scientific-critical-thinking

To critique a research paper, systematically inspect its experimental design, statistical validity, and evidence quality to identify biases like selection or confounding. This structured review process yields severity-ranked concerns, strengths, and actionable recommendations proportional to the data quality.

Checking for p-hacking and statistical validity issues requires analyzing power, effect sizes, multiple comparisons, and missing data. Identifying these red flags separates supported conclusions from weak reasoning and ensures statistical conclusion validity.

Yes, you can grade evidence quality using GRADE and Cochrane-style risk of bias frameworks. The process applies evidence hierarchies and structured logic to assess internal, external, and construct validity, producing evidence-strength reasoning consistent with established scientific evaluation principles.

To identify fallacies in scientific arguments, apply a bias taxonomy check against cognitive, reporting, and analysis biases within the study. This reveals flawed reasoning and methodological threats, allowing you to separate robust scientific claims from weak or unsupported conclusions.

Yes, Python and the requests library are required dependencies to run the systematic assessment of experimental design and confounders. This environment supports the structured methodology critique and statistical review needed to evaluate study validity.