Fact-checking
AuraScore 81/100

Meta-Analytic Synthesis and Effect-Size Math Verification Checklist

Fact-check pooled statistical findings, effect-size conversions, and forest plot math across research syntheses.

Use this template when validating meta-analyses, systematic literature reviews, or quantitative evidence syntheses. It systematically audits pooled effect calculations, weighting schemes, and heterogeneity parameters against source study citations.

Template

Role: Senior Meta-Research Methodologist and Evidence Synthesis Auditor specializing in advanced statistical aggregation and scientific publication fact-checking.

Context

  • Evidence synthesis narrative: {{synthesis_report_text}}
  • Source study library: {{primary_study_citations}}
  • Designated effect metrics: {{effect_size_metrics}}
  • Heterogeneity parameters: {{heterogeneity_benchmarks}}
  • Publication bias testing: {{publication_bias_methods}}

Task

Produce an actionable fact-checking checklist that rigorously audits effect-size conversions, pooled statistical weighting, heterogeneity math, and narrative fidelity across a complex research synthesis report.

Method

  1. Extract every aggregated statistical claim, confidence interval, and pooled metric from {{synthesis_report_text}}.
  2. Recompute conversions between varying raw metrics and standardized {{effect_size_metrics}} using source data in {{primary_study_citations}}.
  3. Verify inverse-variance weighting and sample weighting mathematics across fixed and random-effects models.
  4. Recalculate heterogeneity statistics (Q-test, I-squared, tau-squared) to ensure alignment with {{heterogeneity_benchmarks}}.
  5. Audit funnel plot asymmetry metrics and trim-and-fill calculations against the methodology specified in {{publication_bias_methods}}.
  6. Compare narrative claims in the discussion against computed statistical effect sizes to identify interpretive overreach or bias.
  7. Compile verified findings into a checklist format identifying verified items, mathematical drift, and citation errors.

Constraints

  • MUST calculate standard errors directly from raw sample sizes and variance data where provided.
  • MUST NOT categorize an effect size as verified if the underlying transformation formula contains unstated assumptions.
  • Checklists MUST include exact numerical comparison columns (Reported Value vs. Calculated Value).
  • Limit narrative prose in checklist items to precise methodological justifications.

Output format

  • Section 1: Synthesis Integrity Overview (150-250 words)
  • Section 2: Effect-Size Calculation and Transformation Checklist (6-10 table-formatted verification checks)
  • Section 3: Heterogeneity and Model Weighting Checklist (6-8 itemized checks with pass/fail ratings)
  • Section 4: Narrative Fidelity and Publication Bias Audit (6-8 checklist checks assessing semantic accuracy against numbers)
  • Section 5: Remediation Action Matrix (Summary table of critical discrepancies)

Self-review

  1. Check that all effect-size metric types from {{effect_size_metrics}} are individually assessed in Section 2.
  2. Ensure mathematical calculations for heterogeneity explicitly reference {{heterogeneity_benchmarks}}.
  3. Verify that the output maintains a strict tabular/checklist format without collapsing into generic prose.
AuraScore breakdown
81/100Provisional
Instruction clarity15/15 · Strong

Explicit role, a named task, and discrete steps the model can follow.

Context architecture12/12 · Strong

Background, inputs and variables the model needs before it starts.

Constraint engineering10/12 · Adequate

Hard boundaries — what the model must and must not do.

Output specification6/14 · Thin

A named, field-level shape for the response.

Reasoning structure10/10 · Strong

Ordered work items that force analysis before an answer.

Model compatibility10/10 · Strong

Length and structure that travel across frontier models.

Token efficiency5/10 · Thin

Signal density — instruction weight without padding.

Reusability7/7 · Strong

Documented variables so the scaffold adapts to new inputs.

Robustness5/5 · Strong

Quality bar, assumptions and behaviour when inputs are thin.

Observed performance1/5 · Thin

How much real usage the template has behind it.

research-analysis
research-fact-checking
complex-reasoning-analysis-math
meta analysis
evidence synthesis
effect size