R语言数据框 #

一、数据框概述 #

数据框（Data Frame）是R语言中最常用的数据结构，用于存储表格数据。每列可以是不同的数据类型，但同一列必须是相同类型。

二、创建数据框 #

2.1 使用data.frame函数 #

r 
df <- data.frame(
  name = c("张三", "李四", "王五"),
  age = c(25, 30, 22),
  city = c("北京", "上海", "广州"),
  stringsAsFactors = FALSE
)
print(df)

2.2 从向量创建 #

r 
name <- c("张三", "李四", "王五")
age <- c(25, 30, 22)
city <- c("北京", "上海", "广州")

df <- data.frame(name, age, city, stringsAsFactors = FALSE)
print(df)

2.3 从矩阵创建 #

r 
m <- matrix(1:12, nrow = 3, ncol = 4)
df <- as.data.frame(m)
print(df)

2.4 从列表创建 #

r 
my_list <- list(
  name = c("张三", "李四"),
  age = c(25, 30)
)
df <- as.data.frame(my_list)
print(df)

2.5 创建空数据框 #

r 
df <- data.frame()
print(df)

df <- data.frame(
  name = character(),
  age = numeric(),
  city = character()
)
print(df)

三、数据框属性 #

3.1 维度 #

r 
df <- data.frame(
  name = c("张三", "李四", "王五"),
  age = c(25, 30, 22)
)

dim(df)
nrow(df)
ncol(df)
length(df)

3.2 列名和行名 #

r 
df <- data.frame(
  name = c("张三", "李四", "王五"),
  age = c(25, 30, 22)
)

names(df)
colnames(df)
rownames(df)

rownames(df) <- c("A", "B", "C")
print(df)

3.3 结构查看 #

r 
df <- data.frame(
  name = c("张三", "李四", "王五"),
  age = c(25, 30, 22)
)

str(df)
summary(df)
head(df)
tail(df)

四、访问数据 #

4.1 访问列 #

r 
df <- data.frame(
  name = c("张三", "李四", "王五"),
  age = c(25, 30, 22),
  city = c("北京", "上海", "广州")
)

df$name
df[["age"]]
df[[2]]
df["city"]
df[, "name"]
df[, 2]

4.2 访问行 #

r 
df <- data.frame(
  name = c("张三", "李四", "王五"),
  age = c(25, 30, 22)
)

df[1, ]
df[c(1, 3), ]
df[1:2, ]

4.3 访问单元格 #

r 
df <- data.frame(
  name = c("张三", "李四", "王五"),
  age = c(25, 30, 22)
)

df[1, 2]
df[2, "age"]
df$name[1]

4.4 逻辑索引 #

r 
df <- data.frame(
  name = c("张三", "李四", "王五"),
  age = c(25, 30, 22)
)

df[df$age > 24, ]
df[df$name == "张三", ]
df[df$age >= 25 & df$age <= 30, ]
subset(df, age > 24)

五、修改数据框 #

5.1 修改列 #

r 
df <- data.frame(
  name = c("张三", "李四", "王五"),
  age = c(25, 30, 22)
)

df$age <- c(26, 31, 23)
df[["age"]] <- df$age + 1
print(df)

5.2 添加列 #

r 
df <- data.frame(
  name = c("张三", "李四", "王五"),
  age = c(25, 30, 22)
)

df$city <- c("北京", "上海", "广州")
df["gender"] <- c("男", "男", "女")
df <- cbind(df, score = c(85, 90, 78))
print(df)

5.3 删除列 #

r 
df <- data.frame(
  name = c("张三", "李四", "王五"),
  age = c(25, 30, 22),
  city = c("北京", "上海", "广州")
)

df$city <- NULL
df <- df[, -2]
df <- df[, !names(df) %in% c("age")]
print(df)

5.4 添加行 #

r 
df <- data.frame(
  name = c("张三", "李四"),
  age = c(25, 30)
)

new_row <- data.frame(name = "王五", age = 22)
df <- rbind(df, new_row)
print(df)

5.5 删除行 #

r 
df <- data.frame(
  name = c("张三", "李四", "王五"),
  age = c(25, 30, 22)
)

df <- df[-1, ]
df <- df[df$name != "李四", ]
print(df)

六、数据框操作 #

6.1 排序 #

r 
df <- data.frame(
  name = c("张三", "李四", "王五"),
  age = c(25, 30, 22)
)

df[order(df$age), ]
df[order(df$age, decreasing = TRUE), ]
df[order(df$age, df$name), ]

