Module 01 Summary

What this module established

A statistical project needs one folder a data project does not: the analysis plan, dated before the analysis. Choices made after seeing the data are unfalsifiable, because a version of them always exists that gives the answer you hoped for.

Carry forward

• A notebook remembers what you ran but not the order, and a variable whose cell you deleted is still in memory. Restart and run all before believing any notebook result.
• np.random.seed() sets process-wide state that any library call can consume. np.random.default_rng(seed) keeps the state in an object you pass explicitly.
• scipy runs a test; statsmodels builds a model you can add covariates to. For a two-group comparison they give the same answer, and only one of them grows.
• Every exclusion is a rule with a reason and a printed cost. The one in the lesson moved both group means and shrank the gap by five points.

Before moving on

Move on when your plan is dated, your seed is recorded, and your results table contains every comparison you ran.

Mark complete