Welcome to COMAP
 
Search

COMAP
Enter query
  

Order Form


All five dvds with both guides
DVD
Product # 8600
$350.00




View Cart / Checkout

 

 

Decisions Through Data Home

   

Statistics: Decisions Through Data

UNIT 17: Census and Sampling

Summary of the Video

A census is an attempt to gather information about every member of some group (called the population ), while sampling gathers information only about a part (the sample ) to represent the whole. The U.S. Constitution requires a census of the U.S population every ten years. We look at the Census Bureau's attempts to encourage participation and count even those who are reluctant. Barbara Bailar, former head of statistical research at the Census Bureau, and New York Congressman Garcia discuss the politically controversial census undercount. Some experts favor adjusting the census to correct the undercount. The adjustment would be made on the basis of smaller but more intense samples.

Sampling is widely used, as a look at the manufacture of potato chips by Frito-Lay illustrates. A truck carrying raw potatoes arrives at the plant, but is not allowed to unload until a sample of its potatoes has been carefully tested. We see how potato chips are made … and how regular samples taken during production keep quality high. Finished bags are sampled to check their weight, both before and after being packed into cartons. Frito-Lay buyers even purchase sample bags off the store shelves and send them to headquarters for examination.

How should we choose a sample from a large population, say all American adults? The advice columnist Ann Landers asks her readers to mail in answers to questions she asks—that draws a very large sample, but voluntary response like this attracts people who feel strongly. The resulting sample is almost certainly not representative of all adults. Statisticians prefer to let impersonal chance choose the sample in such a way that everyone has the same chance to be included. To do this in practice, use a random digit table.