01
Estimation, bias and variance
How much can I claim to know from a sample?
Core · Module 08
Statistics & Inference
Probability reasons forward from model to data. Statistics reasons backward — which is much harder, and much more often done badly.
By the end of this module
You can tell a real result from a p-hacked one, and you know what a confidence interval does not mean.
0%0 / 4 complete
02
Hypothesis testing and what p-values are not
What question does a p-value actually answer?
Primary source★ LandmarkFree
ASA Statement on p-Values ↗American Statistical Association
The profession's own correction to how its central tool is used.
03
Regression and the danger of causal language
When does a fitted line license a causal claim?
04
Bayesian inference
Should a parameter have a probability distribution?
ReadingFree
Bayes Rules! An Introduction to Applied Bayesian Modeling ↗Johnson, Ott & Dogucu
Free and complete online.
Quiz
Not written yet for this module. The lectures above are complete and the module still counts toward your progress.