如何用pivot_longer/melt映射AGE与SCORE列转换纵向数据框？

方法一：使用tidyr::pivot_longer（推荐） #

这是最简洁的方式，通过正则匹配直接实现AGE与SCORE列的对应映射：

# 先构造示例数据框（替换成你的实际数据）
df <- data.frame(
  ID = c(1, 2, 3),
  AGE1 = c(10, 12, 11),
  AGE2 = c(13, 14, 15),
  AGE3 = c(16, 17, 18),
  AGE4 = c(19, 20, 21),
  SCORE1 = c(80, 85, 78),
  SCORE2 = c(82, 88, 80),
  SCORE3 = c(85, 90, 83),
  SCORE4 = c(88, 92, 85)
)

library(tidyr)

# 转换为适合折线图的长格式
long_df <- df %>%
  pivot_longer(
    cols = -ID,  # 排除ID列，处理其余所有列
    names_to = c(".value", "wave"),  # .value保留前缀作为新列名，wave存储数字后缀
    names_pattern = "(AGE|SCORE)(\\d)"  # 正则匹配提取前缀（AGE/SCORE）和数字编号
  )

转换后的数据结构：

方法二：使用reshape2::melt #

如果习惯用melt，可以拆分年龄和分数列后合并：

library(reshape2)

# 拆分年龄列
age_df <- melt(df, id.vars = "ID", measure.vars = paste0("AGE", 1:4), 
               variable.name = "wave", value.name = "AGE")
# 拆分分数列
score_df <- melt(df, id.vars = "ID", measure.vars = paste0("SCORE", 1:4), 
                 variable.name = "wave", value.name = "SCORE")

# 去除wave列的前缀，保留数字编号
age_df$wave <- gsub("AGE", "", age_df$wave)
score_df$wave <- gsub("SCORE", "", score_df$wave)

# 合并得到对应映射的长格式
long_df_melt <- merge(age_df, score_df, by = c("ID", "wave"))

绘制折线图示例 #

转换完成后，直接用ggplot绘制折线图：

library(ggplot2)

ggplot(long_df, aes(x = AGE, y = SCORE, group = ID, color = factor(ID))) +
  geom_line(linewidth = 1) +
  geom_point(size = 2) +
  labs(x = "年龄", y = "分数", color = "个体ID") +
  theme_minimal()

内容的提问来源于stack exchange，提问作者bruteforce