Two simple random samples of 60 passengers are selected on different days. In one sample, 39 passengers rate a service as good. In the other, 32 do so.
Calculate the percentage rating the service as good in each sample.
Explain what the difference shows about using samples to infer a population opinion.