02 / ANALYSIS
Every conclusion has a trail.
Choose a test case and run it. The workflow may pause to confirm the study design.
Planner → Checks → SciPy → Interpreter → Review
Your input is needed
Independent: different units per row. Paired: repeated measurements on the same unit. Clustered: units grouped by class or site. This MVP supports independent observations only.
Observational: no confirmed random assignment. Randomized: treatment was actually assigned at random. Use Unknown when the study background is unavailable.
Date window
Optional. Leave No date filter selected for all rows. Date format follows your browser language.
Analysis plan
Checks & warnings
Actual execution trail
Dataset preview
Continue this session
Choose a new analysis when changing variables, dates or the research question.