, Peru = 51, Mexico = 52, 
    Argentina = 54, Brazil = 55, Chile = 56, Colombia = 57, Malaysia = 60, 
    Australia = 61, Indonesia = 62, Philippines = 63, `New Zealand` = 64, 
    Singapore = 65, Thailand = 66, Japan = 81, Korea = 82, Vietnam = 84, 
    China = 86, Turkey = 90, India = 91, Pakistan = 92, Iran = 98, 
    Canada = 101, Morocco = 212, Algeria = 213, Tunisia = 216, 
    Libya = 218, Ghana = 233, Nigeria = 234, Angola = 244, Barbados = 246, 
    Ethiopia = 251, Uganda = 256, Zambia = 260, Namibia = 264, 
    Malawi = 265, Botswana = 267, Portugal = 351, Luxembourg = 352, 
    Ireland = 353, Iceland = 354, Finland = 358, Lithuania = 370, 
    Latvia = 371, Estonia = 372, Serbia = 381, Montenegro = 382, 
    Croatia = 385, Slovenia = 386, `Bosnia and Herzegovina` = 387, 
    Macedonia = 389, `Czech Republic` = 420, Slovakia = 421, 
    Guatemala = 502, `El Salvador` = 503, `Costa Rica` = 506, 
    Panama = 507, Venezuela = 582, Bolivia = 591, Ecuador = 593, 
    Suriname = 597, Uruguay = 598, `* 'Azores'` = 620, Tonga = 676, 
    Vanuatu = 678, Kazakstan = 701, `Shenzhen*` = 755, `Puerto Rico` = 787, 
    `Dominican Republic` = 809, `Hong Kong` = 852, `Trinidad & Tobago` = 868, 
    Jamaica = 876, Bangladesh = 880, Taiwan = 886, Lebanon = 961, 
    Jordan = 962, Syria = 963, `Saudi Arabia` = 966, Yemen = 967, 
    `West Bank & Gaza Strip` = 970, `United Arab Emirates` = 971, 
    Israel = 972), class = c("haven_labelled", "vctrs_vctr", 
    "double")), weight = structure(c(0.947503767410607, 0.919003654090076, 
    0.924603676356567, 1.01710404415125, 0.716602849315504, 0.83510332049034
    ), label = "Weight provided by data vendor", format.spss = "F8.6", display_width = 10L)), row.names = c(NA, 
-6L), class = c("tbl_df", "tbl", "data.frame"), label = "JN725 - IAE - GEM 2009")

# 第二个数据集
GEM2002 <- structure(list(setid = structure(c(1121800009, 1121800025, 1121800031, 
1121800035, 1121800036, 1121800039), label = "Harmonization ID", format.spss = "F12.0", display_width = 14L), 
    setid_ne = structure(c(2000001121800009, 2000001121800025, 
    2000001121800031, 2000001121800035, 2000001121800036, 2000001121800039
    ), label = "Alternative ID variable to avoid duplicates across years", format.spss = "F15.0", display_width = 17L), 
    yrsurv = structure(c(2002, 2002, 2002, 2002, 2002, 2002), label = "Year survey was administered", format.spss = "F4.0"), 
    country = structure(c(1, 1, 1, 1, 1, 1), label = "Country", format.spss = "F4.0", display_width = 9L, labels = c(`United States` = 1, 
    Russia = 7, Egypt = 20, `South Africa` = 27, Greece = 30, 
    Netherlands = 31, Belgium = 32, France = 33, Spain = 34, 
    Hungary = 36, Italy = 39, Romania = 40, Switzerland = 41, 
    Austria = 43, `United Kingdom` = 44, Denmark = 45, Sweden = 46, 
    Norway = 47, Poland = 48, Germany = 49, Peru = 51, Mexico = 52, 
    Argentina = 54, Brazil = 55, Chile = 56, Colombia = 57, Malaysia = 60, 
    Australia = 61, Indonesia = 62, Philippines = 63, `New Zealand` = 64, 
    Singapore = 65, Thailand = 66, Japan = 81, Korea = 82, Vietnam = 84, 
    China = 86, Turkey = 90, India = 91, Pakistan = 92, Iran = 98, 
    Canada = 101, Morocco = 212, Algeria = 213, Tunisia = 216, 
    Libya = 218, Ghana = 233, Nigeria = 234, Angola = 244, Barbados = 246, 
    Ethiopia = 251, Uganda = 256, Zambia = 260, Namibia = 264, 
    Malawi = 265, Botswana = 267, Portugal = 351, Luxembourg = 352, 
    Ireland = 353, Iceland = 354, Finland = 358, Lithuania = 370, 
    Latvia = 371, Estonia = 372, Serbia = 381, Montenegro = 382, 
    Croatia = 385, Slovenia = 386, `Bosnia and Herzegovina` = 387, 
    Macedonia = 389, `Czech Republic` = 420, Slovakia = 421, 
    Guatemala = 502, `El Salvador` = 503, `Costa Rica` = 506, 
    Panama = 507, Venezuela = 582, Bolivia = 591, Ecuador = 593, 
    Suriname = 597, Uruguay = 598, `* 'Azores'` = 620, Tonga = 676, 
    Vanuatu = 678, Kazakstan = 701, `Shenzhen*` = 755, `Puerto Rico` = 787, 
    `Dominican Republic` = 809, `Hong Kong` = 852, `Trinidad & Tobago` = 868, 
    Jamaica = 876, Bangladesh = 880, Taiwan = 886, Lebanon = 961, 
    Jordan = 962, Syria = 963, `Saudi Arabia` = 966, Yemen = 967, 
    `West Bank & Gaza Strip` = 970, `United Arab Emirates` = 971, 
    Israel = 972), class = c("haven_labelled", "vctrs_vctr", 
    "double")), weight = structure(c(0.666652666946661, 1.35532689346212, 
    0.886868262634747, 0.247242055158897, 1.7567198656027, 0.595583088338233
    ), label = "Weight according to vendor", format.spss = "F8.6", display_width = 10L)), row.names = c(NA, 
-6L), class = c("tbl_df", "tbl", "data.frame"), label = "JN725 - IAE - GEM 2009")

内容的提问来源于stack exchange，提问作者flxflks