STAT 301 - Business Statistics
Lecture Review #12 - Confidence Intervals

Review questions:

1) What is a confidence interval? What is it used for?

2) What effect does increasing/decreasing the sample size have on the width of the confidence interval? What is the effect of increasing/decreasing the confidence level?

3) Why can’t we (feasibly) have a 100% confidence interval?

Computational exercises:

1) "Nielsen ratings" are estimates of the percentage of households in the country in which a television set was tuned to a particular program. They are based upon a nationwide sample of households. Let's suppose that Dr. Rasp's favorite television program, Who Wants to Be a Statistician?, had a Nielsen rating of 25 last week. That is, 25% of the households in the sample watched the program. If the Nielsen rating is based upon a sample of 1600 households, what is a 95% confidence interval for the percentage of all households in the country which watched the program?

2) A recent Gallup poll asked a random sample of 800 Florida voters whether they favored passage of a bill, currently being debated by the state legislature, that would name the palmetto bug the official state bird. ("Them critters fly. And they is a big part of the real Floriduh," said the bill's sponsor, State Rep. Balph Snerdwell, a Republicrat from the town of Kudzu.) Of those surveyed, 440 said "yes." Find a 95% confidence interval for the percentage of all Florida voters supporting the bill. May we conclude (with 95% confidence) that a majority of voters favor passage of this legislation? Why/why not?

SOLUTIONS:
1) .25 + .021
2) .55 + .034; yes