Unit 7: Sampling Distributions and Estimations
"Four out of five dentists surveyed recommend sugarless gum for their patients who chew gum.”
This is a statement that has been part of a television commercial for many years. Reread the statement. Does this mean that they asked just 5 dentists, and 4 of them recommended sugarless gum? Or did they ask 50 dentists, and 40 of them recommended sugarless gum? Or did they ask 50,000 dentists? Does it matter? Would you trust the statement more if they had asked 50,000 dentists instead of 5?
In Unit 7, you will learn that it certainly does make a difference. You will find out that if you have a representative sample, a large sample size will yield more trustworthy results. This is the realm of sampling distributions and confidence intervals. These unwieldy sounding concepts provide basic tools for statisticians to use when they decide how big a sample they need and how much "error” they can be comfortable with.
Political pollsters use margins of error when they report approval ratings for politicians (e.g., 42% approval rate, ± 3%) to give a complete picture of opinion data that they have collected. You will learn why these are important and how to calculate them yourself.
Completing this unit should take you approximately 10 hours.
- Receive a grade