McDaniel & Gates – Marketing Research, 9th Edition Instructor’s Manual
B. Examples of Quasi-Experiments
1. Interrupted Time-Series Designs–repeated measurement of an effect both before and after a
treatment is introduced that “interrupts” previous data patterns
O1 O2 O3 O4 X O5 O6 O7 O8
a. Primary Weakness–inability to control history and possibility of interactive effects of testing
2. Multiple Time-Series Designs–interrupted time-series design with a control group–shown
symbolically as follows:
Experimental Group: O1 O2 O3 O4 X O5 O6 O7 O8
Control Group O1 O2 O3 O4 O5 O6 O7 O8
See Global Research: What Happens When Your Selection Bias Is the Interviewer’s
Gender? (pg. 255)
Recently, German researcher Stefan Althoff found that surveys sent by women or with women’s
names get better responses than men’s.
• In 2004, Lufthansa Technik in Hamburg, Germany commenced an online customer survey and
observed a very low response. A week later they sent out reminders, but this time the writer
• In March, 2006, the researchers surveyed 460 registered users of UNIpark and split them into
four groups and asked them to evaluate UNIpark’s Web site homepage and the utility of
Discussion Questions:
1. What types of possible (even if subtle) bias might result from male respondents answering an
e-mail survey sent or signed by a female?
2. Althoff planned to test the impact of different women’s names on response rate. Consider
which three female names might be the most attractive and least attractive to male respondents.
10. TEST MARKETS
I. Test Market
A. Test Market Defined–loosely refers to any research that involves testing a new product or
change in an existing marketing strategy (e.g., product, price, place, promotion) in a single
B. Designed to Provide Information
1. Provides information in regard to the following issues:
a. Estimates of market share and volume–projected to the total market
b. Cannibalization rate–effects that the new product will have on sales of similar products
already marketed by the company
c. Characteristics of consumers who buy the product–the psychographic and lifestyle
characteristics of target customers will provide valuable insights into how to position the product
and the types of promotional messages that will appeal to them
d. Behavior of competitors–during the test
See Global Research: What Makes a Good Test Market? How about Calgary? (pg. 260)
At least five major companies have test-marketed products in Calgary in recent years: Labatt
Discussion Questions:
2. Develop a demographics and psychological profile for the typical person, age 35, who is a
member of this new “hotly pursued age group.”
II. Types of Test Markets
A. Four Types of Test Markets
1. Traditional or standard test market–testing the product and other elements of the marketing
mix through a firm’s regular channels of distribution
2. Scanner or electron test markets–scanner panel research firms have panels of consumers
3. Controlled test markets–research suppliers who ensure that the product is distributed through
the agreed upon types and numbers of distributors
4. Simulated Test Markets–alternative to traditional test markets a. Number of different
1) Sample of consumers is selected based on the expected or known characteristics of the target
consumer for the product
2) Those sampled are recruited to come to a central location testing facility to view commercials
5) This information is used with the proprietary model of the company to generate estimates of
sales
volume, market share, and other key market metrics
III. Costs of Test Marketing
A. Direct Costs
1. Production of commercials
3. Media time, charged at a higher rate because of low volume
5. Customized research information and associated data analysis
7. Coupons and sampling
8. Higher trade allowances to obtain distribution
B. Indirect Costs
1. Cost of management time spent on the test market
3. Possible negative impact of a test market failure on other products with the same family brand
5. Cost of letting competitors know what the firm is doing, thereby allowing them to develop a
better strategy or beat the firm to the national market
IV. Decision to Conduct Test Marketing
A. Benefits
1. The test market provides a vehicle by which the firm can obtain a good estimate of a product’s
sales potential under realistic market conditions
2. 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
B. Major Factors to Consider Before Conducting a Test Market
2. Consider the likelihood and speed with which competitors can copy your product and
introduce it on a national basis
4. Consider how much damage an unsuccessful new product launch would inflict on the firm’s
reputation
V. Steps in a Test Market Study
A. Steps that Must Be Conducted
1. Step One: Define the Objective–test market objectives include the following:
McDaniel & Gates – Marketing Research, 9th Edition Instructor’s Manual
2. Step Two: Select a Basic Approach–the appropriate type of test market method should be
selected. There are four basic approaches
a. Standard, or traditional, test market
b. Scanner or electronic test market
c. Controlled test market
d. Stimulated test market
3. Step Three: Develop Detailed Test Procedures–manufacturing and distribution decisions
must be made to ensure that adequate product is available and that it is available in most stores of
the type that sell that particular product class.
