Preview Extract
Chapter 12: Advanced Topics, Research Frontiers, and Preparing the Final Report
TRUE/FALSE
1) The concept of heterogeneity is very important to marketing because people respond
differently to the same marketing variable.
Answer: True
Heterogeneity implies people are different and will, therefore, react differently to the same
stimuli.
2) Product recommendation systems use heterogeneity, hierarchical Bayes modeling, and
sample selection bias correction in a single application to predict products a customer will
like.
Answer: True
A product recommendation system is a statistical model that predicts which products a target
customer may like by examining the preferences and purchases of similar consumers.
3) The vast majority of statistical models used in marketing yield only one value for each “β,”
which marketers assume is valid for everyone; yet this concept is in direct contrast to the idea
of heterogeneity.
Answer: True
Although heterogeneity implies people react differently, most marketing models assume
everyone will react the same way when a particular situation is faced or a variable is
modified.
4) One of the most important analytical advances in marketing research in recent years has
been the advent of hierarchical models and Bayesian estimation.
Answer: True
This advance has been especially important when conjoined in the form of Hierarchical
Bayes, or simply HB, models.
5) When there is little data for an individual respondent, nonhierarchical methods permit the
borrowing of information from other respondents to obtain better estimates.
Answer: False
This is an important element of hierarchical methods and why hierarchical methods are
popular.
6) Although hierarchical models are excellent for allowing for individual differences among
respondents and measuring that difference in a meaningful way, they cannot provide
information about where it comes from.
Answer: False
One of the strongest benefits of hierarchical models is that they do provide information about
where the heterogeneity comes from.
7) The predictions from homogenous models are far superior to those from hierarchical
models.
Answer: False
Predictions arising from hierarchical models have been found to be far superior to those from
simple homogeneous models, in terms of providing a better fit to the actual data.
8) With the Markov chain Monte Carlo method of estimating parameters, the process finds
the best fit for each parameter before it moves on to the next parameter.
Answer: False
This process immediately goes to the next parameter, assuming the previous guess was the
right value. It continues to cycle through the parameters thousands of times.
9) The Markov chain Monte Carlo process will yield a single optimal value for each
parameter.
Answer: False
It produces a distribution for each parameter, called a posterior distribution.
10) It is important to remember that a sample that is collected non-randomly will always have
a bias.
Answer: False
Non-random selection does not necessarily mean that the sample is biased.
11) If a sample is collected through a non-random means, it is likely to have a selection bias.
Answer: True
It is rare that an overt selection bias has no effect on a study.
12) With end-point selection bias, data collection begins as soon as data fall above or below
an acceptable threshold.
Answer: True
This is a data-based selection bias that occurs when the researcher makes a judgment that
values are looking good or bad before beginning to record them.
13) When respondents can decide whether or not they want to participate, it is called unequal
selection bias.
Answer: False
It is called self-selection bias.
14) The selection bias of “data mining” occurs when researchers perform a huge number of
analyses and present the most significant results as if they were the only a priori hypotheses.
Answer: True
This is the definition of “data mining” or “fishing” among the study-based selection biases.
15) Generally speaking, selection biases are the most problematic when items are selected
based on values of the outcome variables.
Answer: True
By selecting items based on an outcome variable, what researchers wish to study.
16) The Heckman model can be thought of in terms of two equations: a “selection model”
and a “prediction model.”
Answer: True
Selection describes whether the variable is seen, and the prediction involves how well it
predicts.
17) The selection equation of the Heckman model is an ordered logit model.
Answer: False
It is a binary probit model.
18) For prediction, the Heckman model can use any regression-based model.
Answer: True
The type of regression model does not matter as long as the error term is normally
distributed.
19) In the Heckman model, if the selection model and the prediction model are estimated
separately, the two errors remain uncorrelated, and selection bias is automatically controlled
for.
Answer: False
The error terms remain uncorrelated, which means that the selection biases also remain.
20) In the Heckman model, the more correlated are the errors between the selection model
and the prediction model, the greater the evidence of a selection bias.
Answer: True
A high correlation between the errors indicates the presence of a selection bias.
