Try it on paper first. A hint, the key idea and the worked solution are here when you need them.
Before a vote on a new park, a newspaper runs an online poll and gets replies, with in favour. A survey company asks a random sample of 1000 adults, with a high response rate, and finds in favour. For each poll, find the usual luck in percentage points, , where is the poll's percentage and its number of replies. Give both to two decimal places. Then decide which poll to trust and explain why the giant poll's tiny luck does not help it.
When two groups reply at different rates and , the share of the first group among the replies is . A larger sample shrinks the luck, about percentage points, but never the bias.
Have you tried it?
Work it out on paper first. It is fine to get stuck or make a mistake: that is how we learn. When you are ready, go through the solution one step at a time, and check each step against your own.
Step 1. The online poll. percentage points.
Step 2. The random poll. percentage points.
Step 3. Which to trust. The online poll is a voluntary response sample, so it is probably biased. Its luck is tiny, so it is very sure of its own answer, but the two polls differ by 18 percentage points. Luck cannot explain that gap. The random poll is fair, so its answer is "about 52%, give or take about 3 points".
Answer: about and percentage points. Trust the random poll.
Watch out. A small band of luck shows precision, not truth. A biased poll can be precise and wrong at the same time.
Log in to join the discussion.
Loading…