Package {gendertext}

Title:	Detect Gendered Words in Text and Suggest Neutral Alternatives
Version:	0.1.0
Description:	Identifies gendered words and phrases in text using a built in dictionary of more than two hundred gendered terms paired with gender neutral alternatives. Reports the share of gendered language in a text, lists every gendered term found together with its suggested neutral replacement, and can rewrite a text in gender neutral form. Plain text files are read with base R, while other document formats such as PDF and Word are supported through the optional 'readtext' package. The dictionary is informed by published guidance on gender inclusive language, including the United Nations guidelines https://www.un.org/en/gender-inclusive-language/ and the European Parliament guidance on gender neutral language.
License:	MIT + file LICENSE
URL:	https://github.com/mashrur-ayon/gendertext
BugReports:	https://github.com/mashrur-ayon/gendertext/issues
Depends:	R (≥ 3.5)
Suggests:	knitr, readtext, rmarkdown, testthat (≥ 3.0.0)
VignetteBuilder:	knitr
Config/testthat/edition:	3
Encoding:	UTF-8
Language:	en-GB
LazyData:	true
RoxygenNote:	7.3.3
NeedsCompilation:	no
Packaged:	2026-06-11 14:36:06 UTC; mashr
Author:	S M Mashrur Arafin Ayon [aut, cre, cph], Rodaba Zaman Adrita [aut, dtc] (Word collection and gender dictionary curation)
Maintainer:	S M Mashrur Arafin Ayon <mashrur399@gmail.com>
Repository:	CRAN
Date/Publication:	2026-06-18 14:30:02 UTC

gendertext: Detect Gendered Words in Text and Suggest Neutral Alternatives

Description

Tools for identifying gendered language in text and in documents, measuring how much of a text is gendered, and suggesting or applying gender neutral alternatives. The package follows a transparent, dictionary based approach built around the gender_dictionary dataset.

Details

The main functions are:

• gender_score(): share of gendered language in a text or file.

• gender_suggestions(): detected gendered terms with suggested gender neutral alternatives.

• gender_replace(): rewrite a text using the neutral alternatives.

• read_text(): read a document into a single character string.

Author(s)

Maintainer: S M Mashrur Arafin Ayon mashrur399@gmail.com (ORCID) [copyright holder]

Authors:

• Rodaba Zaman Adrita Zamanrodaba@gmail.com (Word collection and gender dictionary curation) [data contributor]

See Also

Useful links:

• https://github.com/mashrur-ayon/gendertext

• Report bugs at https://github.com/mashrur-ayon/gendertext/issues

Dictionary of Gendered Terms and Gender Neutral Alternatives

Description

A curated dictionary of gendered English words and phrases, each paired with a suggested gender neutral alternative. The dictionary covers gendered occupational titles (for example "chairman" and "stewardess"), gendered pronouns, forms of address, family and relationship terms, and common idioms and compounds built on gendered words. It powers gender_score(), gender_suggestions(), and gender_replace().

Usage

gender_dictionary

Format

A data frame with 208 rows and 2 variables:

gendered

Character. A gendered word or phrase, in lower case.

neutral

Character. The suggested gender neutral alternative.

Details

All entries are stored in lower case. Matching in the package functions is case insensitive and tolerant of possessive forms, so "Chairman's" in a text is matched by the entry "chairman".

The selection of terms and replacements is informed by published guidance on gender inclusive language, including the United Nations guidelines for gender inclusive language in English and the European Parliament guidance on gender neutral language.

Source

Curated by the package authors, informed by the United Nations guidelines for gender inclusive language (https://www.un.org/en/gender-inclusive-language/) and the European Parliament guidance on gender neutral language (https://www.europarl.europa.eu/cmsdata/151780/GNL_Guidelines_EN.pdf).

Examples

data(gender_dictionary)
head(gender_dictionary)
nrow(gender_dictionary)

Rewrite Text with Gender Neutral Alternatives

Description

Replaces gendered terms and phrases in a text or in a file with the gender neutral alternatives from the built in dictionary gender_dictionary or from a user supplied dictionary. Longer phrases are replaced before shorter ones and matching is case insensitive. The capitalisation of each replacement follows the matched text: an all caps match yields an all caps replacement and a match starting with a capital letter yields a capitalised replacement.