6.2 去重 #

r 
df <- data.frame(
  name = c("张三", "李四", "张三"),
  age = c(25, 30, 25)
)

unique(df)
df[!duplicated(df), ]
df[!duplicated(df$name), ]

6.3 合并数据框 #

r 
df1 <- data.frame(
  id = c(1, 2, 3),
  name = c("张三", "李四", "王五")
)

df2 <- data.frame(
  id = c(1, 2, 4),
  score = c(85, 90, 78)
)

merge(df1, df2, by = "id")
merge(df1, df2, by = "id", all = TRUE)
merge(df1, df2, by = "id", all.x = TRUE)
merge(df1, df2, by = "id", all.y = TRUE)

6.4 分组统计 #

r 
df <- data.frame(
  group = c("A", "A", "B", "B", "A"),
  value = c(10, 20, 30, 40, 50)
)

aggregate(value ~ group, data = df, mean)
aggregate(value ~ group, data = df, sum)
tapply(df$value, df$group, mean)

七、处理缺失值 #

7.1 检查缺失值 #

r 
df <- data.frame(
  name = c("张三", "李四", NA),
  age = c(25, NA, 22)
)

is.na(df)
colSums(is.na(df))
rowSums(is.na(df))
anyNA(df)

7.2 删除缺失值 #

r 
df <- data.frame(
  name = c("张三", "李四", NA),
  age = c(25, NA, 22)
)

na.omit(df)
df[complete.cases(df), ]
df[!is.na(df$name), ]

7.3 替换缺失值 #

r 
df <- data.frame(
  name = c("张三", "李四", NA),
  age = c(25, NA, 22)
)

df$age[is.na(df$age)] <- mean(df$age, na.rm = TRUE)
df$name[is.na(df$name)] <- "未知"
print(df)

八、数据框转换 #

8.1 转置 #

r 
df <- data.frame(
  name = c("张三", "李四"),
  age = c(25, 30)
)

t(df)

8.2 转换为矩阵 #

r 
df <- data.frame(
  a = c(1, 2, 3),
  b = c(4, 5, 6)
)

as.matrix(df)
data.matrix(df)

8.3 转换为列表 #

r 
df <- data.frame(
  name = c("张三", "李四"),
  age = c(25, 30)
)

as.list(df)
split(df, df$name)

九、apply函数族 #

9.1 apply #

r 
df <- data.frame(
  a = c(1, 2, 3),
  b = c(4, 5, 6)
)

apply(df, 1, sum)
apply(df, 2, sum)
apply(df, 2, mean)

9.2 lapply和sapply #

r 
df <- data.frame(
  a = c(1, 2, 3),
  b = c(4, 5, 6)
)

lapply(df, sum)
sapply(df, mean)
sapply(df, class)

十、实践示例 #

10.1 学生成绩管理 #

r 
students <- data.frame(
  id = 1:5,
  name = c("张三", "李四", "王五", "赵六", "钱七"),
  math = c(85, 90, 78, 92, 88),
  english = c(80, 85, 90, 75, 82)
)

students$total <- students$math + students$english
students$average <- students$total / 2
students$grade <- ifelse(students$average >= 85, "优秀",
                         ifelse(students$average >= 70, "良好", "及格"))

print(students)

10.2 数据清洗 #

r 
raw_data <- data.frame(
  name = c("张三", "李四", "", "王五"),
  age = c(25, -1, 30, 22),
  score = c(85, 90, NA, 78)
)

raw_data$name[raw_data$name == ""] <- NA
raw_data$age[raw_data$age < 0] <- NA
raw_data <- na.omit(raw_data)

print(raw_data)

10.3 数据汇总 #

r 
sales <- data.frame(
  region = c("北", "南", "北", "南", "北"),
  product = c("A", "A", "B", "B", "A"),
  amount = c(100, 150, 200, 180, 120)
)

aggregate(amount ~ region, data = sales, sum)
aggregate(amount ~ product, data = sales, mean)
aggregate(amount ~ region + product, data = sales, sum)

十一、总结 #

本章学习了：

  • 数据框的创建方法
  • 数据框属性查看
  • 数据访问：列、行、单元格
  • 数据修改：增删改
  • 排序、去重、合并操作
  • 缺失值处理
  • 数据框转换
  • apply函数族的应用

数据框是R语言数据分析的核心数据结构，掌握数据框操作是数据科学的基础！

最后更新：2026-03-27