# Crowdsourcing for Language Learning (2020–2022): Published Table Aggregates

Version 1.0. Compiler: P. Taskov, University of Information Science and Technology "St. Paul the Apostle" (UIST), Ohrid, North Macedonia. Prepared with AI-assisted repository transcription and documentation. Source extraction was prepared in October 2026 and the deposit documentation was finalized on 9 October 2026.

## Contents and purpose

166 selected aggregate table cells with explicit response/category labels and source locators from Tables 1a, 1b, 2a, 2b and 3. P1 denotes Spring 2020 and P2 Spring 2022. The source describes student questionnaires and descriptive quantitative/qualitative analysis.

## Source and attribution

Ç. Hatipoğlu, N. Delibegović Džanić, E. Gajek and L. Miloshevska (2022), Crowdsourcing and language learning habits and practices in Turkey, Bosnia and Herzegovina, the Republic of North Macedonia and Poland in the pre-pandemic and pandemic periods.

Source DOI: https://doi.org/10.4312/slo2.0.2022.2.132-183

UIST publication record: https://repository.uist.edu.mk/items/0de15561-ad68-4141-aa70-7268c6fcbe10

The author list credits the source study and published results. This repository package is a derived extraction; no source-author endorsement of the compilation is claimed.

## Compilation and validation

Transcribed printed counts and percentages, added explicit machine-readable category labels, and checked each selected cell against rendered source tables. Table locations: PDF pages 12, 13, 14 and 17 (printed pages 143, 144, 145 and 148). Normalized the malformed printed TUR percentage 72.,1 to 72.1 with an explicit note; retained reported precision and source inconsistencies.

The deposit CSV uses the validated archived bytes from the preparation package. A local checkout changed CSV line endings in two packages without changing any cell values; archived bytes were retained for fixity. SHA256SUMS.txt records the exact deposit bytes.

## Scope and limitations

Partial published aggregate extraction, not respondent-level data or a complete replication package. Source period totals 211 and 185 are not known question-specific denominators. Table 1a RNM total 41 conflicts with prose stating 42; table values are retained. Table 3 total 468 counts multiple-response resource mentions, not unique respondents. Blank cells are omitted, and missing categories, denominators and records are not inferred. The printed page-range citation differs from the DOI/PDF extent.

## Files and use

- crowdsourcing_published_aggregates_v1.csv: UTF-8 comma-separated data with a header row; 166 data rows and 10 columns. No proprietary software is needed.
- DATA_DICTIONARY.md: column definitions and interpretation notes.
- README.md: provenance, methods, scope and limitations.
- LICENSE.md: reuse conditions, source credit and changes.
- SHA256SUMS.txt: SHA-256 checksums of the four data/documentation files.

## Access and reuse

Intended access is open download without registration or embargo. Reuse follows CC BY-SA 4.0 (https://creativecommons.org/licenses/by-sa/4.0/); see LICENSE.md. Public descriptive repository metadata is subject to the repository metadata policy separately.

Suggested citation: Ç. Hatipoğlu, N. Delibegović Džanić, E. Gajek and L. Miloshevska (2022). Crowdsourcing and language learning habits and practices in Turkey, Bosnia and Herzegovina, the Republic of North Macedonia and Poland in the pre-pandemic and pandemic periods. https://doi.org/10.4312/slo2.0.2022.2.132-183. Published aggregate extraction: Crowdsourcing for Language Learning (2020–2022): Published Table Aggregates, version 1.0 (2026). UIST Digital Repository. https://hdl.handle.net/20.500.15029/279.