Usage

gender_replace(text = NULL, path = NULL, dictionary = NULL)

Arguments

Details

The function performs plain dictionary substitution. It does not adjust the surrounding grammar, so a replacement such as "they" for "he" may require manual revision of verb forms. It is intended as a drafting aid, not a fully automatic rewriter.

Value

A length one character string containing the rewritten text.

See Also

gender_suggestions() to preview the replacements, gender_score() for an overall share, and gender_dictionary for the built in dictionary.

Examples

gender_replace(text = "The chairman called the policeman.")

# Capitalisation is preserved
gender_replace(text = "Chairman Smith spoke. THE FIREMAN AGREED.")

# Use a custom dictionary
my_dict <- data.frame(gendered = "dude", neutral = "person")
gender_replace(text = "Hey dude!", dictionary = my_dict)

Gendered Language Score

Description

Computes the share of gendered language in a text or in a file, based on the built in dictionary gender_dictionary or on a user supplied dictionary. Multi word phrases are matched before single words and each piece of text is counted at most once, so a phrase such as "ladies and gentlemen" is never counted again as "ladies" plus "gentlemen".

Usage

gender_score(
  text = NULL,
  path = NULL,
  unit = c("tokens", "matches"),
  dictionary = NULL
)

Arguments

Details

The reported neutral share is a proxy. It is the proportion of tokens that are not matched by any dictionary entry, not a comprehensive linguistic measure of neutrality. When unit = "tokens", a matched multi word phrase contributes one gendered unit for every word it spans, so the gendered and neutral percentages always sum to 100.

Value

A data frame with one row and the following columns:

total_units

Total number of units counted (tokens or matches, depending on unit).

gendered_units

Number of gendered units detected.

neutral_units

For unit = "tokens", the difference total_units - gendered_units. Otherwise NA.

gendered_percent

Percentage of gendered units.

neutral_percent

Percentage of unmatched (proxy neutral) units, or NA for unit = "matches".

See Also

gender_suggestions() to list the terms behind the score, gender_replace() to rewrite the text, and gender_dictionary for the built in dictionary.

Examples

# Direct text input
gender_score(text = "The chairman said he will call the policeman.")

# Count matches only
gender_score(text = "The chairman spoke.", unit = "matches")

# Analyse a file shipped with the package
txt <- system.file("extdata", "test.txt", package = "gendertext")
gender_score(path = txt)

# Use a custom dictionary
my_dict <- data.frame(
  gendered = c("dude"),
  neutral = c("person")
)
gender_score(text = "Hey dude!", dictionary = my_dict)

Gender Neutral Suggestions for Detected Gendered Terms

Description

Identifies the gendered terms and phrases that occur in a text or in a file, using the built in dictionary gender_dictionary or a user supplied dictionary, and returns the suggested gender neutral alternative for each detected term, optionally with occurrence counts.

Usage

gender_suggestions(
  text = NULL,
  path = NULL,
  include_counts = TRUE,
  dictionary = NULL
)

Arguments

Value

A data frame with one row per detected term, sorted by decreasing count and then alphabetically:

gendered

Detected gendered term or phrase from the dictionary.

suggested_neutral

Suggested gender neutral replacement.

count

Number of occurrences in the text (only when include_counts = TRUE).

See Also

gender_score() for an overall share, gender_replace() to apply the suggestions, and gender_dictionary for the built in dictionary.

Examples

gender_suggestions(text = "Our chairman said he will email the mailman.")

# Without counts
gender_suggestions(
  text = "The fireman and the policeman arrived.",
  include_counts = FALSE
)

# Analyse a file shipped with the package
txt <- system.file("extdata", "test.txt", package = "gendertext")
head(gender_suggestions(path = txt))

Read Text from a File

Description

Reads text content from a document and returns it as a single character string for downstream analysis in gender_score(), gender_suggestions(), and gender_replace(). Plain text files (extensions txt, text, md, and rmd) are read with base R. Other formats such as pdf, docx, rtf, odt, csv, and json are read with the suggested 'readtext' package when it is installed.

Usage

read_text(path)

Arguments

Value

A length one character string containing the extracted text.

See Also

gender_score(), gender_suggestions(), gender_replace()

Examples

txt <- system.file("extdata", "test.txt", package = "gendertext")
substr(read_text(txt), 1, 60)

if (requireNamespace("readtext", quietly = TRUE)) {
  pdf <- system.file("extdata", "test.pdf", package = "gendertext")
  substr(read_text(pdf), 1, 60)
}