21) In preparing a written report, a simple graph can convey far more than a table with the
same information.
Answer: True
Most individuals reading a written report can quickly gather the message presented in a
graph.
22) When writing a research report, it is important to remember that the report is designed to
provide information to decision makers, not to showcase intriguing findings of the research.
Answer: True
Research reports are written for decision makers so they can make intelligent decisions.
23) In terms of presenting the research results in the written paper, objectivity refers to the
lack of bias while framing the results so as to make them palatable to management.
Answer: False
Objectivity refers to the lack of bias while stating in clear, unequivocal terms what the
research implies.
24) When writing a report, use plenty of clichés and marketing jargon, because this is the
language that management understands.
Answer: False
Avoid clichés; they convey lazy, hackneyed thinking.
25) The executive summary is a condensed, accurate statement of the report’s most crucial
findings that is the essential element of nearly all research reports.
Answer: True
Because many managers read only the executive summary, it is crucial that this section be
accurate, concise, and well written.
MULTIPLE CHOICE
1) The fact that different people respond differently to the same sale, advertisement or other
marketing variables is
a. Bayes model of reaction
b. heterogeneity
c. homogeneity
d. a results of selection bias
Answer: B
Heterogeneity refers to the fact that not everything is identical; in marketing research, it
refers to the difference in individual’s responses.
2) Within marketing research, “heterogeneity” is generally used to refer to
a. differences in behavior
b. differences in sensitivity
c. differences in outcome
d. differences in characteristics
Answer: B
In marketing research, it is primarily differences in sensitivity or responsiveness that is called
heterogeneity. This is beta or coefficient heterogeneity.
3) In examining the differences between men’s and women’s weight, it can be said that
women have a lower baseline weight then men. This is an example of
a. intercept heterogeneity
b. slope heterogeneity
c. beta heterogeneity
d. coefficient heterogeneity
Answer: A
The intercept heterogeneity in this example refers to the different points where the regression
line would intercept the Y-axis for men and for women.
4) In examining the differences between men’s and women’s weight, it can be said that each
inch of height adds more to men’s weight than to women’s. This is an example of
a. intercept heterogeneity
b. slope heterogeneity
c. Bayes heterogeneity
d. hierarchical heterogeneity
Answer: B
In this example, the difference refers to the difference in the slope of the regression line.
5) Running a separate regression model for each person would yield the best results and
effectively deal with heterogeneity. We do not do this for all of the following reasons except
a. researchers don’t ordinarily have many pieces of data for each individual
b. researchers are looking for a unified explanation of a phenomena
c. individuals tend to give the same answers anyway
d. it is expensive to prepare a separate marketing approach for each individual
Answer: C
Heterogeneity implies that people do not tend to give the same answers.
6) Heterogeneity models go by all of the following names except
a. discrete heterogeneity
b. continuous heterogeneity
c. latent class model
d. Bayesian estimation model
Answer: D
Heterogeneity models include discrete (also called latent class or finite mixture model) and
continuous types.
7) Suppose that the distribution of the value of b for a sample of respondents were as follows:
23% have b = 2
17% have b = 3
30% have b = 4
30% have b = 5
Which type of heterogeneity model does this example fit?
a. discrete
b. continuous
c. Bayes
d. Heckman distribution
Answer: A
Because there are distinct, discrete values for the beta, it fits the discrete heterogeneity model.
8) The most general form of heterogeneity model in current use is called
a. discrete heterogeneity
b. Bayesian hierarchical heterogeneity
c. latent class model
d. finite normal mixture heterogeneity
Answer: D
Finite normal mixture heterogeneity is the most general form of heterogeneity model.
9) In real applications, all of the following complexities of heterogeneity are common except
a. a large number of humps for each coefficient
b. many coefficients
c. homogenous solutions are more accurate
d. the coefficients are correlated
Answer: C
In real applications, it is not uncommon to have a large number of humps for each coefficient,
to have many coefficients, and also to find that the coefficients may be correlated (e.g.,
people who care more about price may care less about styling).
