如何将SPSS（sav）数据转换为适配HH::likert的格式？

Tidyverse 解决方案 #

通过宽转长→标签映射→长转宽的三步流程即可完成格式转换，代码如下：

library(tidyverse)

# 原数据集
cheeses <- tribble(
  ~Cheddar, ~Brie, ~Stilton,
  1, 4, 2, 
  2, 4, 1,
  1, 1, 3,
  4, 1, 5,
  3, 2, 4,
  5, 3, 1,
  1, 5, 2
)

# 格式转换
cheeseslikert <- cheeses %>%
  # 把每款奶酪的评分拆分为行数据
  pivot_longer(everything(), names_to = "Question", values_to = "rating") %>%
  # 将数字评分映射为对应文本标签
  mutate(rating_label = case_when(
    rating == 1 ~ "Strongly like",
    rating == 2 ~ "Like",
    rating == 3 ~ "Unsure",
    rating == 4 ~ "Dislike",
    rating == 5 ~ "Strongly dislike"
  )) %>%
  # 按奶酪和标签统计响应数量
  count(Question, rating_label) %>%
  # 把标签转为列，缺失的评分项填充0
  pivot_wider(names_from = rating_label, values_from = n, values_fill = 0) %>%
  # 调整列顺序，匹配目标格式
  select(Question, "Strongly like", "Like", "Unsure", "Dislike", "Strongly dislike")

# 查看转换结果
cheeseslikert

Base R 解决方案 #

无需依赖tidyverse，用基础函数也能实现：

# 原数据集
cheeses <- tribble(
  ~Cheddar, ~Brie, ~Stilton,
  1, 4, 2, 
  2, 4, 1,
  1, 1, 3,
  4, 1, 5,
  3, 2, 4,
  5, 3, 1,
  1, 5, 2
)

# 将宽格式转为长格式
long_data <- stack(cheeses)
colnames(long_data) <- c("rating", "Question")

# 统计交叉频数
freq_table <- table(long_data$Question, long_data$rating)

# 转换为数据框并调整格式
cheeseslikert_base <- as.data.frame.matrix(freq_table)
cheeseslikert_base$Question <- rownames(cheeseslikert_base)
colnames(cheeseslikert_base) <- c("Strongly like", "Like", "Unsure", "Dislike", "Strongly dislike", "Question")
cheeseslikert_base <- cheeseslikert_base[, c("Question", "Strongly like", "Like", "Unsure", "Dislike", "Strongly dislike")]

# 查看转换结果
cheeseslikert_base

验证绘图 #

转换完成后，直接使用你提供的代码即可生成正确的Likert图：

library(HH)
HH::likert(Question~.,
           cheeseslikert,
           positive.order=TRUE,
           as.percent = TRUE,
           main="Cheese preferences.",
           xlab="percentage",
           ylab=""
)

内容的提问来源于stack exchange，提问作者donnek