Biocultural vulnerability of traditional agricultural knowledge and systems: a quantitative meta-analysis
Description
Background. The dataset was compiled through a PRISMA 2020-compliant systematic review of Scopus and Web of Science Core Collection (search period: inception–2024). Forty-eight primary studies were screened, coded, and classified into a four-tier evidence hierarchy (Tier 1: experimental/quasi-experimental; Tier 2: quantitative observational; Tier 3: mixed-methods; Tier 4: qualitative-only). Six functional dimensions of biocultural vulnerability were assessed: intergenerational erosion (V1), biocultural complexity (V2), territorial singularity (V3), documentation status (V4), legal vulnerability (V5), and social organization (V6). Contents. The repository is organized into three folders: (1) 1-DATABASE contains the extraction database with coded variables and effect sizes for all included studies, the qualitative coding spreadsheet with ordinal-to-lnRR conversion parameters, the PRISMA flow diagram data, and the consolidated BibTeX file; (2) 2-SCRIPTS contains 14 R and Python scripts that reproduce the full pipeline, from data extraction and tier-specific variance inflation through random-effects meta-analysis, multivariate meta-regression (geographic congruence, community type, intervention mode), publication bias diagnostics (Egger, Begg, trim-and-fill), and sensitivity analyses (tier, rho, DL vs. REML); (3) 3-RESULTS contains 11 spreadsheets with summary meta-analytic outputs per dimension, meta-regression coefficients, leave-one-out diagnostics, subgroup analyses, and vulnerability rankings. Software. Analyses were conducted in R (v4.3+) using the packages metafor, dmetar, brms, and ggplot2, among others. The script 00_setup.R installs all required dependencies.