如何在R语言生成的三维交叉表中纳入统计计数？

Hey there! I see exactly where you're stuck—when you use table() on your grouped 3waydata, it's just counting how many times each duration/education/sex combo appears in that data frame (which is 1 for each, since you already grouped and summarised), not pulling in your calculated count values. Let's fix that with the right tool for the job: xtabs().

Step-by-Step Solution #

Your 3waydata is already perfectly structured for this—you just need to tell R to use your count column as the weight for the cross-tabulation. Here's how:

# Use xtabs to build the 3-way table with your custom counts
three_way_table <- xtabs(count ~ duration + sex + education, data = 3waydata)

# Optional: Convert to a standard table object if needed
three_way_table <- as.table(three_way_table)

Why This Works #

The xtabs() function is built specifically for creating contingency tables from pre-summarized data. The formula count ~ duration + sex + education tells R:

• Use the count values to fill the table cells
• Cross-tabulate using duration, sex, and education as the three dimensions

Example to Test #

Let’s use sample data to demonstrate how this comes together:

library(dplyr)

# Sample input data
data <- data.frame(
  duration = c("short", "long", "short", "long", "short"),
  education = c("high", "low", "high", "low", "medium"),
  sex = c("M", "F", "M", "F", "F"),
  number = c(5, 3, 2, 4, 6)
)

# Your existing grouping/summarizing code
3waydata <- data %>% 
  group_by(duration, education, sex) %>% 
  summarise(count = sum(number), .groups = "drop")

Running the xtabs() code above will generate a 3-way table where each cell holds the sum of number for that combination, not just a row count.

Bonus: Make the Table More Readable #

If you want a flattened, easier-to-read output of your 3-way table, wrap the result in ftable():

ftable(three_way_table)

This will print the table in a compact format that’s simpler to parse than the default array structure.

内容的提问来源于stack exchange，提问作者user8132864