How We Score Studies
Our scoring system evaluates scientific research on four key dimensions, then adjusts for study type to give you a clear picture of evidence quality.
The Four Pillars of Quality
Every study is evaluated on four weighted components that capture different aspects of scientific rigor:
Methodology
Evaluates study design: randomization, blinding, control groups, sample size, and follow-up duration.
Statistical Rigor
Checks for p-values, effect sizes, confidence intervals, and pre-registration of the study.
Reporting Transparency
Assesses conflict of interest disclosure, data availability, and code sharing practices.
Publication Quality
Baseline score for peer-reviewed publication status and journal reputation factors.
What We Look For
Methodology Factors
Full randomization earns maximum points; controlled but non-randomized designs earn partial credit
Double-blind studies earn full points; single-blind earn 60%
Presence of a proper control group for comparison
Uses exponential decay formula—larger samples score higher, with diminishing returns
Studies with documented follow-up periods
Statistical Analysis Factors
Clear reporting of the magnitude of findings
Statistical uncertainty properly quantified
Statistical significance testing included
Study protocol registered before data collection
Transparency Factors
Authors clearly state funding sources and potential conflicts
Raw data accessible for verification
Analysis code shared for reproducibility
Study Type Adjustments
After calculating the base score, we apply a multiplier based on study type. This reflects that not all evidence is equally applicable to human health decisions.
Clinical trials and observational studies in humans
Literature reviews without systematic methodology
Cancer research in animal models
Brain and nervous system research in animals
Other animal model research
Cell and molecular level experiments
Lab dish experiments
Expert opinions without primary research
Why the difference? A perfectly designed mouse study might score 100 on methodology, but that doesn't mean the findings apply to humans. The multiplier helps you understand how directly the evidence relates to human health outcomes.
Interpreting Final Scores
Strong methodology, good statistical reporting, and transparent practices. Human studies in this range represent solid evidence.
Some limitations in design or reporting. Evidence should be considered alongside other studies. Common for animal research.
Significant methodological issues or very preliminary research. Should not be used alone to make health decisions.
Opinions, editorials, or studies with major flaws. Useful for generating hypotheses but not for drawing conclusions.
Our Commitment
Objective Criteria
Every factor in our scoring system is based on established scientific quality indicators. No subjective judgments about whether we "like" the findings.
Consistent Application
The same algorithm scores every study. A nutrition study and a pharmacology study are evaluated by identical criteria.
Full Transparency
Every analyzed study shows the complete score breakdown. You can see exactly why a study received its score.