基于给定R数据框绘制两组并排分组条形图求助

步骤1：数据预处理 #

原数据框的数值列均为字符类型，需转换为可计算的数值格式，同时修正数据中的小疏漏（如Biv_E_activity里的"60"补全为"60%"）：

# 转换数据类型
df.new$Correlation_task_persistence <- as.numeric(df.new$Correlation_task_persistence)
df.new$Biv_A_task_persistence <- as.numeric(sub("%", "", df.new$Biv_A_task_persistence))
df.new$Biv_E_task_persistence <- as.numeric(sub("%", "", df.new$Biv_E_task_persistence))
df.new$Correlation_activity <- as.numeric(df.new$Correlation_activity)
df.new$Biv_A_activity <- as.numeric(sub("%", "", df.new$Biv_A_activity))
df.new$Biv_E_activity <- as.numeric(sub("%", "", ifelse(df.new$Biv_E_activity == "60", "60%", df.new$Biv_E_activity)))
# 设置年龄显示顺序，与目标图对齐
df.new$age <- factor(df.new$age, levels = c("12", "18", "24", "30"))

步骤2：转换为长数据格式 #

ggplot更适配长数据结构，使用tidyr和dplyr完成格式转换：

library(tidyr)
library(dplyr)

df_long <- df.new %>%
  pivot_longer(
    cols = -age, 
    names_to = c("metric", "domain"), 
    names_sep = "_",
    values_to = "value"
  ) %>%
  # 修正拆分后的分类名称，匹配原数据逻辑
  mutate(domain = ifelse(domain == "task", "task_persistence", domain))

步骤3：绘制分组条形图 #

用ggplot2实现分面板的分组条形图，还原目标图的核心样式：

library(ggplot2)

ggplot(df_long, aes(x = age, y = value, fill = metric)) +
  # 绘制分组条形，调整间距避免重叠
  geom_col(position = position_dodge(width = 0.8), width = 0.7) +
  # 按领域分面板，y轴自由缩放适配不同数值范围
  facet_wrap(~domain, scales = "free_y") +
  # 设置坐标轴与图例标题
  labs(x = "年龄", y = "数值", fill = "指标类型") +
  # 匹配目标图的灰色系填充色
  scale_fill_manual(values = c("Correlation" = "#636363", "Biv_A" = "#bdbdbd", "Biv_E" = "#f0f0f0")) +
  # 使用简洁主题，去除冗余网格线
  theme_bw() +
  theme(
    panel.grid = element_blank(),
    axis.text = element_text(size = 10),
    axis.title = element_text(size = 12),
    legend.title = element_text(size = 12),
    strip.text = element_text(size = 12)
  )

内容的提问来源于stack exchange，提问作者wooden05