Complete Test Bank Ch.16 Kumar Fundamentals of Data Analysis - Marketing Research 13e Complete Test Bank by V. Kumar. DOCX document preview.
Test Bank
CHAPTER 16 Fundamentals of Data Analysis
True-False
1. Data analysis is a powerful aid to gaining useful knowledge and it
can help an investigator rescue even poorly conceived research.
2. If a research study is well designed, then the researcher is assured
of clear judgments and conclusions leading to good decisions, even
if the data analysis techniques aren't entirely appropriate.
3. The basic purpose of data analysis is to extract meaning and
knowledge from the collected data.
4. The purpose of editing is to identify omissions, ambiguities, and
errors in the responses.
5. Omissions refer simply to the errors resulting from the failure of
interviewers to ask all questions.
6. Ambiguity is the result of an interviewer's failing to understand
the question he must ask his respondents.
7. When problems are identified during the editing stage, the
researcher's only options are to re-contact the respondent or to
throw out the entire questionnaire.
8. The coding of closed-ended questions involves judgment decisions.
9. It is generally easier to code open-ended questions as compared with
closed-ended questions because the researcher has more options with
open-ended questions.
10. Consider the following question: "What are your opinions about the
way the President is handling the economic problems facing the
nation?" This is an example of a closed-ended question.
11. Every questionnaire should have some open-ended questions.
12. A frequency distribution simply reports the number of responses that
each question category receives.
13. It is generally easier for an analyst to interpret the raw numbers
rather than the percentages that have been subsequently derived from
the data.
14. The decision to combine certain categories is dependent upon the
ease with which this can be done and the number of original
categories.
15. Frequency distributions, though sometimes unwieldy, do provide more
information than does the sample mean.
16. When the data is collected on a nominal scale, the researcher can
only use the mean and percentages to make meaningful conclusions.
17. A researcher who uses a frequency distribution or a single number
will necessarily reach the same conclusion.
18. Product usage is a useful variable when a researcher is seeking to
segment the population.
19. The difference between means is concerned with the association
between two questions, the question defining the group (for example,
smokers or nonsmokers) and another question (fear of fires).
20. Hypothesis testing of a difference between means can help
researcher decide whether a certain difference was obtained merely
by chance or because of differences in the underlying populations.
21. If a researcher's objective is to analyze a single question for
various subgroups, based on a frequency distribution, the technique
is termed cross-tabulation.
22. While analyzing his data, Mr. Thorough decided to use a frequency
distribution to study the response to each question. Later he became
interested in finding out if there were any significant differences
between the three subgroups in his study. The appropriate technique
at this stage is the difference between means.
23. When cross-tabulating data, the objective of the researcher is to
learn how the response variable varies from subgroup to subgroup.
24. Cross-tabulation is the analysis of association between two
variables that are nominally scaled.
25. Scale transformation involves the manipulation of scale values to ensure compatibility
with other scales
26. Consider the diagram shown below:
The correlation between sales and advertising is close to 1.0.
27. While inspecting political polling data, you notice that when the
distance jogged by the President (the independent variable)
increases, his popularity rating (the dependent variable) goes down.
This would indicate a positive correlation between the two variables.
28. A researcher made the following claim: "The correlation between the
weight of a car and its mileage is 45." Without additional
information, it is impossible to say whether this claim is true.
29. Multivariate analysis can be used to group variables or people or to
understand the relationship between variables.
30. In weighting, categories that are underrepresented in the sample are given lower
weights, while overrepresented categories are given higher weights
31. Coding for closed-ended questions is much more difficult than open-ended questions
32. The t-test is more sensitive to violations of equal the equal-variance assumption.
33. Nominal-scaled data is the best from the perspective of data analysis
Multiple Choice
- The process that is used to make the sample data more representative
of the population that has been surveyed is called
- editing
- coding
- weighting
- transcribing
- Which one of the following is not an assumption underlying a test statistic?
- the samples are independent
- the characteristic of interest in each population have normal distribution
- the two populations are of equal size
- the populations have equal variances
- The most primitive form of data from data analysis perspective is
- nominal scale
- ordinal scale
- interval scale
- ratio scale
- Analysis of association between two nominally scaled variables is called
- analysis of variance
- cross tabulation
- factor analysis
- cluster analysis
- If there are m levels of qualitative variables, ________ dummy variables are
used to specify them
- m
- m+1
- m-1
- m x 2
6. The process of data analysis involves
1. editing and coding the data.
2. collecting the data.
3. tabulating each question.
4. estimating differences between means.
a. 1
b. 2
c. 3
d. 4
e. 2, 3, and 4
7. The process of editing the data involves
a. detecting and correcting interviewer errors.
b. discovering inconsistencies between responses.
c. identifying omissions, ambiguities, and errors in responses.
d. eliminating data from ineligible respondents.
e. all of these.
8. While editing the data collected in a survey to determine the
psychographic profiles of heavy users of cereals, some of the
responses were found to contain omissions, ambiguities, and errors.
The researcher should do which of the following?
1. Try to recontact the respondents (if feasible) and obtain clarification.
2. Throw out those questionnaires which have many errors
or errors involving crucial questions.
3. Ignore questionable responses while making use of the
remainder of a respondent's data.
