如何移除R语言柱状图中数值为0的柱子的轮廓线？

你的分组柱状图里，数值为0的柱子仍显示黑色轮廓线，要解决这个问题，只需让柱子的轮廓色根据final_value是否为0动态调整：当值为0时，将轮廓色设为与背景一致（theme_classic()的背景是白色），非0值则保留黑色轮廓。

修改后的完整代码 #

library(tidyverse)

#generate data
Correlation_task_persistence <-  c("0.29", "0.29","0.24", "0.20", "0.24", "0.04", "0.00", "0.04")
Biv_A_task_persistence <- c("80%", "80%","44%", "44%", "47%", "72%", "49%", "55%")
Biv_E_task_persistence  <- c("20%", "20%", "57%", "56%", "54%", "29%", "52%", "45%")
Correlation_activity <- c("0.29", "0.29", NA, "0.29", "0.08", "0.08", "0.07", "0.00")
Biv_A_activity <- c("80%", "80%", NA, "80%", "45%", "70%", "20%", "50%")
Biv_E_activity <- c("20%", "20%", NA, "20%", "55%", "40%", "90%", "50%")
age <- c("intercept", "slope", "36", "30", "24", "18", "12", "6")
Correlation_emotionality <-  c("0.19", "0.19","0.34", "0.10", "0.13", "0.04", "0.00", "0.04")
Biv_A_emotionality <- c("80%", "80%","43%", "44%", "47%", "71%", "49%", "55%")
Biv_E_emotionality  <- c("20%", "20%", "57%", "56%", "53%", "29%", "51%", "45%")
df.new <- data.frame(Correlation_task_persistence, Biv_A_task_persistence,Biv_E_task_persistence, Correlation_activity, Biv_A_activity, Biv_E_activity, Correlation_emotionality, Biv_A_emotionality, Biv_E_emotionality, age)

#produce the bar plot
df.new %>%
  mutate(across(Correlation_task_persistence:Biv_E_emotionality, 
                ~if_else(as.numeric(gsub("%", "", .x)) > 1,
                         as.numeric(gsub("%", "", .x, fixed = TRUE))/100, 
                         as.numeric(.x)))) %>% 
  pivot_longer(-c(age, contains("Correlation"))) %>% 
  mutate(Correlation = case_when(
                               grepl("task", name) ~ Correlation_task_persistence, 
                               grepl("activity", name) ~ Correlation_activity,
                               grepl("emotionality", name) ~ Correlation_emotionality),
         final_value = Correlation * value,
         name = gsub("_task", "", name)) %>%
  tidyr::extract("name", c("var","group"), regex = "(.*)_([^_]+)$") %>%
  group_by(age, group) %>%
  mutate(label = scales::percent(final_value / sum(final_value))) %>%
  ungroup() %>%
  mutate(age = factor(age,
                      levels = c("0", "6", "12", "18", "24", "30", "36", "intercept", "slope"),
                      ordered = TRUE)) %>%
  ggplot(aes(x = age, y = final_value, fill = var)) +
  # 核心修改：根据final_value是否为0设置轮廓色
  geom_col(aes(color = final_value != 0)) +
  scale_color_manual(values = c("FALSE" = "white", "TRUE" = "black"), guide = "none") +
  theme_classic() +
  geom_text(aes(label = Correlation, group = age), 
            stat = 'summary', fun = function(x) sum(x) + 0.01 * sign(x), size = 3) +
  geom_text(aes(label = label), size = 3, position = position_stack(vjust = 0.5)) +
  scale_fill_grey(start = 0.475, end = 0.8, na.value = "red") +
  labs(y = "Correlation") +
  labs(x = "") +
  theme(legend.position = "bottom",
        legend.title = element_blank()) + 
  facet_wrap(~group)

关键修改点说明 #

1. 修正原代码逻辑错误：将if_else多分支判断改为case_when，避免语法报错
2. 动态设置轮廓色：把geom_col(color="black")改为geom_col(aes(color = final_value != 0))，让轮廓色绑定到final_value是否非0的逻辑值
3. 自定义颜色映射：添加scale_color_manual()，将0值对应的轮廓色设为白色（与背景一致），非0值设为黑色，同时隐藏额外图例

这样处理后，0值的柱子轮廓会和背景完全融合，不会再显示多余的线条。

内容的提问来源于stack exchange，提问作者wooden05