R语言中ftable三维表格的变量重命名与格式化方法问询

Hey there! Let's walk through how to rename your 3-way frequency table variables and format it nicely for your report—beyond just using dnn in xtabs. I'll cover both renaming strategies and formatting tools that work like kable (and even better in some cases).

Rename Variables (Beyond dnn) #

1. Modify dimnames Directly on the ftable Object #

Since an ftable is essentially a matrix with hierarchical dimension names, you can directly overwrite its dimnames attribute to rename both the variables and their levels. This is quick if you already have the table generated:

# First, create your original table
tab_ft <- with(dataset, ftable(xtabs(count ~ dos + sex + edu)))

# Rename the variable labels (the top-level dimension names)
names(dimnames(tab_ft)) <- c("Duration of Stay", "Gender", "Education Level")

# Rename the level labels for each variable to clean up text
dimnames(tab_ft)[[1]] <- c("5-10 Years", "≤5 Years", ">10 Years", "Native Born", "Unknown")
dimnames(tab_ft)[[2]] <- c("Female", "Male")
dimnames(tab_ft)[[3]] <- c("High", "Low", "Medium", "Unknown")

# Check the updated table
tab_ft

2. Use Tidy Data Tools (dplyr + tidyr) for Flexible Renaming #

If you prefer a reproducible, readable workflow (especially if you need to clean up messy level labels), convert the table to a tidy data frame, rename everything, then convert back to a frequency table:

library(dplyr)
library(tidyr)
library(stringr)

# Convert the xtabs result to a data frame, then rename and clean labels
tab_tidy <- as.data.frame(with(dataset, xtabs(count ~ dos + sex + edu))) %>%
  rename(
    `Duration of Stay` = dos,
    Gender = sex,
    `Education Level` = edu,
    Count = Freq
  ) %>%
  mutate(
    # Recode messy level names to user-friendly ones
    `Duration of Stay` = case_match(
      `Duration of Stay`,
      "five-to-ten-years" ~ "5-10 Years",
      "five-years-or-less" ~ "≤5 Years",
      "more-than-ten-years" ~ ">10 Years",
      "native-born" ~ "Native Born",
      "unknown" ~ "Unknown"
    ),
    Gender = str_to_title(Gender),
    `Education Level` = str_to_title(`Education Level`)
  )

# Convert back to ftable if needed, or keep as tidy data for formatting
tab_formatted <- ftable(xtabs(Count ~ `Duration of Stay` + Gender + `Education Level`, data = tab_tidy))

3. Combine dnn with Level Renaming in xtabs #

You already know about dnn for variable names, but you can pair it with factor() to rename levels directly in the xtabs call—one step to clean labels and variable names:

tab <- with(dataset, xtabs(
  count ~ factor(dos, 
                 levels = c("five-to-ten-years", "five-years-or-less", "more-than-ten-years", "native-born", "unknown"),
                 labels = c("5-10 Years", "≤5 Years", ">10 Years", "Native Born", "Unknown")) + 
         factor(sex, levels = c("female", "male"), labels = c("Female", "Male")) + 
         factor(edu, levels = c("high", "low", "medium", "unknown"), labels = c("High", "Low", "Medium", "Unknown")),
  dnn = c("Duration of Stay", "Gender", "Education Level")
))
ftable(tab)

Formatting the Table for Reports #

Once your table is renamed, here are the best tools to make it report-ready—similar to kable but with more flexibility:

1. knitr::kable + kableExtra (Markdown/PDF Reports) #

This is the go-to for R Markdown documents. Convert your ftable to a data frame first, then style it:

library(knitr)
library(kableExtra)

# Convert ftable to a data frame (preserves hierarchical structure)
tab_df <- as.data.frame(tab_formatted)

# Format with kable for clean, readable output
kable(tab_df, 
      caption = "Frequency Distribution by Duration of Stay, Gender, and Education Level",
      col.names = c("Duration of Stay", "Gender", "High Ed", "Low Ed", "Medium Ed", "Unknown Ed")) %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed"), 
                full_width = FALSE,
                position = "left") %>%
  column_spec(3:6, width = "10em") %>%
  row_spec(0, bold = TRUE)

2. gt (Beautiful HTML/PDF Tables) #

The gt package makes stunning, customizable tables with minimal code. It’s great for interactive reports or polished PDFs:

library(gt)

tab_df %>%
  gt() %>%
  tab_header(
    title = md("**Frequency Distribution**"),
    subtitle = "Grouped by Stay Duration, Gender, and Education"
  ) %>%
  fmt_number(columns = 3:6, sep_mark = ",", decimals = 0) %>%
  tab_style(
    style = cell_text(weight = "bold"),
    locations = cells_column_labels()
  ) %>%
  opt_table_lines() %>%
  tab_options(table.width = pct(80))

3. flextable (Word/PDF Export) #

If you need to export to Microsoft Word or a professional PDF, flextable is perfect—it integrates seamlessly with Office formats:

library(flextable)

# Create and style the flextable
tab_flex <- flextable(tab_df) %>%
  set_caption(caption = "Frequency Table: Stay Duration × Gender × Education") %>%
  bold(part = "header") %>%
  autofit() %>%
  colformat_num(j = 3:6, big.mark = ",") %>%
  hline_top(part = "header", border = fp_border(width = 2))

# Export to Word
save_as_docx(tab_flex, path = "education_frequency_table.docx")

# Or print directly in R Markdown for PDF/Word output
tab_flex

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