Data and Reproducibility Materials for Governing State-Sponsored Scholarship Programs: A Systematic Review

This repository contains the derived data, reporting materials, tables, figures, and reproducibility code supporting the systematic review entitled “Governing State-Sponsored Scholarship Programmes for Human Capital Development: A Systematic Review of Legal Instruments, Return Migration, and Programme Management.” Review Overview & PRISMA 2020 Selection Flow The review searched Scopus, ProQuest, and ERIC on 1 September 2026. Total records identified: 9,263 (Scopus: 3,006; ProQuest: 3,082; ERIC: 3,175) Pre-screening removals (out-of-range ERIC): 1,367 Internal duplicates removed: 2,912 Unique records screened after deduplication: 4,984 Records excluded at title and abstract screening: 4,115 Reports sought for retrieval: 307 Reports not retrieved: 47 (extension) + 11 (baseline) Reports excluded prior to full-text assessment (preliminary screening): 49 Reports assessed for full-text eligibility: 209 (211 retrieved minus 2 unparseable files) Reports excluded at full-text assessment: 117, documented with explicit substantive rationales and codes in 07_fulltext-excluded-reports.csv Studies included in the systematic review: 166 (74 baseline cohort + 92 extension cohort) Synthesis & Methodological Profile Critical appraisal profile: 60 Low concern (36.1%), 91 Moderate concern (54.8%), and 15 High concern (9.0%) across five appraisal domains adapted from JBI and MMAT principles. Primary research-question distribution: Legal and regulatory frameworks (RQ1): 12 studies Return migration and brain circulation (RQ2): 29 studies Programme management and governance (RQ3): 119 studies Information technology systems (RQ4): 6 studies Governance mechanism coverage across all 166 studies: Monitoring: 68.7% Selection: 65.1% Equity: 54.8% Service obligations: 39.2% Technology: 37.3% Deposited Repository Files (Flat Numbered Structure) All materials are provided as direct, numbered individual files: Protocols & Documentation 01_README.md 02_prisma-2020-checklist.docx 03_database-search-strategy.csv 04_deduplication-report.docx 05_inclusion-exclusion-criteria.csv Screening & Extraction Registers 06_full-text-disposition.csv 07_fulltext-excluded-reports.csv 08_included-studies-extraction.csv 09_quality-appraisal.csv 10_research-question-mapping.csv Source Tables 11_table-1-inclusion-exclusion.csv to 16_table-6-evidence-structure.csv Publication Figures 17_fig-1-prisma-flowchart.pdf 18_fig-1-prisma-flowchart.png 19_fig-2-corpus-descriptives.pdf 20_fig-2-corpus-descriptives.png 21_fig-3-quality-appraisal.pdf 22_fig-3-quality-appraisal.png Reproducibility Code & Verification 23_evidence-structure-analysis.py 24_generate_figures.py 25_LICENSE-DATA.txt 26_LICENSE.txt 27_MANIFEST-SHA256.txt Data Conventions NR means not reported in the source study. NA means not applicable to the study design or question. Quantitative values were transcribed only when explicitly reported in the source article. Methodological Transparency The review protocol was specified and locked internally prior to screening. The review was not prospectively registered in PROSPERO because reviews without direct health outcomes are not accepted by the register. Unretrieved reports were retained under dedicated unresolved retrieval statuses rather than treated as substantive exclusions. No statistical meta-analysis was conducted because the included studies evaluated diverse policy schemes without a shared quantitative estimand. Rights and Licensing Data, documentation, tables, and figures are licensed under the Creative Commons Attribution 4.0 International License (CC BY 4.0). Analysis scripts are provided under the MIT Licence (26_LICENSE.txt). Publisher PDFs, extracted full texts, database-provided abstracts, original licensed database exports, personal data, reviewer identities, local filesystem paths, chat histories, and internal working notes are not redistributed in order to comply with database licensing conditions.