4. Use intuition and common sense to revise erroneous
responses, based on an educated assessment of how
respondent should have answered.
a. 1
b. 2
c. 3
d. 4
e. 1, 2, or 3
9. The coding of open-ended questions
a. is much more difficult than for closed-ended questions.
b. could require the coder to make a judgmental decision.
c. and the problems involved make it advisable to avoid open-ended
questions as much as possible.
d. can be difficult when the handwriting of the respondents is illegible.
e. all of these.
Use the following information for questions 10 and 11.
Intention to Buy | Number |
Will buy as soon as product is introduced | 80 |
Will see the price of the product at introduction and then Decide | 120 |
Will Purchase if others using it seem satisfied Will certainly not purchase | 75 125 |
10. The table above is an example of
a difference between means.
b. a frequency distribution.
c. a two-way tabulation.
d. a cross-tabulation.
e. an interval-scaled question.
11. The percentage of people who say that they will buy the product as
soon as it is introduced is: (1) less than that of those who will
purchase if others seem satisfied; (2) 20 percent; (3) 2/3 of the
percentage of people who will check the price before purchasing.
a. (1)
b. (1) and (2)
c. (1), (2), and (3)
d. (2) and (3)
e. (1) and (3)
12. Which of the questions listed below would be the easiest to code?
a. "What are the attributes that a good president must have?"
b. "What is your opinion of the influence of the Hare Krisna
group (a religious sect) on the minds of its younger members?"
c. "What influence, if any, do you think TV commercials have on children?"
d. "What are the major achievements and failures of the system of American government?"
e. "What type of oven do you have: (1) gas, (2) electric, (3) microwave, (4) other?"
13. Consider the following table:
Height | Number of people |
74 in. 72 70 68 66 | 10 5 12 5 10 |
Total | 42 |
a. The percentage of people with a height of 74 in. is close to 24 percent.
b. The average height of the sample is 70 in.
c. Most of the people in the sample are taller than or as tall as 70 in.
d. The most common height (mode) in the sample is 70 in.
e. All of these are true.
14. Consider the following table:
Height | Sample Size | Actual number in The population | Number in the sample who are members of the basketball team |
74 in. 72 70 68 66 | 10 5 12 5 10 | 150 200 250 200 250 | 7 4 6 2 3 |
(This table represents the heights and basketball team memberships
of high school students in a small mid-western town.)
1. The expected percentages of basketball players in the
population are 70, 80, 50, 40, and 30, respectively.
2. The estimated number of students in the city's high schools
playing basketball, based on the various samples, is 105,
160, 125, 80, and 75.
3. The estimated total number of basketball players in the city's high schools is 545.
4. The (weighted) average height of the city's basketball
players is different from the average height of the sample.
a. 1
b. 2
c. 3
d. 4
e. 1, 2, and 3
15. ABC Incorporated asked a sample of residents in each of two cities
to indicate the brand of shampoo most preferred by them. They found
that 40 percent of the people in City 1 preferred Brand A, as
compared with 30 percent in City 2. The results indicate that
1. Brand A is more popular with the sample respondents in City 1.
2. Brand A is more popular with the residents in City 1.
3. a hypothesis test of the significance of the difference in
sample percentages is needed before any conclusions about
differences in population can be made.
a. 1
b. 2
c. 3
d. 1 and 3
e. 1 and 2
Use the following information for questions 16 through 18
Type of oven owned | Income Less than More than $10,000 year $10,000/year | Total |
Conventional Microwave | 150 100 20 75 | 250 125 |
Total | 200 175 | 375 |
16. The table shown above is an example of
a. one-way classification.
b. four-way classification.
c. six-way classification.
d. cross-tabulation.
e. none of the above.
18. What inferences about the relation between income and oven ownership
may be drawn from the data above?
a. No inferences can be made without performing a hypothesis test.
b. The higher the income, the fewer the number of ovens owned.
c. A higher proportion of people in the lower income bracket own microwave ovens.
d. A higher proportion of those in the higher income bracket own microwave ovens.
e. In absolute terms, the ownership of conventional ovens seems to
decline with income, while that of microwaves appears to increase.
Use the following information for questions 19 and 20
Sales | Ad Budget for radio |
100,000 120,000 150,000 70,000 | 8,000 8,200 9,000 6,700 |
19. The correlation between sales and radio ad budget
a. is 0.
b. is negative.
c. is approximately .9.
d. is less than .75.
e. cannot be calculated, because the number of data points is too small.
20. Which of the following would be the most likely plot of the data points?
a. 1
b. 2
c. 3
d. 4
e. 5
21. A researcher might want to do multivariate analysis for which of the
following legitimate reasons?
1. To be able to group variables or people
2. To help improve predictive validity
3. To compare the difference between two means
4. To make the results look more meaningful and impressive to management
a. 1
b. 2
c. 3
d. 4
e. 1 and 2
22. A procedure in which the variables are brought to a mean of 0, and a standard
deviation of 1 is called ________________
a. Data modification
b. Variable Re-specification
c. Scale Transformation
d. Standardization
23. A procedure in which the existing data with a large number of variables are collapsed into
fewer variables, is called ________
a. Weighting
b. Standardization
c. Scale Transformation
d. Variable Re-specification