cb practice test 5 section 4 question number 15
The key here is that while you can generalize from a random sampling to the exact population that the random sample came from (so you would generally around 78% of adults from this city to be satisfied with air quality), the question asks which things must be true, and all surveys of samples have the potential for error. It’s not likely the 30% of adults in the city overall are satisfied with the air quality and this one random sample just happened to pick people who were outside the average. But doesn’t it seem at least possible that among the city overall the satisfaction is 76%, not 78%? So we simply can’t conclude that the same 78% must apply elsewhere; the answer is none.
how the margin of error became 2%
I am not calculating a margin of error. I’m just saying that surveys can give you a probabilistic idea of a population, but a statement that the survey MUST predict the population perfectly is too strong.