如何在ggplot中计算双Y轴刻度比例并修复图表异常

以下是可直接运行的完整解决方案，同时针对你遇到的X轴顺序错误、折线不显示、次Y轴异常问题逐一修复：

# 1. 准备数据集并修正X轴因子顺序
QY = c("3Q22", "4Q22", "1Q23", "2Q23", "3Q23")
shipment = c(50400.10, 60268.39, 46853.50, 45867.33, 50555.89)
growth =  c(0.15759993, 0.19579914, -0.22258583, -0.02104799, 0.10222003)

# 关键：指定factor的levels强制保留原始季度顺序，避免X轴乱序
df <- data.frame(QY = factor(QY, levels = QY), shipment, growth)

# 2. 自定义主题（保留你的原代码）
mystyle <- function(){ 
  font <- "Calibri"   
  
  theme_bw() %+replace%    
    theme(
      panel.grid.major.x = element_blank(),
      panel.grid.minor = element_blank(),
      panel.border = element_blank(),
      axis.line = element_line(color = 'black'),
      plot.title = element_text(
        family = font,
        size = 12,
        face = 'bold',
        hjust = 0,
        vjust = 2),
      plot.subtitle = element_text(
        family = font,
        size = 10),
      plot.caption = element_text(
        family = font,
        size = 9,
        hjust = 1),
      axis.title = element_text(
        family = font,
        face = "bold",
        size = 10),
      axis.text = element_text(
        family = font,
        size = 9),
      axis.text.x = element_text(
        margin=margin(5, b = 10)),
      axis.title.x = element_blank(),
      legend.text = element_text(
        family = font,
        size = 9),
      legend.title = element_blank()
    )
}

# 3. 计算双Y轴转换系数（核心逻辑）
# 主Y轴（shipment）的数值范围
min_ship <- min(df$shipment)
max_ship <- max(df$shipment)
# 次Y轴（growth）的数值范围
min_growth <- min(df$growth)
max_growth <- max(df$growth)
# 计算转换系数：将growth的范围映射到shipment的范围
coef <- (max_ship - min_ship) / (max_growth - min_growth)
# 转换growth数据，使其落在shipment的数值区间内（解决折线不显示问题）
df$growth_transformed <- (df$growth - min_growth) * coef + min_ship

# 4. 绘制最终图表
library(ggplot2)

ggplot(df, aes(x = QY)) +
  # 主Y轴：柱状图（对应shipment数据）
  geom_col(aes(y = shipment, fill = "Shipment"), width = 0.6) +
  # 次Y轴：折线图（对应转换后的growth数据）
  geom_line(aes(y = growth_transformed, color = "Growth"), group = 1, size = 1.2) +
  geom_point(aes(y = growth_transformed, color = "Growth"), size = 3) +
  # 设置主、次Y轴刻度与标签
  scale_y_continuous(
    name = "Shipment Volume",
    # 次Y轴：将主Y轴刻度反向转换回growth原始值，格式化为百分比
    sec.axis = sec_axis(
      ~ (. - min_ship)/coef + min_growth,
      name = "Growth Rate (%)",
      labels = function(x) paste0(round(x*100, 1), "%")
    )
  ) +
  # 自定义颜色
  scale_fill_manual(values = "#4a6fe3") +
  scale_color_manual(values = "#e34a6f") +
  # 添加标题与说明
  labs(title = "Quarterly Shipment & Growth Trend",
       subtitle = "3Q22 to 3Q23",
       caption = "Data Source: Sample Dataset") +
  # 应用自定义主题
  mystyle() +
  # 调整图例位置
  theme(legend.position = "top")

关键问题修复与核心逻辑讲解 #

1. X轴因子顺序错误 #

默认factor()会按字母排序（比如"1Q23"会排在"3Q22"前面），解决方法是创建因子时指定levels = QY，强制保留原始季度顺序：

df <- data.frame(QY = factor(QY, levels = QY), shipment, growth)

2. 折线不显示 #

growth的数值范围（-0.22_{0.2）远小于shipment的范围（45867}60268），直接绘制会导致折线贴在X轴上无法看见。通过数据转换将growth映射到shipment的数值区间内，就能让折线正常显示在图表中。

3. 双Y轴刻度比例计算逻辑（重点） #

ggplot本身不支持原生双Y轴，必须通过「数据转换+刻度反向映射」实现，具体步骤：

1. 确定两个轴的数值范围：分别取shipment和growth的最小值、最大值
2. 计算转换系数：

   coef <- (max_ship - min_ship) / (max_growth - min_growth)
   

   这个系数的作用是把growth的数值范围拉伸/压缩到和shipment完全一致的区间。
3. 转换growth数据：

   df$growth_transformed <- (df$growth - min_growth) * coef + min_ship
   

   公式逻辑：先把growth的最小值移到0，乘以系数拉伸到shipment的范围，再加上shipment的最小值，确保转换后的数据刚好落在shipment的min和max之间。
4. 次Y轴刻度反向转换：
   在sec_axis()中，用主Y轴的刻度值反向计算出对应的growth原始值，再格式化为百分比，让次Y轴显示真实的增长率：

   ~ (. - min_ship)/coef + min_growth
   

内容的提问来源于stack exchange，提问作者Helmasan