120 SAMPLING FROM BINOMIAL POPULATIONS Ch. 5 



PROBLEMS 



1. Suppose that you were sent out to ascertain the public opinion in a cer- 

 tain community regarding the necessity for flood control of a certain type in 

 that area. How would you obtain your sample so that it would be representa- 

 tive of the whole community? 



2. Referring to problem 1, would it make any difference in the manner in 

 which you took your sample if it were taken in July of 1936 during a severe 

 drouth or in July of 1951 right after a record flood? Justify your answer. 



3. Suppose that a roving reporter goes into a city with the intention of 

 ascertaining public opinion on a matter of foreign policy. He is going to walk 

 about the streets asking persons at random a specific question requiring one 

 of three answers: Yes, No, or No Opinion. Will it make any difference what 

 hours of the day, between 7 a.m. and 6 p.m., he does this? Would the day of 

 the week matter? Would the type of city — industrial, college site, farming 

 community, rich suburb, and the like — have anything to do with the answers 

 to these questions? 



4. Suppose that a company which is manufacturing candies develops a new 

 product whose originators believe is especially good. Which of the following 

 possible ways of testing the public's reaction to this new confection would you 

 prefer to use? Why? 



(a) Sit back and see how the sales go. 



(b) Have some trained persons take samples out to the public and ask peo- 

 ple to taste the candy, to record their reactions, and to give these records to 

 the field representatives directly. 



(c) In each of the first 10,000 packages manufactured, place a stamped and 

 addressed card requesting that the purchaser record his opinion of the candy 

 and mail the card to the company. 



(d) Ask a panel of expert candy tasters to decide the matter. 



(e) Do as in d, first, then a. 

 (/) Do as in d, first, then b. 

 (g) Do as in d, first, then c. 



(h) Have all the firm's employees record their opinions of the candy and 

 decide from these records if mass production is wise. 

 (i) Combine h and a. 

 (;') Combine h and b. 

 (k) Combine h and c. 

 (I) Combine others above. Specify. 

 (to) Specify another method if you have one you prefer. 



5. Suppose that some engineering concern wishes to test the strength of two 

 types of structural beams, each produced and recommended by a different com- 

 pany. Which of the following sampling procedures would you recommend if 

 the engineering group has two laboratories, each with its operating personnel, 

 available for the tests? 



(a) Ask each company to send a specified number of beams for testing and 

 have each laboratory test half of each company's product. 



