- Which of the following are sources of sampling error and which are sources of nonsampling error? Explain your answers.
- The subject lies when questioned.
This is a nonsampling error since it is a response error.
- A typing error is made in recording the data.
This is a nonsampling error since it is a processing error.
- Data are gathered by asking people to mail in a coupon printed in a newspaper.
This is a sampling error because this is an example of nonrandom sampling error (convenience sampling).
- On page 43 of the textbook is part of a newspaper report of a Gallup Poll, reprinted from the New York Times of January 1, 1990. At the end of Section 5 are some questions that should be answered by a careful account of a sample survey. Which of these questions does this newspaper report answer, and which not? Give the answers whenever the article contains them.
- Population: This is vague, given only as "Americans" and "adults." That means U. S. residents age 18 and over.
- Sampling Method: This is not stated, though we know that Gallup uses random sampling and the statement of a margin of error confirms this.
- Sample Size: 1234 persons. A margin of error (+/- 4%) is also given.
- How Contacted: This is done "by telephone."
- Timing: "Nov. 16-19" (of 1989).
- Exact Questions: These are not explicitly given, but suggested by the wording such as "said they expected the overall quality of their own life to be better." Because these may be the newswriter's words, we would like to know just how the poll posed its questions.
- You wish to determine whether students at your school think that faculty members are sufficiently available to students outside the classroom. You will select an SRS of 200 students.
- What is the exact population? (Will you include part-time students? Graduate students?)
- How will you obtain a sampling frame?
- How will you contact subjects? (Is door-to-door interviewing allowed in campus residence halls?)
- What specific question or questions will you ask?
This is a discussion question, answers to which depend on local circumstances. In (d), students often forget to include demographic questions such age, sex, major, etc.
- Comment on each of the following as a potential sample survey question. If any are unclear or slanted, restate the question in better words.
- Does your family use food stamps?
It used to be thought that food stamps were the bonus green stamps given out by many grocery stores. Now that government food stamps are better known, the vagueness is less. However, a better wording is "The Department of Agriculture provides food stamps to enable low income families to buy food at a lower cost. Does your family use these food stamps?"
- Which of these best represents your opinion on gun control?
- The government should confiscate our guns.
- We have the right to keep and bear arms.
The National Rifle Association is forever sending out surveys with questions like this. It contrasts an extreme measure with a vague constitutional statement. A good question should concern one or more specific proposals on gun control, such as "Which of these gun control measures do you support?"
- In view of escalating environmental degradation and predictions of serious resource depletion, would you favor economic incentives for recycling of resource-intensive consumer goods?
This is unreadable, but also vague. A better wording is "Would you be willing to pay more for the products you buy if the extra cost were used to conserve resources by encouraging recycling?" That's still a bit vague, but less so, and it is written in English.
Please direct all questions regarding STAT-110 to your instructor or to the director of STAT-110, Dr. Todd Ogden at
ogden@stat.sc.edu.
© 1996 by W. Scott Street, IV
