'LimeSurvey' '.lss' Questionnaires to and from Word Documents

Turn a 'LimeSurvey' '.lss' survey export into a publication-quality questionnaire document in Word ('.docx') or PDF, with up to four of the survey's own languages side by side. Every label the package adds around that content -- column headers, type names, the audit section -- is written in English, French, German, Spanish or Italian, whatever the survey languages are. A rule-based audit flags missing translations, forward filter references, duplicate codes, array-scale inconsistencies and orphan structural references. Questionnaires travel the other way too: describe one in R, or fill in a Word form, and write a '.lss' file ready to import. Meant for the people who work on questionnaires -- researchers, methodologists, ethics committees, translators and reviewers -- and fully local: the source file is the only input, and no questionnaire content is uploaded to a third-party service.

lssdoc: LimeSurvey questionnaires to and from Word

lssdoc turns a LimeSurvey .lss export into a polished Word (.docx) or PDF questionnaire document for anyone working with a LimeSurvey survey: researchers, survey methodologists, ethics committees, translators and reviewers. It renders the questionnaire content side by side in up to four languages, runs an automated integrity audit, and produces a layout that reads as a published instrument – not a developer dump.

It also works the other way around. Write a questionnaire in a Word form, or describe it in R, and lssdoc writes the .lss file you import into LimeSurvey – so the people who design a questionnaire never have to build it by hand in a web interface.

Two output templates:

• "cards" – one detached block per item (meta table + question table), stacked vertically. Closest to the printed questionnaire a respondent would see.

  

• "table" – one dense table covering the whole document: every variable is one tinted Question row carrying its metadata, followed by one or more white Value rows holding the answer codes and their labels per language. Group banners (with their description, when present), welcome text, survey description and end text become labelled rows of the same table so the document reads as a single artifact.

  

Other things lssdoc takes care of for you:

• Multilingual chrome. Column headers, row labels, type labels (Single choice / Multiple choice / Text / Number …), Value descriptors, the audit section – everything outside the survey content – is translated to English, French, German, Spanish or Italian. Independent from the survey’s content languages, so you can keep your FR/DE questionnaire and switch the document scaffolding to English (or vice versa) with a single argument.
• Variable-centric compound rendering, matching the data file. Every response variable is documented under the exact column name of the LimeSurvey CSV / Excel export, so the variable index lines up with the raw data file column for column: arrays and multiple choice as parent[subq] / parent[other], dual-scale arrays split into one block per scale (parent[subq][1]), two-dimensional arrays as the row x column cross-product (parent[row_col]), ranking expanded into one column per position (parent[answerid]), and list-with-comment into parent[_Ccomment]. Pass variable_names = "underscore" for the Expression Manager / SPSS / Stata code form (parent_subq) instead.
• Methodological type labels. UI distinctions like “List (radio)” vs “List (dropdown)” collapse into “Single choice”, because the response semantics are identical and the actual codes appear in the Value section. The Type cell reads in the methodologist’s vocabulary, not LimeSurvey’s.
• Automated audit. Missing translations, duplicate codes, forward references in routing filters, array-scale inconsistencies, orphan structural references, types missing their required options or subquestions – all flagged inline on the rendered item and, when you ask for it, in a dedicated section. No AI; rule-based and reviewable.
• Local-only processing. No network calls, no third-party services. The questionnaire content never leaves the machine where R runs.

Installation

Install the released version from CRAN:

install.packages("lssdoc")

Or the latest build from R-universe (pre-built, no compiler needed):

install.packages(
  "lssdoc",
  repos = c("https://amaltawfik.r-universe.dev", "https://cloud.r-project.org")
)

Or the development version from GitHub, with pak:

# install.packages("pak")
pak::pak("amaltawfik/lssdoc")

Or with remotes:

# install.packages("remotes")
remotes::install_github("amaltawfik/lssdoc")

The package depends on officer and flextable (declared as Suggests – the parse and audit paths work without them).

Quick tour

Reading, auditing and rendering:

Writing a questionnaire (experimental):

audit_lss(), render_questionnaire() and render_audit() accept input as either a path to a .lss file or a pre-parsed lss object. The output format for the two renderers is inferred from the extension of output (.docx or .pdf).

A one-line pipeline

library(lssdoc)

# Parse the .lss and render the Word document in a single call.
render_questionnaire("survey.lss", "review.docx")

The output uses the "cards" template by default, with chrome_lang following the survey’s primary content language. Open review.docx in Word, click Yes when prompted to update fields (page numbers and the table of contents).

Table layout

render_questionnaire(
  "survey.lss",
  "review.docx",
  languages   = c("en", "fr"),
  template    = "table",
  chrome_lang = "en"
)

languages controls which content columns appear and in which order; the first one is the primary language used for the table of contents and group headings. chrome_lang is independent of languages: keep chrome_lang = "en" to get English column headers and row labels even if your survey content runs in French and German – pass chrome_lang = "fr" (or "de" / "es" / "it") when you want the scaffolding to follow the content.

