基于R ggplot绘制分试验得分区间计数直方图的问题求助

我有一个包含三列的数据集：参与者编号、试验编号、参与者在该试验中的得分。数据集涉及100名参与者与160次试验，我希望绘制一幅直方图，统计每个试验中得分处于0-2、2-4、4-6……以及大于16区间的参与者数量。

我尝试了如下代码：

max(scores.data$value)
  
scores.data %>% ggplot(aes(score, fill = trial)) + 
  geom_bar(color = NA, position = position_dodge()) +
  scale_y_continuous(expand = c(0,0), limits = c(0, 100), breaks = seq(0, 100, by = 5)) +
  scale_x_binned(limits = c(0,45), breaks = c(2,4,6,8,10,12,14,16,45), labels = c("2","4","6","8","10","12","14","16","45")) + 
  labs(x = "time bins per trial",
       y = "count"
    
  ) + 
  facet_grid(~trial, scales = "free_x", switch = "x") +
  theme_classic() + 
  theme(axis.text.x = element_blank(),
        axis.ticks.x = element_blank(),
        legend.position = "none",
        panel.grid.major.y = element_line(color = "lightgray", size = 0.25),
        panel.spacing    = unit(0, "points"),
        strip.background = element_blank(),
        strip.text       = element_blank())

但生成的图不符合预期：

以下是可复现数据：

structure(list(participant = c("10000", "10000", "10000", "10000", 
"10000", "10000", "10000", "10000", "10000", "10000"), trial = 
c("1", "2", "3", "4", "5", "6", "7", "8", "9", "10"), score = 
c(5.409, 4.079, 4.355, 4.245, 3.43, 3.685, 4.808, 3.256, 7.038, 
3.714)), row.names = c(NA, -10L), class = c("tbl_df", "tbl", 
"data.frame"))

原代码的核心问题：geom_bar搭配position_dodge会错误错开同区间的柱子，且facet_grid的设置隐藏了试验标签，未正确展示每个试验的得分区间分布。以下是两种可行的修正方案：

方法1：手动分箱统计后绘图（可控性强） #

先对数据按试验和得分区间做统计，再用柱状图展示：

library(tidyverse)

# 定义得分区间与对应标签
score_bins <- c(0, 2, 4, 6, 8, 10, 12, 14, 16, Inf)
bin_labels <- c("0-2", "2-4", "4-6", "6-8", "8-10", "10-12", "12-14", "14-16", ">16")

# 分箱并统计每个试验各区间的参与者数量
scores_summary <- scores.data %>%
  mutate(score_bin = cut(score, breaks = score_bins, labels = bin_labels, right = FALSE)) %>%
  group_by(trial, score_bin) %>%
  summarise(count = n(), .groups = "drop") %>%
  # 补全无数据的区间，保证所有子图x轴一致
  complete(trial, score_bin, fill = list(count = 0))

# 绘图
ggplot(scores_summary, aes(x = score_bin, y = count, fill = score_bin)) +
  geom_col(color = "black", width = 0.8) +
  # 按试验分面，可根据试验总数调整nrow/ncol
  facet_wrap(~trial, nrow = 10) +
  scale_y_continuous(expand = c(0, 0), limits = c(0, 100), breaks = seq(0, 100, 5)) +
  labs(x = "得分区间", y = "参与者数量", title = "各试验得分区间参与者分布") +
  theme_classic() +
  theme(
    legend.position = "none",
    panel.grid.major.y = element_line(color = "lightgray", size = 0.25),
    axis.text.x = element_text(angle = 45, hjust = 1)
  )

方法2：直接用直方图分面（简洁高效） #

无需手动统计，直接通过geom_histogram指定分箱规则后分面：

ggplot(scores.data, aes(x = score)) +
  geom_histogram(
    breaks = score_bins,
    color = "black", fill = "steelblue",
    closed = "left" # 左闭右开区间，匹配0-2包含0、不包含2的需求
  ) +
  facet_wrap(~trial, nrow = 10) +
  scale_y_continuous(expand = c(0, 0), limits = c(0, 100), breaks = seq(0, 100, 5)) +
  scale_x_continuous(breaks = score_bins, labels = bin_labels) +
  labs(x = "得分区间", y = "参与者数量", title = "各试验得分区间参与者分布") +
  theme_classic() +
  theme(
    panel.grid.major.y = element_line(color = "lightgray", size = 0.25),
    axis.text.x = element_text(angle = 45, hjust = 1)
  )

关键调整说明 #

• 明确区间划分：用cut或breaks指定目标区间，right=FALSE/closed="left"确保区间逻辑符合预期。
• 合理分面展示：用facet_wrap替代facet_grid，保留试验标签并灵活控制布局。
• 补全缺失区间：手动统计时用complete补全无数据的区间，避免子图柱子缺失、x轴不一致。
• 移除冗余设置：原代码中fill=trial和position_dodge完全多余，每个分面已对应单个试验，无需额外区分。

内容的提问来源于stack exchange，提问作者AnneZ