# Statistics

Jeanine Stewart
QNT/561-Dr. J. Rudin-Instructor
Homework Assignment-Individual Week 2

Chp 8.   Exercise 21

What is a sampling error? Could the value of sampling error be zero? If it were zero what would this mean ?

The sampling error is the difference between a population parameter and a sample statistic.
The value of the sampling error could be zero but this would mean that we would have an exact representation of the population and that we knew all the characteristics of the entire population.

Exercise 22
List the reasons for sampling?   Give an example for each   reason for sampling?
Per the information in the chapter reading reasons for sampling involve to contact the whole population is time consuming (ex. A candidate running for national office- use of a poll and a few staff members are more efficient and can be done in 1 to 2 days versus years and lots of manpower to get this done.)   Another reason is the cost of doing such a study of all the items in a population may be expensive. (Ex. Companies can test cereal, perfume cat food etc using \$2,000 people in the population with a cost of approximately \$40,000 versus studying 60 million people at a cost of about \$1 billion).   It is also physically impossible to check all the items in a population. (Ex.   Attempting to check all the water in Lake Erie for bacterial levels is not physically capable but select water from a specific section of the lake.)   The destructive nature of some tests. (If wine tasters drink all the wine in the winery to determine their vintage they would consume the entire crop and have nothing to sale).   Finally, the sample results are adequate. (If the U.S.   Government tested all the grocery stores in the population there is a very small chance that it would affect the price index versus using a small sample of stores).

Exercise 34
1. Information from the American Institute of Insurance indicates the mean amount of life insurance per household in the United States...