Chapter 9: Primary Data Collection: Experimentation and
Test Markets
Multiple Choice
1. In an experiment, which of the following variables would probably not be the
dependent variable?
a. total sales
b. market share
c. advertising
d. percent increase in market penetration
e. All of the above would probably be dependent variables.
2. The “treatment” is also referred to as which of the following?
a. increase in sales
b. change in market share
c. gross margin
d. independent variable
e. none of the above
3. Which of the following types of research allows the researcher to show
causation?
a. survey
b. observation
c. experiment
d. none of the above
103
4. Which of the following is not a precondition for causation?
a. the effect variable
b. correlation
c. appropriate time order of occurrence
d. elimination of other possible causal factors
e. All of the above are preconditions for causation.
5. Research that is designed to determine whether a change in one variable likely
caused change in another variable is referred to as _______________.
a. exploratory research
b. causal research
c. descriptive research
d. none of the above
6. Concomitant variation occurs when 2 variables are _______.
a. causally related
b. experimentally related
c. correlated
d. none of the above
7. In order to show that variable A causes variable B, A must occur:
a. after B.
b. before B.
104
c. simultaneously with B.
d. none of the above
8. Suppose a manufacturer wants to determine whether or not changing the
packaging of his/her major product will bring about a change in sales. Which
type of research will be most appropriate?
a. causal research
b. exploratory research
c. descriptive research
d. qualitative research
e. none of the above
9. To demonstrate that an increase in advertising causes a sales increase, one must
be able to show that the increase in advertising occurred before the increase in
sales. Which of the preconditions for causation is being illustrated?
a. correlation
b. appropriate time order of occurrence
c. elimination of all possible causal factors
d. elimination the effects of the independent variable
e. none of the above
10. An analysis of company financial records shows that in every year that the
company increased its advertising expenditures, sales increased. What can you
say about the relationship between the two factors/variables?
a. Advertising expenditure and sales exhibited concomitant variation.
b. Increasing advertising expenditures led to increased sales.
c. Increased sales led to an increase in advertising expenditures.
d. Advertising and sales are inversely related.
e. none of the above
11. In an experiment where the marketer is interested in finding out the impact of
shelf position on a product’s sales, the shelf position is considered the
__________variable.
a. correlation
b. dependent
c. independent
d. concomitant
e. none of the above
12. Experiments are especially effective in showing causality because they are able to
do what?
a. show correlation
b. demonstrate concomitant variation
c. demonstrate an appropriate time order of occurrence
d. eliminate other possible causal factors
e. none of the above
13. Which of the following is the main disadvantage of laboratory experiments?
a. Exploratory research is enhanced.
b. The laboratory setting may not be a good representation of the real-world
setting.
c. Concomitant variation is not possible.
d. Elimination of causal factors is increased.
e. none of the above
14. Test markets are which of the following type of experiment?
a. controlled experiment
b. exploratory experiment
c. field experiment
d. laboratory experiment
e. none of the above
15. If the researcher can show that the experimental or treatment variable actually
produced the differences observed in the dependent variable, then the experiment
can be said to be which of the following?
a. internally valid
b. extraneously valid
c. externally valid
d. laboratory valid
e. none of the above
16. An experiment is designed to test how consumers go about choosing life
insurance policies. If this experiment uses a sample of college students, it would
likely not have _______.
a. internal validity
b. extraneous validity
c. external validity
d. laboratory validity
17. Which of the following has the highest degree of external validity?
a. field experiment
b. laboratory experiment
c. concomitant experiment
d. none of the above
18. __________ refers to the intervention between the beginning and end of an
experiment, involving outside variables or events that might change the dependent
variable.
a. Maturation
b. History
c. Selection bias
d. Mortality
e. none of the above
19. __________ refers to changes in subjects throughout the course of the experiment
that are a function of time, such as getting older, tiring, etc.
a. Maturation
b. History
c. Selection bias
d. Mortality
f. none of the above
20. When respondents drop out of an experiment before it is completed, this is
referred to as ________________.
a. maturation
b. history
c. instrument variation
d. mortality
e. selection bias
21. Suppose during an experiment, a researcher used several observers to record the
results. The results of the experiment were perplexing to the researcher. In such
a case one might expect ___________ to have occurred, distorting the results of
the experiment.
a. maturation
b. history
c. instrument variation
d. mortality
e. selection bias
109
22. If extraneous factors in an experiment can be identified and measured,
__________ can be used to control these factors.
a. randomization
b. physical control
c. design control
d. statistical control
e. none of the above
23. When measures of recall and brand attitudes are only taken after an advertising
campaign is implemented, this is an example of a:
a. one-shot case study design.
