ggplot使用facet_wrap指定布局时如何去除多余布局序号

我使用ggplot的facet_wrap绘制分面图时，为了指定分面布局编写了如下代码：

df$layout_order <- factor(rep(1:ceiling(nrow(df)/4), each = 4), levels = unique(rep(1:ceiling(nrow(df)/4), each = 4))[1:nrow(df)])

ggplot(df, aes(x = Time, y = nPVI, group = Time, fill = Group)) +
  geom_bar(stat = "identity", position = "dodge", color = "black") +
  labs(title = "nPVI scores by Participant, Time and Group",
       x = "Time",
       y = "nPVI",
       group = "Time") +
  facet_wrap(~ Participant + layout_order, scales = "free", ncol = 4)+
  coord_cartesian(ylim = c(20, 60)) +
  scale_fill_manual(values = c("Oral" = "#86CB92", "Prosody" = "#0E7C7B")) +
  theme_minimal()

代码实现了布局指定功能，但每个分面顶部的参与者（Participant）名称下方会显示布局序号，请问如何去除这些多余序号？

要去掉这些多余的布局序号，试试这两种实用方法：

方法1：自定义分面标签，只保留参与者名称 #

在facet_wrap里添加labeller参数，用自定义函数把layout_order对应的标签内容删掉即可：

# 自定义标签函数，移除layout_order的标签列
custom_label <- function(labels) {
  labels$layout_order <- NULL
  labels
}

ggplot(df, aes(x = Time, y = nPVI, group = Time, fill = Group)) +
  geom_bar(stat = "identity", position = "dodge", color = "black") +
  labs(title = "nPVI scores by Participant, Time and Group",
       x = "Time",
       y = "nPVI",
       group = "Time") +
  facet_wrap(~ Participant + layout_order, scales = "free", ncol = 4,
             labeller = custom_label) +
  coord_cartesian(ylim = c(20, 60)) +
  scale_fill_manual(values = c("Oral" = "#86CB92", "Prosody" = "#0E7C7B")) +
  theme_minimal()

方法2：调整参与者的因子顺序，无需引入layout_order到分面 #

直接把Participant的因子顺序设置为你想要的布局顺序，这样facet_wrap只按Participant分面，自然不会出现多余序号：

# 按layout_order的逻辑生成Participant的显示顺序
participant_order <- df$Participant[order(df$layout_order)]
# 将Participant转换为因子并指定顺序
df$Participant <- factor(df$Participant, levels = unique(participant_order))

ggplot(df, aes(x = Time, y = nPVI, group = Time, fill = Group)) +
  geom_bar(stat = "identity", position = "dodge", color = "black") +
  labs(title = "nPVI scores by Participant, Time and Group",
       x = "Time",
       y = "nPVI",
       group = "Time") +
  facet_wrap(~ Participant, scales = "free", ncol = 4) +
  coord_cartesian(ylim = c(20, 60)) +
  scale_fill_manual(values = c("Oral" = "#86CB92", "Prosody" = "#0E7C7B")) +
  theme_minimal()

这两种方法都能保留你想要的布局效果，同时分面顶部只显示参与者名称，不会再出现多余的布局序号。

内容的提问来源于stack exchange，提问作者Lucie J.