A fictional turnout report
A fictional municipality compares two districts with the same number of eligible voters. District A records turnout of sixty percent, while District B records forty percent. A newspaper concludes that every low-income resident is less interested in democracy. Its report combines district income averages with official aggregate turnout, but includes no individual interviews or linked household records. A voluntary online poll then asks readers whether voting should be compulsory; most respondents favor the proposal. The poll does not disclose how many people saw the link, whether participants voted more than once or how readers differ from the population. A student panel must prepare a briefing that distinguishes what the observed data show from what the newspaper and poll cannot establish. No real municipality or current election result is being described.
1. What inference is justified, and what is missing?
Read the worked answer
The data establish a district turnout difference. They do not identify individual income effects or motives; linking group averages to every resident is an ecological inference. Linked representative individual evidence and contextual variables would be needed.
2. How should the online poll enter a policy judgment?
Read the worked answer
Treat it as evidence of views among respondents, not a representative estimate of municipal opinion. Evaluate selection and duplicate-response risks, then weigh participation, liberty and unequal burdens with explicit criteria.