See Global Research: Best Global Test Markets? Danes Are Quick, Koreans Love Tech
(pg. 264-265)
It takes only 3.3 years from product launch to widespread sales and market entrenchment in
Denmark, compared to Germany, France, or England, where it might take two to three times
longer, or up to 9.3 years in Portugal. South Korea, the most wired nation in Asia, is the test
market for high tech. Two-thirds of the population’s 16 million households have broadband
Internet service, and 68 percent already have advanced cell phones capable of high-speed data
transmission.
Discussion Questions:
1. Identify and discuss other examples of culture-specific references that are taken for granted in
the United States but that would seem strange or incomprehensible in other countries.
2. Identify three cutting-edge high-tech products (e.g., in the category of new products like
4. Step Four: Select Test Markets–a number of factors must be taken into account in making
this decision. First there are the overall standards:
a. There should be a minimum of two test markets, in addition to a control market, for an
existing national brand, or a minimum of three markets for testing a new brand
5. Step Five: Execute the plan–researcher can now begin execution–but a key decision has to
be made–how long should the test run? The average test runs for 6 to 12 months.
6. Step Six: Analyze the Test Results–the data produced by an experiment should be evaluated
throughout the test period. After completion of the experiment, a more careful and thorough
evaluation of the data must be performed. This analysis will focus on four areas:
a. Purchase data
b. Awareness data
c. Competitive response
McDaniel & Gates – Marketing Research, 9th Edition Instructor’s Manual
d. Source of sales
VI. Other Types of Product Tests
A. Rolling Rollout–product is launched in a certain region rather than in one or two cities
B. Lead Country Strategy–try a product out in a foreign market before rolling it out globally
See Practicing Marketing Research: Quick Scans of Product-Specific Test Markets (pg.
267)
• Miller Brewing Company of Milwaukee, Wisconsin, test marketed three kinds of low-calorie,
craft-style light beers in only four cities starting in 2008, including Minneapolis, Minnesota;
Charlotte, North Carolina; San Diego, California; and Baltimore, Maryland.
Discussion Questions:
1. What characteristics do the cities of Minneapolis, Charlotte, San Diego, and Baltimore possess
that make them good markets for introducing premium-level beers?
2. How would you tailor a test market experimentation for Brazilian cuisine in a Mormon
saturated region whose members nonetheless are well-traveled and versed in Brazilian foods?
11. SUMMARY
QUESTIONS FOR REVIEW AND CRITICAL THINKING
1. Divide the class into as many as six groups, as appropriate. Each group will have the task
of recommending a test market design and addressing the associated questions for one of
the following scenarios.
• Design a test of a new pricing strategy for orange juice concentrate. The brand is an
established brand, and we are only interested in testing the effect of a 5 percent price
increase and a 5 percent price decrease. All other elements of the marketing mix will
remain the same.
Randomly select three experimental groups. Expose Group 1 to the product at the 5%
price increase, expose Group 2 to the 5% price decrease, and expose Group 3 (the control
group) to the original price. Use a questionnaire to measure post-test attitudes about
product pricing and purchase intentions.
• A national value-priced hotel chain needs to understand the business impact of including
a free buffet style breakfast to guests. Design and justify a test that will do this. Do some
preliminary investigation to determine what items the target market of guests would
2. Tico Taco, a national chain of Mexican fast-food restaurants, has developed the “Super
Sonic Taco,” which is the largest taco in the market and sells for $1.19. Tico Taco has
identified its target customers for this new product as men under 30 who are not concerned
about health issues, such as fat content or calories. It wants to test the product in at least
four regional markets before making a decision to introduce it nationally. Which cities
would you recommend using? Why would you
recommend these cities?
