Chapter 13 Implementing Basic Differences Tests
1) Iams marketing over 20 different types of dog food, Toyota marketing over 20 different
types of cars, and Boeing having five different types of commercial jets and a separate
business jets division are all examples of:
A) differing markets strategy.
B) marketing variation.
C) market segmentation.
D) market separation.
E) contrasting markets.
Answer: C
2) In your text, you learn to test for significant differences between:
A) two percentages or two means.
B) two percentages or three percentages.
C) two means or three percentages.
D) three or more percentages.
E) differences between a percentage and a mean.
Answer: A
3) A winery sees the market in three distinct segments based upon usage of wine; a dog food
manufacturer sees the dog food market in 12 segments; Toyota offers 17 different models
ranging from a two-seater sports car to the Avalon luxury car. These are all examples
illustrating how:
A) marketers can take advantage of differences between market segments, simply by a
conceptual analysis of differences in the marketplace.
B) marketers should search for all differences between each of their products or services.
C) marketers could take better advantage of differences between market segments if they had
a method of determining insignificant differences in the marketplace.
D) marketers can take advantages of differences between market segments, and differences
analysis can be used to discovery statistically significant, meaningful, and stable differences.
E) there are differences in the market, and by knowing statistical methods that focus on data
summarization, marketers can take advantage of these differences.
Answer: D
4) In order to be potentially useful for a marketing manager, differences must at minimum be:
A) at a ratio of 2 to 1.
B) statistically significant.
C) new and not part of existing knowledge.
D) at a ratio of 10 to 1.
E) None of the above; any difference should be important to a manager.
Answer: B
5) Which best represents the correct meaning of statistical significance of differences?
A) The differences found in sample data may be found in other sample data.
B) The differences found in sample data may be assumed to exist in the remainder of the
sample.
C) The differences found in the sample may be assumed to exist in the population.
D) The differences found in the population may be assumed to exist in the sample.
E) A significant difference means that managers will be able to use the difference in order to
develop effective marketing strategies.
Answer: C
6) One of the reasons that all statistically significant differences may not be meaningful is
that, to a great extent, statistical significance is determined by:
A) competitors who are trying to sabotage research studies.
B) only a few customers.
C) statistical formulas that are unstable and unpredictable themselves.
D) sample size.
E) data mining software that is too new to be trustworthy.
Answer: D
7) A meaningful difference is one that the marketing manager:
A) can potentially use as a basis for marketing decisions.
B) deems to be statistically significant.
C) finds to be at a ratio of 2 to 1.
D) finds to be stable over at least a week or so.
E) may or may not be statistically significant.
Answer: A
8) The fact that different segments of flu or cold sufferers will consistently seek different
medications targeted at their symptoms (sore throat, nasal congestion, cough, etc.) illustrates
the use of:
A) differences that are significant.
B) differences that are meaningful.
C) differences that are stable.
D) differences that are actionable.
E) differences that are found in the sample.
Answer: C
9) To be useful to the marketing researcher or manager, differences must be:
A) meaningful.
B) stable.
C) actionable.
D) statistically significant.
E) All of the above are needed.
Answer: E
10) Which of the following requirements of differences means that the marketer can focus
various marketing strategies and tactics, such as product design or advertising, on the market
segments to accentuate the differences between the segments?
A) actionable
B) meaningful
C) stable
D) statistically significant
E) different
Answer: A
11) The t value is used for many tests instead of the z value because:
A) it is easier to calculate and to interpret.
B) it is more widely known among statisticians.
C) assumptions of the z value are violated if sample size is 30 or less.
D) it is available on statistical software packages.
E) assumptions of the t value are violated if the sample size is 10 or less.
Answer: C
12) Fortunately, when testing for differences, you will not need to be concerned whether to
use the t or z test because:
A) the z is easier to calculate and to interpret and should always be used.
B) the t is more widely known among statisticians and should always be used.
C) assumptions of the z and t are minimal and are of concern only to statisticians.
D) SPSS, and most statistical software packages, are programmed to use the correct statistic.
E) clients will not know the difference.
Answer: D
13) P values are identified by what terms on computer output by statisticians?
A) propositional values and levels of confidence
B) levels of confidence and Pearson coefficients
C) waving flags
D) significance (Sig) and probability (Prob)
E) either .0001 or .0005
Answer: D
14) P values can take on a range of:
A) +1.00 to -1.00.
B) 0 to 100.
C) 95 to 99.