Ethics committee submission

render_questionnaire(
  "survey.lss",
  "ethics_review.docx",
  languages              = c("en", "fr"),
  template               = "table",
  chrome_lang            = "en",
  show_privacy_settings  = TRUE,   # anonymized / save partial / IP / referrer / timestamp
  show_admin_settings    = TRUE,   # alias / end URL / active
  authors = list(
    list(name        = "Jane Doe",
         affiliation = "HESAV",
         orcid       = "0009-0001-2345-6789"),
    list(name        = "John Doe",
         affiliation = "HESAV",
         orcid       = "0009-0002-3456-7890")
  ),
  description = paste0(
    "Validated as part of the SNSF project XYZ. ",
    "Reference: https://example.org/projects/xyz"
  )
)

The cover page picks up the authors block (with clickable ORCID hyperlinks), the description (with URLs auto-detected and made clickable), and the privacy / admin settings rows in the metadata table.

Minimal questionnaire – just the variables

render_questionnaire(
  "survey.lss",
  "minimal.docx",
  template         = "table",
  show_welcome     = FALSE,
  show_endtext     = FALSE,
  show_description = FALSE,
  show_groups      = FALSE,
  show_toc         = FALSE,
  show_audit       = FALSE,
  show_index       = FALSE
)

Pure data: one row per variable header, one row per value code, nothing else.

Parse once, render many

# Parse a single time, then audit and render different variants
# without re-reading the .lss.
lss   <- read_lss("survey.lss")
audit <- audit_lss(lss)
print(audit)  # console summary of flagged anomalies

render_questionnaire(lss, "review_en.docx", languages = "en")
render_questionnaire(lss, "review_fr.docx", languages = "fr")
render_audit(lss, "qa_followup.docx")

See the audit in action

A deliberately flawed survey ships with the package, so you can see what audit_lss() catches without supplying your own file:

demo <- system.file("extdata", "audit_demo.lss", package = "lssdoc")
audit_lss(read_lss(demo))
#> -- lssdoc audit ----------------------------------------------------
#> 12 findings: 5 errors, 7 warnings, 0 notes.
#> x Survey: duplicate question code 'age'
#> x Question 'age': filter references 'income', asked later (forward reference)
#> x Answer 'X': points to a question id that does not exist
#> x Subquestion 'orphan_sq': points to a question id that does not exist
#> x Question 'blank_q': text empty in every language
#> ! Question 'satisf': expects answer options, but none are defined
#> ! Question 'income': text missing in 'fr'
#> ! Question 'comment ': code carries whitespace
#> ... plus an empty group name, missing subquestions and an array-scale mismatch.

PDF output

# Same call, just pass a .pdf path. The package writes the .docx
# into a temp file and converts it locally via LibreOffice
# (must be installed and on PATH).
render_questionnaire("survey.lss", "review.pdf", template = "table")
render_audit("survey.lss", "qa.pdf")

Write a questionnaire in Word

Many questionnaires are written by people who live in Word, not in R. Hand them a blank form, get it back filled in, and turn it into a LimeSurvey file.

# 1. a blank form, in the language of your choice
lss_template_docx("questionnaire.docx", lang = "fr")

# 2. the author fills it in, in Word

# 3. check it, then read it
check_form_docx("questionnaire.docx")   # every problem at once
spec <- read_form_docx("questionnaire.docx")

# 4. write the file to import into LimeSurvey
write_lss(spec, "questionnaire.lss")

The form is one small table per survey, group, question and quota, carrying the fields each question type needs with its defaults already filled in. Blank forms can also be downloaded from the website, so an author who does not use R never has to install anything.

An existing survey can enter the same loop: write_form_docx() renders it as the form, so a questionnaire can be exported from LimeSurvey, edited in Word and re-imported. See vignette("write-in-word") for the conventions, the question types and the filter syntax.

Templates and palette

The rendered .docx uses an editorial petrol-blue palette tuned for publication-grade output (#133B52 primary, #3A7C8C accent, #E9F2F6 and #F4F8FA for tinted bands, #D3DCE2 for the soft grid). Calibri body / Consolas monospace by default; the font and font_code arguments accept any string to override.

Citation

Run citation("lssdoc") for the up-to-date citation, or cite as:

Tawfik A (2026). _lssdoc: 'LimeSurvey' '.lss' Questionnaires to and
from Word Documents_. doi:10.32614/CRAN.package.lssdoc
<https://doi.org/10.32614/CRAN.package.lssdoc>. R package version
0.3.0, <https://CRAN.R-project.org/package=lssdoc>.

License

MIT. See LICENSE for the full notice.