Help for package tabbitR

Title:

Weighted Cross-Tabulations Exported to 'Excel'

Version:

0.1.3

Description:

Produces weighted cross-tabulation tables for one or more outcome variables across one or more breakdown variables, and exports them directly to 'Excel'. For each outcome-by-breakdown combination, the package creates a weighted percentage table and a corresponding unweighted count table, with transparent handling of missing values and light, readable formatting. Designed to support social survey analysis workflows that require large sets of consistent, publication-ready tables.

Language:

en-GB

License:

MIT + file LICENSE

Encoding:

UTF-8

Imports:

haven, openxlsx, stats

RoxygenNote:

7.3.3

URL:

https://github.com/smmcandrew/tabbitR

BugReports:

https://github.com/smmcandrew/tabbitR/issues

Suggests:

knitr, rmarkdown

VignetteBuilder:

knitr

NeedsCompilation:

Packaged:

2026-02-05 21:46:01 UTC; siobh

Author:

Siobhan McAndrew [aut, cre]

Maintainer:

Siobhan McAndrew <siobhan.mcandrew@gmail.com>

Repository:

CRAN

Date/Publication:

2026-02-09 13:50:02 UTC

Write Weighted Crosstab Tables To An Excel Workbook

Description

tabbit_excel() produces weighted percentage tables and unweighted counts for one or more outcome variables, optionally including missing responses and/or row percentages. Output is written to an Excel file with one sheet per breakdown variable by default (or all results in a single sheet if by_breakdown = FALSE).

Usage

tabbit_excel(
  data,
  vars,
  breakdown,
  file,
  wtvar,
  row_pct = FALSE,
  decimals = 1L,
  nooverall = FALSE,
  nototal = FALSE,
  missingasrow = FALSE,
  nomissing = FALSE,
  by_breakdown = TRUE,
  sheet_base = "Frequencies",
  ...
)

Arguments

data

A data frame.

vars

Character vector of outcome variable names.

breakdown

Character vector of breakdown variables.

file

Path to the Excel file to create.

wtvar

Name of the weight variable (string). Must be present in data.

row_pct

Logical. If FALSE (default), tables show column percentages. If TRUE, tables show row percentages.

decimals

Integer. Number of decimal places for percentages (0-6; default 1).

nooverall

Logical. If TRUE, suppress the "Overall" column (or "Overall column percentage (valid responses)" table).

nototal

Logical. If TRUE, suppress the "Total percent" row in the percentage table (column mode only).

missingasrow

Logical. If TRUE, include "Response missing" as an explicit row in the main percentage table.

nomissing

Logical. If TRUE, drop missing responses from the unweighted N table (and from the percentage table unless missingasrow = TRUE).

by_breakdown

Logical. If TRUE (default), create one sheet per breakdown variable. If FALSE, stack all results into a single sheet called sheet_base.

sheet_base

Sheet name to use when by_breakdown = FALSE.

...

For future extension.

Details

For each outcome variable in vars and each breakdown variable in breakdown, tabbit_excel():

computes weighted percentages (by column or by row),
optionally adds an overall distribution across breakdowns,
optionally adds a "Total percent" row (column mode only),
handles missing outcomes either as a separate "Response missing" row or by excluding them, with or without a separate "Missing percent" line, and
writes a corresponding unweighted N table including (or excluding) missing responses.

A light formatting layer is applied using the openxlsx package:

table headers: bold, with top and bottom borders,
row labels: bold,
total rows ("Total percent" and "Column totals"): bold, with top and bottom borders, and
missing-percentage row: italic.

Value

Invisibly, the file path of the created workbook (a character string).

Examples

out_file <- tempfile(fileext = ".xlsx")

df <- data.frame(
  courteous = factor(c("Definitely true","Mostly true", NA, "Mostly false")),
  listener  = factor(c("Often","Sometimes","Never", NA)),
  sex       = factor(c("Male","Female","Female","Male")),
  agegrp1   = factor(c("18-34","35-54","18-34","55+")),
  weight = c(1, 1.5, 0.8, 1.2)
)

tabbit_excel(
  data         = df,
  vars         = c("courteous", "listener"),
  breakdown    = c("sex", "agegrp1"),
  file         = out_file,
  wtvar        = "weight",
  row_pct      = FALSE,
  decimals     = 1L,
  nooverall    = FALSE,
  nototal      = FALSE,
  missingasrow = FALSE,
  nomissing    = FALSE
)