D) 0.000 to 1.00.
E) .0001 to .0005.
Answer: D
15) If we adopt a 95 percent level of confidence, we need a P value to be significant (i.e., flag
is waving) if it is:
A) less than .05.
B) less than or equal to .05.
C) greater than .05.
D) greater than or equal to .05.
E) .90 or greater.
Answer: B
16) When a researcher is determining if two groups are statistically significant, he or she is
considering the two groups as two separate populations. The question is whether or not the
two different populations':
A) z scores are the same.
B) t scores are the same.
C) parameters are different.
D) associations are different.
E) summarization values are the same.
Answer: C
17) When making a comparison between two groups of respondents to determine whether or
not they are statistically different, in concept, the researcher is considering the two groups as
two:
A) different populations.
B) different answers.
C) different tests.
D) common groups.
E) related groups.
Answer: A
18) To test whether a true difference exists between two group percentages, we test the
________ hypothesis.
A) z
B) t
C) z or t hypothesis; it does not matter
D) null
E) alternative
Answer: D
19) Which of the following states that the difference between the population parameters
between two groups is zero?
A) null parameter
B) null hypothesis
C) alternative hypothesis
D) null alternative hypothesis
E) zero hypothesis
Answer: B
20) We have two percentages and we want to know if they are statistically different. We
calculate our z and find that it is 4.21. This means that:
A) the two percentages are the same.
B) the two percentages are not statistically different.
C) the two percentages have a 421 percent chance of not being different.
D) the two percentage are statistically different.
E) the two percentages have a 421 percent chance of being different.
Answer: B
21) When a computed z value (for a test for differences between two percentages), 4.21, is
larger than the standard z value, 1.96, then this amounts to:
A) support for the null hypothesis; the two percentages are different.
B) no support for the null hypothesis; the two percentages are not different.
C) support for the null hypothesis, the two percentages are not different.
D) no support for the null hypothesis; the two percentages are different.
E) None of the above; a z value is inappropriate for testing the differences between two
percentages.
Answer: D
22) Which of the following is NOT true of a differences test?
A) With the null hypothesis, the arithmetic difference between one parameter and another is
0.
B) With both hypotheses, it is impossible to "guess" differences before testing.
C) With the null hypothesis, the statistical test begins with the assumption that the two
percentages (or means) are exactly the same value.
D) The alternative hypothesis is that the two percentages (or means ) are not the same value.
E) With the alternative hypothesis, the difference between parameters is not 0.
Answer: B
23) Which of the following SPSS commands would be used to test for the significant
differences between two percentages?
A) ANALYZE; COMPARE PERCENTAGES; DIFFERENCES
B) ANALYZE; COMPARE PERCENTAGES; DIFFERENCES; TWO
C) ANALYZE; COMPARE PERCENTAGES; PAIRED SAMPLES t TEST
D) DIFFERENCES; TWO SAMPLE; PERCENTAGES
E) None of the above; SPSS does not test for differences between two percentages.
Answer: E
24) If we were comparing the difference between the mean number of sports drinks
consumed by males versus females, and we calculated a z value of 6.43, we would conclude
that:
A) the probability of support of the null hypothesis of no difference is less than .01 because
6.43 is greater than 2.58.
B) the probability of support of the alternative hypothesis of no difference is less than .01
because 6.43 is greater than 2.58.
C) the probability of support of the null hypothesis of no difference is less than .01 because
6.43 is less than 2.58.
D) the probability of support of the alternative hypothesis of no difference is less than .01
because 6.43 is less than 2.58.
E) there is a 6.43 percent chance that the two means are not significantly different.
Answer: A
25) Which of the following commands would we use in SPSS to test for a difference between
two means — one generated from a sample of heavy users of our brand and the other
generated from a sample of nonusers of our brand?
A) ANALYZE; COMPARE MEANS; DIFFERENCES
B) ANALYZE; COMPARE MEANS; DIFFERENCES; TWO
C) ANALYZE; COMPARE MEANS; PAIRED SAMPLES t TEST
D) ANALYZE; COMPARE MEANS; INDEPENDENT-SAMPLES t TEST
E) ANALYZE; COMPARE MEANS; HEAVY v NON
Answer: D
26) The assumption of equal variances between two samples for differences testing, is
evaluated by:
A) the test of variances.
B) Levene's Test for Equality of Variances.
C) Levene's Test for Heterogeneity.
