Association and causal claims
Two variables may be associated because of a third variable, selection or chance. A causal argument needs design and mechanism evidence beyond a rising trend or a strong correlation. Work through the example in two forms, such as an annotated calculation and a labeled diagram, table or verbal explanation. Name the assumption that makes the operation valid, and compare a tempting incorrect approach with the justified method. Keep an exact value until the problem asks for rounding; a final explanation should connect the result to the quantities being compared.
A worked example
A scatter plot slopes upward as x increases. What association does it suggest? Worked solution: Use the displayed evidence and model conditions. A statistical association or fitted rule has limits and does not by itself establish causation. The result can be checked against the stated quantities and units before it is used in a recommendation.
- Explore association and causal claims using the worked scenario. Sketch or tabulate the quantities, label the unknown, and predict the kind of answer before calculating.
- Resolve this new case without copying the model: A strong association appears between ice-cream sales and swimming. What must a causal claim consider? Show each equality or representation change and explain its purpose to a partner or in writing.
- Complete the two practice questions about association and causal claims. For one tempting wrong choice, describe the likely reasoning error and construct a small counterexample.
- Create and solve an original association and causal claims problem with new quantities. State its assumptions, include a representation, and check that the requested answer has appropriate units.
- Answer the independent check questions without viewing the worked solution. Then compare methods, revise one explanation, and record which idea needs another example.
Open the printable learning material