From raw data to published results.
Upload your data
Drop a CSV or Excel file and describe your study in plain language. DataDoc reads the structure, interprets each column, and confirms your research question.
Audit and clean
An exhaustive audit flags impossible values, duplicates, and missing patterns. Proposed fixes are shown in full — you approve every change before it applies.
Confirm the plan
DataDoc proposes a statistical model and explains why. Review the variables, assumptions, and approach before any calculation runs.
Read the results
Receive APA-formatted output with clinical interpretation — tables, effect sizes, and assumption checks — ready for your manuscript.
Every number comes from validated software. The AI never touches the math.
The AI
- Reads your dataset and understands your research question
- Flags data quality problems and proposes fixes for your review
- Selects the appropriate statistical approach and explains why
- Writes the clinical interpretation after results are computed
The calculation engine
- Computes every p-value, coefficient, and confidence interval
- Runs the same validated math as established research software
- Same input always produces identical output — no randomness
- Fully separate from the AI and cannot be influenced by it
Statistics you can stand behind.
No AI in your statistics
The calculation engine runs on statsmodels, scipy, and lifelines — no language models. The same dataset produces the same output every time. Verifiable, reproducible, publishable.
You confirm before it runs
The analysis plan is shown in full before any calculation starts. Review the model, the variable assignments, the assumption checks — then confirm. Nothing runs automatically.
Formatted for peer review
Tables follow APA 7th edition. Results include effect sizes, confidence intervals, and assumption check results. The interpretation is in clinical language, not statistical shorthand.