D) The Staniklos-Shockley Test of Equal Common Variance.
E) None of the above; there is no such assumption.
Answer: B
27) If we are examining the SPSS output for testing the differences between two means from
independent samples (Independent Samples t Test) and we show a Sig. of .001 for Equal
variances assumed under Levene's Test for Equality of Variances, this means we:
A) should use the "equal variances assumed" output.
B) should use the "equal variances not assumed" output.
C) should not assume anything; run the test properly.
D) should conclude that the two means are significantly different.
E) should conclude that the two means are not significantly different.
Answer: B
28) Assume you have two questions of the same format, a 5-point scale measuring
importance level. One question measures importance of prices in the selection of a restaurant
and the other measures food quality. If you want to know if the two mean scores to each
question are significantly different, you would use:
A) a z test of the differences between two percentages.
B) a paired samples test for the differences between two means.
C) an independent samples t test.
D) a two mean comparison z test.
E) a two question, mean comparison test.
Answer: B
29) Let's say we wanted to see if there was a significant difference between your class's mean
GPA when you were all freshmen and your class's mean GPA today. What is the proper
SPSS command that you would use to test this difference?
A) STATISTICS; COMPARE MEANS; SAME SAMPLE
B) STATISTICS; COMPARE MEANS; TWO-CLASS SAMPLES
C) STATISTICS; COMPARE MEANS; INDEPENDENT SAMPLES t TEST
D) STATISTICS; COMPARE MEANS; PAIRED SAMPLES t TEST
E) STATISTICS: TWO QUESTION; MEAN COMPARISON TEST
Answer: D
30) Let's assume there are juniors, seniors, and graduate students in your class and we want to
know if their average GPAs differ. What is the proper statistical test?
A) t test
B) means/differences test
C) Levene's test
D) ANOVA
E) Z test
Answer: D
31) ANOVA is called a "signal flag" procedure, meaning that if it shows significance:
A) differences between means are known.
B) there are no significant differences among the means.
C) there is a significant difference between at least one pair of means.
D) there is a significant difference between at most one pair of means.
E) there is a significant difference between all percentages.
Answer: C
32) If we wanted to test for the mean likelihood that shoppers will shop again in different
departments such as home and garden versus sporting goods, versus automotive, versus
electronics, we would use:
A) a z test for differences between percentages.
B) independent samples t test.
C) paired samples t test.
D) ANOVA.
E) departmental means test.
Answer: D
33) Which of the following is NOT part of the advantages that ANOVA has over performing
multiple t tests of the significance of the differences between means?
A) It immediately notifies the researcher if there is any significant difference.
B) All the researcher needs to do is look at the "Sig."
C) It arranges the means so the significant differences can be located.
D) It arranges the means so the significant differences can be interpreted easily.
E) SPSS will only do ANOVA tests, not t-tests of the differences between means.
Answer: E
34) Duncan's multiple range test, Scheffe's test, and Tukey's test are used for:
A) determining the significance of association among groups of related variables.
B) determining the differences between means when ANOVA has produced a significant F
value.
C) determining the differences between means when ANOVA has produced an insignificant
F value.
D) determining significant differences among three or more percentages.
E) None of the above; none of these are statistical tests.
Answer: B
35) Which of the following SPSS commands is used to run ANOVA?
A) ANALYZE; COMPARE MEANS; ANOVA
B) ANALYZE; ANOVA; POST HOC
C) ANALYZE; COMPARE MEANS; ONE-WAY ANOVA
D) ANALYZE; COMPARE MEANS; n-WAY ANOVA
E) ANALYZE; COMPARE MEANS; ANALYSIS OF MEANS
Answer: C
36) Suppose Hershey's chocolate managers wanted to test for differences between usage of
coupons among different combinations of two or more grouping variables, such as age groups
and occupation groups. The appropriate procedure to test for differences among the means of
these groups would be:
A) ANOVA, Multiple Groups Analysis.
B) n-Way ANOVA.
C) GroupANOVA.
D) Percentage-Way ANOVA.
E) None of the above; this cannot be done.
Answer: B
37) There are different statistical tests for differences between means and differences between
percentages.
Answer: True
38) Although there are separate differences tests for means versus percentages, all these test
for differences between two or more values. As long as you have at least two percentages or
two means, any one test will be appropriate.
Answer: False
39) Market segmentation holds that different types of consumers have different requirements.
Answer: True
40) An astute marketer will realize that it is impossible to customize the marketing mix to
each target market's unique situation.