10) The hierarchical Bayes model is called “hierarchical” because
a. it consists of a number of levels
b. it represses the will of the grassroots
c. the error term and slope are related
d. the intercept is known in advance
Answer: A
It is called “hierarchical” because it consists of a number of levels.
11) The hierarchical model allows for all of the following except
a. it allows for different slopes for different respondents
b. it includes individual regressions for each respondent
c. it allows for an explanation as to where the heterogeneity comes from
d. it allows for borrowing missing information from other respondents
Answer: B
The purpose of the hierarchical model is to be able to get away from individual regressions
for each person, yet capture the heterogeneity that is present.
12) With _______________, there is no search for the best parameter values, but instead
guesses are made about the value of each parameter repetitively to arrive at a solution.
a. Bayesian estimation
b. the Heckman model
c. latent model estimation
d. parsimonious modeling
Answer: A
By definition, the Bayesian estimation process guesses the value of the parameters in an
iterative process until a solution is found.
13) In Bayesian analysis, guesses follow a rigorous process called
a. latent Monte Carlo
b. Markov chain Monte Carlo
c. Heckman chain Monte Carlo
d. hierarchical Markov estimation
Answer: B
The guessing process used is Markov chain Monte Carlo, which is set of rigorous guidelines
on how the guessing should progress.
14) The result of the estimation process through rigorous guessing procedures is
a. the best beta value for each parameter
b. the best guess value for each parameter
c. a distribution of the various values of each parameter
d. an estimation of the distance between each data point and the normalized centroid
Answer: C
The Markov chain Monte Carlo process produces a posterior distribution of the values for
each parameter.
15) Having used a Bayesian estimation process, to determine the best value for a parameter,
researchers look
a. at the standard errors plot
b. at the highest point of the posterior distribution
c. for the highest factor loading
d. for the highest beta value
Answer: B
The “best” value is the highest point of the posterior distribution for that parameter.
16) A hierarchical model that is estimated by Bayesian methods is called a
a. hierarchical Bayes model
b. a Heckman model
c. Bayesian heterogeneity model
d. latent class model
Answer: A
By definition, the hierarchical Bayes model is a hierarchical model that is estimated using a
Bayesian method.
17) It is rare that an overt selection bias will
a. have an effect on the outcome of a study
b. have no effect on the outcome of a study
c. have a significant impact on a study
d. will influence a management decision advised by a study
Answer: B
An overt selection bias is extremely likely to have some type of effect on a study’s results.
18) All of the following are data-based selection biases except
a. unequal selection criteria
b. ad hoc rejection
c. endpoint selection
d. premature termination
Answer: A
Unequal selection criteria is a participant-based selection criteria.
19) _______________ is a data-based selection bias that occurs when data are deemed “bad”
using arbitrary criteria, often as they come in, instead of using explicitly-stated or previously
agree-upon standards.
a. Endpoint selection
b. Ad hoc rejection
c. Premature termination
d. Drop-out bias
Answer: B
With ad hoc rejection, the researcher decides arbitrarily that a piece of data looks bad and is,
therefore, rejected.
20) All of the following are participant-based selection biases except
a. unequal selection criteria
b. premature termination
c. drop-out bias
d. self-selection
Answer: B
Premature termination is a data-based selection bias.
21) _______________ occurs when researchers pre-screen potential participants on an
outcome variable.
a. Self-selection
b. Ad hoc rejection
c. Unequal selection criteria
d. Drop-out bias
Answer: C
With unequal selection criteria, a researcher uses an outcome variable to determine who
should participate in the study or in what capacity.
22) An example of a study-based selection bias is
a. endpoint selection
b. self-selection
c. unequal selection criteria
d. file drawer effect
Answer: D
The two types of study-based selection bias are file drawer effect and fishing (data mining).
23) Performing multiple, related studies and reporting only those that support the favored
hypotheses is called
a. drop-out bias
b. self-selection
c. unequal selection criteria
d. file drawer effect
Answer: D
Reporting only a study that supports a favored hypothesis or repeating a study until the
desired effects are obtained is called the file drawer effect.
