Sample Surveys

5. Designing a Sample

Exercise 2

Read the statement below and select the appropriate response to the question asked. Once you answer this question, you will be asked a series of related questions.

Statement:
Suppose you need to conduct an inventory survey for a large company that has its products stored in 5 warehouses. The company generally puts products of similar sizes into warehouses together. In this case, would stratification be appropriate?

Select YES or NO

Answer:
Yes. By taking separate samples in the 5 warehouses the survey would deliberately include a range of products. The overall estimate of the inventory characteristics should be much better than if a simple random sample had been used.

Select the “Continue” button to proceed to the next part of the exercise.

Question 1 out 3: Which sampling goals can stratification address?

a) Writing a question about household income
b) Obtaining a minimum sample size for a subpopulation
c) Changing sample sizes after the survey has started
d) Increasing the spread of the sample across the range of characteristics in the population
e) Reducing travel costs
f) Reducing the margin of error for estimates

The user can select the ‘How did I do’ and then ‘Restart Question 1’ or Show correct answers’ button.

Answers:

a) Writing a question about household income NO
b) Obtaining a minimum sample size for a subpopulation YES
c) Changing sample sizes after the survey has started NO
d) Increasing the spread of the sample across the range of characteristics in the population YES
e) Reducing travel costs NO
f) Reducing the margin of error for estimates YES

Question 2 of 3: Which sampling design tends to be associated with higher standard errors relative to simple random sampling?

a) Systematic sampling
b) Cluster sampling
c) Stratified random sampling
d) Sampling with probability proportion to size

Answers:

a) Systematic sampling NO
b) Cluster sampling YES
c) Stratified random sampling NO
d) Sampling with probability proportion to size NO

The user can select the ‘How did I do’ and then ‘Restart Question 2’ or Show correct answers’ button.

Question 3 of 3: Which sampling design tends to be associated with lower standard errors relative to simple random sampling?

a) Systematic sampling
b) Cluster sampling
c) Stratified random sampling
d) Sampling with probability proportion to size

The user can select the ‘Restart Question 3’ or ‘Show Correct Answers’ and the ‘Finish’ button. Once the user has finished the exercise they can select the ‘Restart Exercise’ button.

Answers:

a) Systematic sampling YES
b) Cluster sampling NO
c) Stratified random sampling YES
d) Sampling with probability proportion to size YES

Good work! You have completed this exercise!