Test cities should be chosen to represent the targeted market. This would mean finding cities that
have a large population of the non-health-conscious men under age 30. Also, these cities should
be dispersed geographically to try to control for regional cultural differences. One important
3. Of the primary data collection techniques available to the researcher (survey,
observation, experimentation), why is the experiment the only one that can provide
conclusive evidence of causal relationships? Of the various types of experiments, which
type or type provides the best evidence of causation or non-causation?
4. What are some of the important independent variables that must be dealt with in an
experiment to test consumer reaction to a pilot for a new TV series? Explain why these
variables are important.
What cities will be used for the test? How will the material be presented; in a studio, in peoples’
homes, at a gala grand opening? Who are we going to select as the target market? What sort of
5. Managers of the student center at your university or college are considering three
alternative brands of frozen pizza to be offered on the menu. They only want to offer one of
the three and want to find out which one students prefer. Design an experiment to
determine which brand of pizza the students prefer.
The student center could use several types of experimental design. One appropriate method
would be a pretest-posttest with control group. Students could be randomly assigned to the
6. Night students at the university or college are much older than day students. Introduce
an explicit control for day versus night students in the preceding experiment.
The randomized block design could be incorporated into this design so that students are blocked
7. Why are quasi-experiments much more popular in marketing research than true
experiments?
McDaniel & Gates – Marketing Research, 9th Edition Instructor’s Manual
Quasi-experimental designs generally are more feasible in field settings than are true
8. How does history differ from maturation? What specific actions might you take to deal
with each in an experiment?
History considers special events or variables outside the experiment that may take place between
9. The manufacturer of microwave ovens has designed an improved model that will reduce
energy costs and cook food evenly throughout. However, this new model will increase the
product’s price by 30 percent because of extra components and engineering design
changes. The company wants to determine what effect the new model will have on sales of
its microwave ovens. Propose an appropriate experimental design that can provide the
information for management. Why did you select this design?
A two factor factorial design, or a number of other designs, might be used. The two factor
factorial design might be used with two price levels – current and 30 percent increase – and two
packages of features – current features and new features. In the experiment, randomly selected
subjects would be randomly
10. Discuss various methods by which extraneous causal factors can be controlled.
Extraneous causal factors can be controlled by randomization of testing units, multiple measures
11. Discuss the alternatives to traditional test marketing. Explain their advantages and
disadvantages.
Simulated Test Markets (STMs) are an alternative to traditional test marketing. The advantages
include a low probability of your competitors knowing about your testing, the speed at which
tests can be performed, relatively low expense, and accuracy of the models. Another alternative
WORKING THE NET
1. Visit www.questionpro.com/akira/showLibrary.do?categoryID=16&mode=1 to take a
psychographics profile with 269 focused questions distributed into 15 categories. What
kind of consumer are you? How would you design a test market to attract a customer like
yourself?
2. Consult www.city-data.com Research city demographics for Cedar Rapids, Iowa; Eau
Claire, Wisconsin; and Grand Junction, Colorado, to evaluate why Advertising Age (2005)
ranked them among the top seven most popular test market sites in the United States for
matching the average American demographic profile.
Data from the website indicates that these cities consistently mirror the demographic profile of
McDaniel & Gates – Marketing Research, 9th Edition Instructor’s Manual
REAL-LIFE RESEARCH
Case 9.1 Texas Red Soft Drinks (pg. 275)
Key Points:
• Texas Red is the oldest soft-drink company in Texas. During the 1980s, the company
went through bankruptcy because of its inability to compete with the big national soft
drink brands. In 2003, a group of investors purchased the brand name from the
bankruptcy court and reincarnated it with distribution primarily through specialty grocery
stores and selected theme restaurants. The company is currently available in 267 grocery
Questions
1. Is the test market approach the best way to tackle this problem, all things considered?
Are there other viable options and, if so, what are the options?
2. Which of the experimental designs discussed in the chapter would be most feasible for
this project? Why that design?
3. How many markets should be used for the test? What should be the characteristics of
these markets?
4. What sort of evidence, from the test, would definitively break the deadlock between the
two groups with different visions as the company?