24) The problem of selection bias was long been recognized by statisticians as a problem
until the introduction of the
a. Markov chain Monte Carlo method
b. hierarchical Bayesian estimation method
c. latent heterogeneity model
d. Heckman model
Answer: D
In 2000, James Heckman was awarded the Nobel Prize in Economics for his development of
the Heckman model to deal with selection bias.
25) Correcting for selection bias relies on the _______________ for selection
a. binary probit model
b. binary logit model
c. ordered probit model
d. ordered logit model
Answer: A
For the selection, the Heckman model uses the binary probit model because its normallydistributed error is somewhat easier to understand.
26) The model for selection in question #23 is used instead of others because
a. it calculates a more accurate error term
b. its normally-distributed error is somewhat easier to work with
c. it predicts more accurately the factor loading
d. its Gumbel-distributed error is somewhat easier to work with
Answer: B
Because the binary probit model produces an error that is normally-distributed and easier to
work with, it is used in the Heckman model.
27) The coefficients of selection bias correction models will automatically correct for any
selection bias if the
a. selection model is estimated before the prediction model
b. prediction model is estimated before the selection model
c. the selection and prediction models are estimated jointly
d. population standard deviation is known
Answer: C
By estimating the two equations jointly, the Heckman model automatically corrects for any
bias.
28) The most common type of selection bias correction model used by researchers is the
a. hierarchical Bayes model
b. fixed normal mixture model
c. Heckman model
d. bias estimation model
Answer: C
The Heckman model is the most frequently used selection bias correction model.
NOTE: The following results from the Application Case Study presented in the text show
means for selection of movies (Selection Distribution) and how they were rated (Prediction
Distribution). Refer to this table to answer the following questions.
29) From the following list of movies, which one is rated the lowest?
a. Classic
b. Family
c. Horror
d. Thriller
Answer: B
Family was rated the lowest (-0.389 mean).
30) Which gender is less likely to select a movie?
a. males
b. females
c. men and women are equally likely to select a movie
d. it cannot be determined from the given data
Answer: A
The -0.022 indicates males are slightly less likely to select a movie.
31) Which type of movie is rated the highest?
a. Action
b. Animation
c. Romance
d. Drama
Answer: B
Animation had a mean rating of 2.110, the highest.
32) Which types of movies are not selected often, yet are rated highly when selected?
a. Classic and Art
b. Thriller and Classic
c. Family and Horror
d. Horror and Art
Answer: A
Classic movies have a selection mean of -2.070, but a rating mean of 1.375; Art movies have
a selection mean of -0.523, but a rating mean of 1.462.
33) Which kind of movies are selected the least often?
a. Drama and Comedy
b. Family and Thriller
c. Family and Thriller
d. Classic and Thriller
Answer: D
Classic and Thriller have the highest negative means for selection.
34) In terms of selection, which two movies have the highest heterogeneity?
a. Action and Animation
b. Animation and Art
c. Family and Classic
d. Art and Horror
Answer: C
The standard deviation for the selection distribution is highest for Family (0.888) and Classic
(1.118) movies.
35) In terms of ratings, which movies have the highest heterogeneity?
a. Action and Animation
b. Animation and Art
c. Family and Classic
d. Art and Horror
Answer: D
The standard deviation for the ratings is highest for Art (0.809) and Horror (0.905).
36) Since many managers only read the _______________ of a written research report, it is
important for it to be accurate, concise, and well-written
a. table of contents
b. executive summary
c. body
d. conclusions and recommendations
Answer: B
The executive summary should be 1-2 pages in length and provide a brief summary of the
crucial findings of the report.
37) A discussion of the methodology used in the research would be found in which section of
the written report?
a. table of contents
b. executive summary
c. body
d. conclusions and recommendations
Answer: C
The methodology is part of the body of the report.
38) The methodology component of the written report should contain all of the following
information except
a. research design
b. sampling plan
c. data analysis procedure
d. research findings
Answer: D
The research findings should be presented in the results component of the report.
39) In using pie charts to illustrate data in written reports, it is advisable not to exceed
a. 3-4 slices
b. 5-6 slices
c. 10-15 slices
d. 20-25 slices
Answer: B
It’s best to stick to a maximum of 5 or 6 slices or the pie chart becomes confusing and
difficult to read.
