meta-analysis-forge

Meta-Analysis Forge

Use this skill when evidence synthesis requires statistical pooling of primary-study effects.

Core Principle

A meta-analysis is valid only when the effect sizes being combined are conceptually and statistically comparable enough for the target inference.

Separate:

• effect-size extraction;
• effect-size conversion;
• dependence among effects;
• model choice;
• heterogeneity interpretation;
• publication-bias diagnostics;
• substantive conclusion.

Intake

Identify:

• outcome construct;
• effect-size metric;
• standard error, confidence interval, p-value, or sample size availability;
• number of studies;
• multiple effects per study;
• study designs;
• expected heterogeneity;
• moderators;
• field norms.
• whether raw, participant-level, sample-level, or harmonized derived data are available.

Load:

• references/effect-sizes.md for effect metrics and extraction.
• references/soil-fauna-carbon-meta.md when the project pools ecological effects on both carbon stocks and carbon fluxes and needs trait or climate moderators without collapsing incompatible outcome families.
• references/ecological-meta-ml-path-model-paradigm.md when the project combines meta-analysis, mixed-effects meta-regression, random forest variable ranking, and PLS-PM/SEM-family path modeling.
• references/high-value-paper-reproducibility-audit.md when a strong published meta-analysis should become a reusable template and the task requires checking code, data-table structure, rma.mv, random forest, PLS-PM/SEM-family modeling, and reproducibility.
• references/ipd-and-mega-analysis.md when the task involves individual participant data, multi-site raw/derived data harmonization, small-sample dataset integration, or mega-analysis.
• references/synthesis-models.md for model choice and diagnostics.
• references/network-meta-analysis.md when comparing three or more treatments across a connected evidence network (frequentist NMA, P-score ranking, and node-splitting inconsistency with an explicit trust check).
• references/ml-moderator-analysis.md for exploratory interpretable-ML moderator analysis (MetaForest/SHAP/EBM/GAM) with the small-k honesty guardrails — complements, never replaces, pre-specified meta-regression.
• references/meta-analysis-quality-gates.md for pre-pooling checks.
• templates/coding-schema.csv and templates/validation-rules.md for machine-readable coding-sheet structure and validation.
• scripts/validate_coding_sheet.py before statistical execution.
• scripts/effect_size_helpers.R for transparent mechanical conversions during extraction.
• scripts/run_meta_analysis.R only after coding validity and pooling appropriateness have been checked.
• scripts/install_r_packages.R when setting up the minimal R environment.