Answer: False
41) When testing for differences, a manager must first determine if the differences are
meaningful and then act on the differences. Only then should he or she decide if the
differences are significantly different.
Answer: False
42) Research determines a significant difference between cold remedy preferences depending
on symptoms to be relieved. When the marketing manager markets several different
remedies, each designed to reduce a different symptom, we can say the differences found in
the research are actionable.
Answer: True
43) Statisticians have found that data do not meet the assumption of being normally
distributed with sample sizes of 300 or more.
Answer: False
44) The t test is the statistical inference test to be used with small sample sizes less than or
equal to 30.
Answer: True
45) SPSS will report analysis of differences tests in terms of t tests.
Answer: True
46) The statistical significance "flag" in SPSS is called the p value by statisticians and is
denoted with "significance" or "probability" or their abbreviations.
Answer: True
47) When we test for the differences between two groups, the null hypothesis is that the
difference in the two groups' population parameters is more than 30.
Answer: False
48) The formula for calculating the difference between two percentages calculates a z value.
Answer: True
49) When testing the difference between two percentages from two independent samples, a z
value larger than 1.96 indicates the difference is NOT significantly different.
Answer: False
50) When testing the difference between two percentages from two independent samples,
SPSS will not make this calculation. You must do it by hand.
Answer: True
51) If you want to determine the mean score on a research question measuring whether
likelihood to purchase a new product differs between respondents in two separate income
groups, you would use the independent samples t test.
Answer: True
52) In the test to determine significance difference between two means generated by
independent samples, there is no provision for testing a directional hypothesis such as "males
drink more soft drinks per week than females."
Answer: False
53) In the independent samples t test, Levene's test determines whether the variances between
the two samples are approximately equal.
Answer: True
54) A group comparison table is often used to summarize significant differences when
reporting group differences to clients.
Answer: True
55) While researchers are expected to report research results accurately and honestly, it is not
necessary to include any results that may seem contradictory or unfavorable.
Answer: False
56) The findings of a research study presented in the textbook comparing purchasing
impulses of French and Swedish supermarket shoppers reveal that marketers who seek to
stimulate impulse shopping in supermarkets must vary their strategies according to the
customs in each country in which they are operating.
Answer: True
57) Different tests are required to test the differences between the means of two samples
depending upon whether or not the two samples are independent or are paired.
Answer: True
58) As long as different, independent samples are being tested, it is appropriate to run the
paired samples test for the differences between two means on one sample's mean height
(measured in feet/inches), versus another sample's mean weight (measured in
pounds/ounces).
Answer: False
59) Unfortunately, SPSS cannot run the paired samples t test.
Answer: False
60) ANOVA is used to determine if there is any correlation among three or more paired
groups or samples.
Answer: False
61) The word "variance" in the name "analysis of variance" is misleading.
Answer: True
62) Analysis of variance is a "signal flag" procedure. By this, we mean that it alerts the
researcher to the fact that there are many differences that are statistically significant and it
flags him or her to those differences which are most important, those which should be acted
upon first.
Answer: False
63) ANOVA's null hypothesis is that no single pair of means is significantly different.
Answer: True
64) If the F value in ANOVA produces a significantly high p value (aka "Sig." in SPSS) of
10 or more, then it is appropriate to proceed with an ad hoc test.
Answer: False
65) The Duncan's Multiple Range test is a post hoc test that allows the researcher to
determine among which pairs of means significant differences exist.
Answer: True
66) In interpreting the output for the Duncan's multiple range test, different means are listed
as different subsets.
Answer: True
67) n-Way ANOVA is a procedure that determines differences between three or more
percentages.
Answer: False
68) Tommy Prothro, a marketing manager for Golden Snack Bars, has commissioned
marketing research to determine if one recipe of snack bar is superior to another recipe. More
than 400 persons who were "snack bar eaters" were involved in taste tests and, after tasting
both recipes, they were asked which recipe they would purchase the next time they purchased
snack bars. Tommy is now looking at the data and he sees that recipe A had 53 percent
stating a preference, whereas recipe B had 47 percent. Tommy's brand manager felt this was
"significant" evidence that the firm should produce recipe A. But Tommy wanted more
evidence so he asked the research firm to run a test to determine if there was a significant
difference between the two recipes. By doing this, Tommy would get information that would
allow him to determine:
A) if there are real differences between the two recipe preferences in the population.
B) if the differences between the recipes are really 6 percent or more.