40) To illustrate multiple comparisons and complex relationships, the best illustrative aid to
use is
a. pie charts
b. histograms
c. bar charts
d. tables
Answer: C
Bar charts are good for multiple comparisons and complex relationships.
SHORT ANSWER
1) Name the two main types of heterogeneity models.
Answer: 1) discrete (or latent class or finite mixture model) heterogeneity
2) continuous heterogeneity
Finite normal mixture heterogeneity is the more general form, which is acceptable as an
answer if the student understands how it relates to these two types.
2) What is the Heckman model, how does it work, and what is it used for?
Answer: The Heckman model is a means of correcting for sample selection bias by
estimating a selection model and a prediction model jointly through Bayesian methods.
3) What increasingly common web feature uses a Heckman model to correct of massive
amounts of missing data, parameter heterogeneity, and Bayesian estimation methods?
Answer: Product recommendation systems.
4) Identify the 4 items that should be summarized in the executive summary section of a
written research report.
Answer: The executive summary should contain:
1) research objectives (and nature of the decision problem)
2) key results
3) conclusions
4) recommendations for action
5) Identify the 4 components of the “body” of the written report.
Answer: The body of the reports should contain:
1) introduction (background)
2) methodology (research design, sampling plan, data collection and analysis procedure)
3) results
4) limitations
ESSAY
1) In the equation
heterogeneity?
, how do the various variables and coefficients relate to
Answer: Heterogeneity of characteristics is denoted by the X variable(s) in regression.
Heterogeneity of behavior or outcomes is denoted by the Y variable in a regression.
Heterogeneity of sensitivities is denoted by
b. This is called b or coefficient or slope heterogeneity, and is generally what is meant by
heterogeneity in marketing research—that different people or objects react differently (in Y)
to the same changes in the X variables. Intercept heterogeneity denotes differences in α.
2) Explain the general procedure of how the Markov chain Monte Carlo works.
Answer: The following steps are followed:
1) Make assumptions (called priors) about what parameters look like. These can be weak
assumptions or even randomly selected.
2) The program starts guessing for each parameter, assuming all other parameters have been
correctly determined.
3) It checks to see if the parameter was a good one by checking the fit with the data. If the
current guess fits the data better than the previous, the value is kept. If not, the program uses
a randomizing device to decide whether to keep it.
4) The procedure them moves to the next parameter, assuming the previous parameter was
correct, and repeats the procedure.
5) The process cycles through all of the parameters in this fashion, usually at least 10,000
times each, until it ends up with distributions of “accepted” guesses for each parameter.
3) Identify and describe the 8 types of selection bias.
Answer: 1) Data-based selection bias: Ad hoc rejection—data deemed as bad is rejected
Endpoint selection—data collection begins when the researcher feels the data has reached an
acceptable threshold
Premature termination—a study is terminated when the results appear to be “clear” to the
researcher
2) Participant-based selection bias: Unequal selection criteria—pre-screening of participants
is done based on one of the outcome variables
Drop-out bias—deleting the data of any participant who did not complete a survey
Self-selection—respondents can decide if they want to participate or not
3) Study-based selection bias: File-drawer effect—when a study is performed multiple times
and only the one that favors the desired results is reported
Data mining—performing a huge number of analysis and presenting the most significant
results as if they were the only a priori hypotheses
4) Discuss 5 general guidelines for a written marketing research report.
Answer: Guidelines for a written report include:
1) consider the audience
2) address promised information needs
3) be concise, yet complete
4) be objective
5) adopt a concise, business-like writing style
5) For most reports, what is the typical format (of 7 sections)?
Answer: 1) Title
2) Table of Contents
3) Table of Tables, Figures, or Graphs
4) Executive Summary
5) Body of Paper
6) Conclusions and Recommendations
7) Appendices
Test Bank for Modern Marketing Research: Concepts, Methods, and Cases
Fred M. Feinberg, Thomas Kinnear, James R. Taylor
9781133188964, 9781133191025, 9780759391710
