使用ggplot绘制柱状图时误差线位置错误的原因排查

问题原因 #

你的数据框是分组的9行数据，每个variable对应3行完全重复的统计值（gfp_mean、ymax、ymin完全一致），导致geom_errorbar在同一个x位置重复绘制3次误差线，视觉上出现重叠偏移；同时分组数据框的默认分组逻辑也会干扰绘图定位。

解决方法 #

方法1：用去重后的统计数据绘图 #

直接保留每个variable的唯一统计行，避免重复绘制：

# 转换为普通数据框并去重
gfp_long_uniq <- gfp_long %>% 
  ungroup() %>% 
  distinct(variable, .keep_all = TRUE)

# 绘图
ggplot(gfp_long_uniq) +
  geom_bar(aes(x = variable, y = gfp_mean), stat = "identity", fill = "skyblue", alpha = 0.7) +
  geom_errorbar(aes(x = variable, ymin = ymin, ymax = ymax), width = 0.4, colour = "orange", alpha = 0.9, size = 1.3)

方法2：重新计算统计量（更规范） #

从原始数据直接分组计算均值和误差，生成仅含统计结果的3行数据框，彻底避免数据冗余：

# 计算统计量
gfp_stats <- gfp_long %>% 
  ungroup() %>% 
  group_by(variable) %>% 
  summarise(
    gfp_mean = mean(gfp),
    stdev = sd(gfp),
    ymax = gfp_mean + stdev,
    ymin = gfp_mean - stdev
  )

# 绘图
ggplot(gfp_stats) +
  geom_bar(aes(x = variable, y = gfp_mean), stat = "identity", fill = "skyblue", alpha = 0.7) +
  geom_errorbar(aes(x = variable, ymin = ymin, ymax = ymax), width = 0.4, colour = "orange", alpha = 0.9, size = 1.3)

方法3：临时取消分组（快速修复） #

不想修改数据的话，可在绘图时临时取消分组，并指定只绘制唯一值：

ggplot(ungroup(gfp_long)) +
  geom_bar(aes(x = variable, y = gfp_mean), stat = "identity", fill = "skyblue", alpha = 0.7) +
  geom_errorbar(aes(x = variable, ymin = ymin, ymax = ymax), 
                width = 0.4, colour = "orange", alpha = 0.9, size = 1.3,
                stat = "unique")

原始问题相关信息 #

异常图表 #

原始数据 #

# A tibble: 9 × 6
# Groups:   variable [3]
  variable       gfp   stdev gfp_mean     ymax     ymin
  <chr>        <dbl>   <dbl>    <dbl>    <dbl>    <dbl>
1 gfp_sums   4286898 466912. 4478746  4945658. 4011834.
2 nc_sums     664845   4378.  662308.  666686.  657930.
3 media_sums  778269  29403.  744335   773738.  714932.
4 gfp_sums   5011021 466912. 4478746  4945658. 4011834.
5 nc_sums     657253   4378.  662308.  666686.  657930.
6 media_sums  726416  29403.  744335   773738.  714932.
7 gfp_sums   4138319 466912. 4478746  4945658. 4011834.
8 nc_sums     664827   4378.  662308.  666686.  657930.
9 media_sums  728320  29403.  744335   773738.  714932.

原始绘图代码 #

ggplot(gfp_long)+
  geom_bar( aes(x=variable, y=gfp_mean), stat="identity", fill="skyblue", alpha=0.7) +
  geom_errorbar( aes(x=variable, ymin=ymin, ymax=ymax), width=0.4, colour="orange", alpha=0.9, size=1.3)

数据框dput内容 #

structure(list(variable = c("gfp_sums", "nc_sums", "media_sums", 
"gfp_sums", "nc_sums", "media_sums", "gfp_sums", "nc_sums", "media_sums"
), gfp = c(4286898, 664845, 778269, 5011021, 657253, 726416, 
4138319, 664827, 728320), stdev = c(466911.593911524, 4378.05634195511, 
29403.1217900413, 466911.593911524, 4378.05634195511, 29403.1217900413, 
466911.593911524, 4378.05634195511, 29403.1217900413), gfp_mean = c(4478746, 
662308.333333333, 744335, 4478746, 662308.333333333, 744335, 
4478746, 662308.333333333, 744335), ymax = c(4945657.59391152, 
666686.389675288, 773738.121790041, 4945657.59391152, 666686.389675288, 
773738.121790041, 4945657.59391152, 666686.389675288, 773738.121790041
), ymin = c(4011834.40608848, 657930.276991378, 714931.878209959, 
4011834.40608848, 657930.276991378, 714931.878209959, 4011834.40608848, 
657930.276991378, 714931.878209959)), class = c("grouped_df", 
"tbl_df", "tbl", "data.frame"), row.names = c(NA, -9L), groups = structure(list(
    variable = c("gfp_sums", "media_sums", "nc_sums"), .rows = structure(list(
        c(1L, 4L, 7L), c(3L, 6L, 9L), c(2L, 5L, 8L)), ptype = integer(0), class = c("vctrs_list_of", 
    "vctrs_vctr", "list"))), class = c("tbl_df", "tbl", "data.frame"
), row.names = c(NA, -3L), .drop = TRUE))

内容的提问来源于stack exchange，提问作者Salt_Industry