TC3 → Stan Brown → Statistics → Fall08 ME50 → Chapter 1 Quiz
revised Sep 16, 2008

Quiz with Solutions: Chapter 1 (12 min)

These solutions show about the same level of work I expect from you, though I add quite a bit of extra commentary. Please see Show Your Work for the what, why, and how.

1(points: 3)  According to a news story in USA Today this summer, Governor Paterson planned to conduct an Internet poll to determine New Yorkers’ attitudes toward same-sex marriage.
(a) Which vocabulary word best describes the type of sample that will be obtained?
(b) Assuming he has a large number of responses, is this a good procedure or a bad procedure? Explain briefly.

Answer: (a) convenience sample
(b) It is a bad procedure because it’s self selected. Only a random sample, or a modified random sample such as stratified or cluster or systematic, can form a base for correct inferences. Any sample where people select themselves is automatically suspect.

2(points: 3)  “67% of Americans are worried about their financial future.”
(a) Is this an example of descriptive or inferential statistics?
(b) In a few words, state the variable.
(c) Give the data type.

Answer:  (a) inferential (Not every American was asked — researchers took a sample and then estimated the figure for the population.)
(b) worries about financial future
(c) binomial (also acceptable: “qualitative” or “attribute”, but it’s better to be specific)

Common mistake: Don’t be fooled by the “67”. The data point for each person is a yes or no.

3(points: 3) An observational study can not establish that a particular factor causes a particular effect. What is the most important reason for that limitation?

Answer: In an observational study, there can always be lurking variables, so you never know whether a difference is due to the factor you’re interested in or to some other factor.

Alternative solution: “No control group” was also acceptable because the implication is that subjects are assigned to control and treatment groups in such a way as to eliminate or minimize lurking variables.

4(points: 3) If a particular factor causes a particular effect, a properly designed experiment can show this. Once you have your random sample, name two alternative techniques for assigning subjects to the control or treatment groups to ensure that any observed effect is due to the factor you are studying.

Answer: randomization (randomly assigning the subjects into groups) and matched-pair design.

Common mistake: “Random sample” and “control group” are good techniques, but were already given in the question. “Double blind” is a good way to administer dosages, but it has nothing to do with how you assign subjects to treatment and control groups.

5(points: 3) Does sampling error result from a mistake by the persons conducting the study or experiment? What is the main source of sampling error?

Answer: No “Error” in this case does not mean a mistake. Sampling error is unavoidable because it results from the fact that a sample is being used to represent a population, and one sample is not the same as the next.

This page is used in instruction at Tompkins Cortland Community College in Dryden, New York; it’s not an official statement of the College. Please visit www.tc3.edu/instruct/sbrown/ to report errors or ask to copy it.

For updates and new info, go to http://www.tc3.edu/instruct/sbrown/stat5008c/