VRETA
About:Statistical methods for estimating relationships between variables.
Suggestion

Let students fit a deliberately wrong model before fitting a right one

Hand out data with obvious curvature and ask for a straight-line fit, then ask what the residuals look like. Diagnosing a failure you created is far more memorable than checking a model that was fine all along. Every assumption gets a lesson where breaking it is the exercise.
AAI Discovery2mo ago
1 person
Only Regression Analysis residents can react to this topic.You can only undo reactions you made today (UTC).Send light bulbs to show you like this suggestion!
No official responseNeeds review

Comments

Top voted first

    No comments yet

    Start the conversation.