C) the number of consumers in each target market preferring recipe A versus B.
D) whether or not the statisticians in the research firm agree with his brand manager.
E) None of the above; there is no statistical test to determine significant differences between
two percentages.
Answer: A
69) Tommy Prothro, a marketing manager for Golden Snack Bars, has commissioned
marketing research to determine if one recipe of snack bar is superior to another recipe. More
than 400 persons who were "snack bar eaters" were involved in taste tests and, after tasting
both recipes, they were asked which recipe they would purchase the next time they purchased
snack bars. Tommy is now looking at the data and he sees that recipe A had 53 percent
stating a preference whereas recipe B had 47 percent. Tommy's brand manager felt this was
"significant" evidence that the firm should produce recipe A. But Tommy wanted more
evidence so he asked the research firm to run a test to determine if there was a significant
difference between the two recipes. When the firm ran the test, they reported a z value of
5.64. This means that:
A) Tommy should follow the advice of his brand manager.
B) there is no statistical significance.
C) there is no statistical difference but there is an actionable difference.
D) there is no statistical difference but there is a meaningful difference.
E) there are real differences between the two recipe preferences in the population and Tommy
will have to determine if this difference is meaningful before going further.
Answer: E
70) Mike Shula is a head football coach. His athletic department spends $30,000 a season on
Lizard-Ade, a flavored drink that supposedly contributes to the performance of his players.
This year, an independent sports testing association has decided to test the merits of LizardAde and Shula's university has been selected as a member of the national sample. The study
is an experiment in which the players, unknown to them, are divided into two groups. Group
1 receives the real Lizard-Ade prior to and during the games. Group 2 receives a placebo,
which is nothing more than sugar-flavored colored water in containers made to make the
sugar-flavored water appear to be Lizard-Ade. It is common practice that, following each
game, the coaches evaluate films and give each player a grade ranging from 0 to 100. After
the season the sports testing association collects the data. They now have a mean score of
performance for each of the two groups for all of the athletic departments participating in the
study. If you are the researcher, what statistical test would you conduct?
A) a z test of the difference between two percentages
B) independent samples t test
C) paired samples t test
D) ANOVA
E) None of the above because the sample size will be too large and will give erroneous
results.
Answer: B
71) Rebecca Sims is the general manager of a chain of auto parts stores. There are 400 stores
in the country and they are divided up into seven divisions based upon geography. Each
division has a division manager. The firm prides itself on keeping close tabs on customer
satisfaction, and Rebecca has decided to take a sample of the 400 to test a new surrogate
measure of customer satisfaction — the dollar value of returned merchandise. Each sample
store reports daily the dollar value of merchandise returned and an average is calculated for
each division. When Rebecca sees the first set of results, she wants to know if the differences
between the division means are real or due to sampling error. Rebecca should use which test?
A) a z test of the difference between two percentages
B) independent samples t test
C) paired samples t test
D) ANOVA
E) division differences test
Answer: D
72) Joy Ward is the director of marketing at Helmsley College. She has been studying
marketing research data collected on a national sample of high school seniors who are
planning on attending college. Joy is trying to determine what appeal she should use in a
direct mail campaign that will be targeted at students with high SAT/ACT scores who live
within 500 miles of Helmsley. She is intrigued with the marketing research data that
measures the students' ratings of importance on a number of factors affecting their decision to
choose a particular college. Some of the factors are (1) programs highly valued in the job
market, (2) small campus atmosphere where professors know students' names, (3) ample
opportunities for an active campus life (i.e., football, campus entertainment for students, etc.),
(4) individualized programs designed around students' needs and interests, and so on. There
are eight different factors and each was rated on the same 5-point importance scale ranging
from "Very Important" to "Very Unimportant." Joy knows that Helmsley could ethically use
either the "small campus" appeal or the "individualized program" appeal. The small campus
appeal has a mean score of 4.3 on the importance scale, while the individualized program
appeal has a mean score of 3.95. Because both of these numbers indicate that these are
important factors for students, she wants to know if the small campus appeal is really more
important than the individualized appeal in the total college bound population. Which of the
following tests should Joy run?
A) a z test of the difference between two percentages
B) independent samples t test
C) paired samples t test
D) ANOVA
E) Duncan's Multiple Range Test
Answer: C
Test Bank for Marketing Research
Alvin C. Burns, Ronald F. Bush
9780133074673, 9780134895406, 9780134167404
