part 5 6 7 Suppose you want to find out how many people support Policy X. A standard polling approach is to just ask Nmany people whether or not they support Policy X, and take the fraction of people who say yes as an estimate ofthe probability that any one person supports the policy. Suppose that the probability someone supports the policyis p, which you do not know. Let py be the mumber of people polled who supported the policy, divided by the totalnumber of people polled N.1) What distribution of N * PN?2) Show that the expected value of pN is p.3) If I want my estimate to be accurate, I want the error of pn to be small. How many people should I poll toguarantee the erpected squared error on py is less than e?4) How many people should I poll to guarantee the expected squared error on pn is less than e, if I don't knowp?5) Just because the expected error is small doesn't mean the actual error is small. How many people should I pollto guarantee that the actual error on pN is less than e with 90% confidence?6) How many people should I poll to guarantee the actual error on pN is less than e with 90% confidence, if Idon't know p? A potential solution to this is the following: suppose that 'YES' is the embarrassing or socially shameful answer; givethe people you are polling the following instructions: flip a coin privately, and if it comes up heads answer honestly,but if it comes up tails answer 'YES' regardless of what the truth is. This gives people plausible deniability aboutwhy they answered yes, if pressed. Again, let py be the fraction of people who said 'YES'. Let p be the probabilitythat a randomly selected person says 'yes'; let q be the probability that a person's true answer is 'yes'. Note again,we have that the expected value of pN is p, but we are actually interested in measuring q.7) What is the relationship between q and p?8) Construct an estimator ĝy from pN so that the expected value of ĝy is q.9) If I want my estimate to be accurate, I want the error on ĝn to be small. How many people should I poll toguarantee that the expected squared error on qn is less than e, when I don't know the value of q? How doesthis compare to the answer in the previous section?10) How many people should I poll to guarantee the actual error on ĝn is less than e, with 90% confidence, if Idon't know q?11) What is the additional 'cost' of accurate polling if I want to preserve people's privacy in this way?

Question

Accepted Answer

To determine the sample size needed to guarantee a certain level of actual error on the estimated…