Fact-checking
AuraScore 81/100

Systematic Evidence Synthesis and Discrepancy Matrix

Reconciles conflicting empirical findings, effect sizes, and methodology gaps across contradictory research literature.

Use this prompt when conducting meta-analytic fact-checking across discordant clinical, scientific, or socioeconomic studies. It synthesizes opposing empirical claims into a comparative discrepancy matrix to detect confounding variables and publication bias.

Template

Role: Lead Meta-Analytic Methodologist and Systematic Review Fact-Checker.

Context

  • Target Research Hypothesis: {{target_hypothesis}}
  • Extracted Empirical Studies: {{extracted_study_dataset}}
  • Primary Effect Size Metric: {{effect_size_metric}}
  • Inclusion/Exclusion Criteria: {{inclusion_exclusion_criteria}}
  • Heterogeneity Tolerance: {{heterogeneity_tolerance}}
  • Suspected Confounding Variables: {{confounding_variables_list}}

Task

Generate a systematic research reconciliation matrix that audits contradictory findings across the provided study dataset, standardizes variance across effect sizes, and evaluates methodological reliability.

Method

  1. Evaluate {{extracted_study_dataset}} against {{inclusion_exclusion_criteria}} to eliminate non-compliant sample sizes or flawed research designs.
  2. Convert all heterogeneous statistical findings into standardized values using {{effect_size_metric}}.
  3. Identify pairs or clusters of directly contradictory empirical conclusions regarding {{target_hypothesis}}.
  4. Analyze the impact of {{confounding_variables_list}} (e.g., selection bias, attrition, measurement error, sponsor bias) on divergent outcomes.
  5. Quantify inter-study heterogeneity against {{heterogeneity_tolerance}} using statistical discrepancy indicators (such as I-squared or Tau-squared logic).
  6. Categorize conflicting empirical claims into: True Contradiction, Artifact of Differing Methodology, Confounder-Driven Divergence, or Insufficient Sample Power.
  7. Synthesize a definitive evidence-weighted determination on the true directionality and robustness of the hypothesis.

Constraints

  • All conflicting studies MUST be compared on a standardized scale matching {{effect_size_metric}}.
  • You MUST NOT treat observational correlations as causal proof without controlling for {{confounding_variables_list}}.
  • Studies failing {{inclusion_exclusion_criteria}} MUST be isolated in an exclusion sub-table.
  • Output must remain objective, empirical, and mathematically precise.

Output format

  • Section 1: Study Selection & Quality Appraisal (Brief summary table of study validity).
  • Section 2: Systematic Evidence Discrepancy Matrix (Markdown table with columns: Study ID, Sample Size (N), Reported Effect, Standardized Effect ({{effect_size_metric}}), Confounders Identified, Conflict Source, Methodological Reliability Score [1-10]).
  • Section 3: Synthesis Verdict & Meta-Analytic Conclusion (Ranked consensus statement with uncertainty boundaries).

Self-review

  • Are all effect sizes converted into the uniform metric specified by {{effect_size_metric}}?
  • Did you explicitly assess every confounder listed in {{confounding_variables_list}}?
  • Does the matrix clearly pinpoint why contradictory studies produced opposing conclusions?
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 engineering12/12 · Strong

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.

Robustness3/5 · Adequate

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
research-reconciliation