b. one group pretest and posttest design.
c. static group comparison design.
d. Solomon four-group design.
e. none of the above
24. This true experimental design involves random assignment of subjects to
experimental and control groups, but no premeasurement of the dependent
variable.
a. Solomon four-group design
b. before and after with control group
c. after only with control group
d. one group pretest and posttest design
e. none of the above
25. When designing a test market, the marketer is often interested in finding out
whether a new product’s sales will take away sales from existing products. What
is this referred to as?
a. controlled test marketing
b. electronic test marketing
c. cannibalization
d. interrupted time series
e. none of the above
26. Which of the following is not a benefit of test marketing?
a. estimate of market potential
b. exposure of the new product to the competition
c. identification of weak areas of the marketing mix
d. provide management with strategic information about product
performance
e. All of the above are benefits of test marketing.
27. Which of the following is often the most important data produced by a test
market?
a. purchase data
b. awareness data
c. competitive response
d. source of sales
e. none of the above
28. A _____________ is a by-product of the research process itself.
a. Testing Effect
b. Research Effect
c. Result
d. Decision
29. The three types of control discussed when dealing with extraneous variables are:
a. Physical, Mental, Statistical
b. Physical, Design, Statistical
c. Design, Mental, Statistical
d. Physical, Emotional, Design
30. _____________ occur(s) when buyers from outside the test area come into the
area to purchase the product being tested, thereby distorting the results of the
experiment.
a. Most accurate results
b. Bias
c. Contamination
d. None of the above
True/False
31. Experimental research is often referred to as causal research.
32. Both surveys and experiments can show causation between two variables.
33. Two variables must be correlated to be causally related.
34. Laboratory experiments are usually conducted in uncontrolled settings.
35. Because laboratory experiments can control extraneous effects, the researcher’s
ability to infer that the independent variable is causing change in the dependent
variable is enhanced.
113
Learning Objective 9.3: To learn about the experimental setting.
36. Field experiments in marketing are conducted in the actual marketplace.
37. The extent to which the results of an experiment can be generalized to represent
what would happen in a population is referred to as internal validity.
38. Selection bias generally is minimized by randomization of subjects to test groups.
39. Laboratory experiments tend to produce higher internal validity than field
experiments.
40. The results of a field experiment are often impacted by competitors’ actions.
114
Ref: Limitations of Experimental Research
Learning Objective 9.5: To learn the limitations of experimentation in marketing
research.
41. Sales are a common dependent variable in marketing experiments.
42. A test market generally occurs after a national rollout of a new product.
43. The extent to which a new product takes away from the innovative company’s
present business is referred to as cannibalization.
44. Full commercialization of a new product is generally less expensive than
conducting a test market.
45. The costs of producing a product for a test market are comparable to the cost
associated with a national launch of the new product.
115
Difficulty: Medium
Response: See page 261
Ref: Test Markets
Learning Objective 9.7: To gain insight into test marketing.
46. Pre-experimental designs offer little or no control over extraneous factors.
47. Quasi-experiments are studies in which the researcher lacks complete control over
the scheduling of treatments or must assign respondents to treatments in a
nonrandom manner.
48. Test markets offer at least two important benefits to the firm conducting the test.
First and foremost, the test should identify weaknesses of the product and the
proposed marketing strategy for the product and give management an opportunity
to correct any weaknesses.
49. The first step in a test market study is to select a basic approach.
Essay Questions
50. Under what conditions can the results of a test market be used to forecast sales for
a region?
51. Your client is interested in determining which of two Internet banner ads is more
effective. Design a field experiment. Be sure to identify the independent and
dependent variables.
52. How would you respond to the following statement: “Advertising and sales are
almost perfectly correlated, as when more is spent on advertising, sales go up, and
vice-versa. Surely our advertising expenditures are ‘causing’ sales to rise.”
53. Marketing Research firms often have exclusive rights to doing interviews in
shopping malls they have contracts with. They also often maintain facilities
within the mall. Such firms might set up an artificial store within the mall to test
advertising or sales promotion associated with such a store. Discuss the type of
experiment being conducted if the marketing research firm offers an incentive to
persons in the mall, such as those participating in mall intercept interviews, who
would be observed as they interact with the promotional materials in artificial
store.
54. Test marketing sounds like a natural step in the progression of a product through
the product development process. However, there are some risks associated with
conducting test markets. Hence, what are the 4 factors that should be taken into
account in determining whether or not to conduct a test market?