PHILOSOPHY OF SCIENCE
“I just don’t understand this philosophy of science stuff! As far as I can tell, all this [philosophy of science] amounts to is a bunch of high brow intellectual snobs getting a bunch of journal publications talking about stuff that nobody really cares about, especially marketing practitioners. Twenty years ago, the marketing discipline did not have any philosophy of science courses and the discipline got along just fine.”
Demonstrate your knowledge and personal opinion of “philosophy of science” and its role in marketing by reacting to the previous statement. (a) Take a position for or against the statement and (b) support your position using relevant literature.
Discuss the relationship between theory, research and science. What is the purpose of each? Given your discussion, is marketing a science? Why or why not?
Early theories of marketing thought were marked by a controversy regarding the universality of marketing theory. On one hand, proponents of a universal theory argued that marketing is a discipline comprising concepts, tools, theories, practices and procedures that are applicable in all contexts. Opponents, on the other hand, argued that marketing thought is context bound. Explain the basis of this controversy and your position on this issue.
All marketing professors take an ocean cruise on the S.S. Titanic II. While on the cruise, the ship accidentally (we think) strikes a boat filled with accounting professors. Unfortunately, the Titanic begins to sink rapidly (the boat with the accounting professors is totally unscathed). There is only one good lifeboat on the Titanic II and it only seats 4 people.
a. Who among the thousands of marketing professors on the Titanic should be saved?
Defend your choice by listing their specific contributions to the marketing literature.
b. Who would be the first four you would throw off the boat?
Many marketers believe that marketing is an applied discipline. Develop in detail the argument in favor of marketing being considered an applied discipline. Others contend that marketing is not an applied discipline. Please develop in detail this position. What are the implications of both positions for marketing research, teaching and practice. What are your personal views concerning this issue?
Several scholars/authors have stated that, “Marketing is an Art, not a Science”. Still others have commented that, “Marketing is a Professional Discipline, not an Academic Discipline”.
a. What is meant by the terms “science”, “art”, “professional” and “academic”?
b. Evaluate and distinguish between these positions.
c. What are the consequences, if any, for marketing teaching and research if this
positions are held?
d. What is the nature of the marketing discipline?
During the last few years there has been an ongoing debate concerning the fundamental philosophical orientation that marketing does (and should) have. Writers such as Peter, Olson, Arndt and Anderson advocate what they call a “Relativistic/Constructionist” approach. They contrast this with the so called “Positivistic/Empiricist” approach, which supposedly “dominates” the marketing discipline today. Other writers have suggested that Peter, Olson and Anderson have misconstrued the fundamental tenets of current approaches to marketing science.
a. What are the fundamental tenets of the approach to science sometimes labeled
“modern” or “logical” empiricism?
b. How do the fundamental tenets of modern empiricism differ from the
“Relativistic/Constructionist” approach advocated by the above cited authors?
c. Which of the two approaches do you believe would be more useful as a guide for
progress in marketing science? Be sure to specify precisely why you choose the
alternative you do.
d. At the present time, the Section Editor for the Measurement Section of the Journal of
Marketing Research is a “relativist”. Do you see any unique problems that a relativist
would have in attempting to fulfill the functions of a Marketing Measurement Section
Editor? If so, what are these problems?
There are many sayings such as “history repeats itself,” “we learn form our mistakes,” and “someone has probably thought about that before”. These sayings, while varied, point out the importance of history. All disciplines have historical roots and development and marketing is no exception. Bartels and others have stated that marketing has been impacted by events which have occurred and the people involved in the “practice” of marketing. While the exact amount and type of impact can be debated, it is true that the present status of the marketing discipline has been impacted in some way, by what has transpired in times past.
Differentiate between normative theory and positive theory. Must the development of positive theory necessarily precede the development of normative theory? Do all normative theories rest upon an essentially positive base?
Summarize the empiricism-rationalism debate. To what extent does the discovery/ justification distinction help us understand the debate? To what extent does the fallible/infallible distinction help us? In your judgment, which side of the debate had the better case?
Many theories and models in marketing consist primarily of a diagram with boxes, each having a single concept or construct and various arrows connecting the boxes. Are these theories? If no, what additionally would be needed to construct a theory? If yes, discuss how they meet the requirements of a theory.
Some analysts think segmentation is a concept that is in the declining stage of its life cycle.
1. What is the basis for this belief and do you agree?
2. Discuss two other marketing concepts or constructs that have had historical significance
to marketing that have gone through the life cycle. Discuss the relevance to marketing
The product life cycle (PLC) concept has been applied to a variety of phenomena in the discipline of marketing. Use the PLC as a framework for discussing marketing research. In doing so, identify streams of scholarly research that you believe to be in each stage of the PLC. Explain and critique each stream, citing examples of relevant research. Be sure to state the criteria you use to assign each stream of research to a life cycle stage. What leads to rapid growth of research in a particular stage? Use examples in your answer.
Compare and contrast the functional school of marketing thought and the managerial school.
GENERAL MARKETING KNOWLEDGE
Select one of the following major areas of marketing and identify the ten most significant events and individuals which have had the most impact on its present state of development. Include mention of books, articles, monographs, etc. which have been published that reflect those significant events and people. The major area you select must come from the following:
Marketing Theory Promotional mix
Consumer Behavior New product development
Sales Management Marketing research
Make certain that when appropriate, you link the various events, people, publications, etc. together to show the pattern of thought development in the area you have selected.
Discuss any five (5) of the following concepts, writing about 2 pages on each. Be sure to cite any relevant research
Satisfaction Transaction Cost Analysis & Principal-agent theory
Store image Behavioral decision theory
Comparative advertising Reference pricing & its effect on choice
E-commerce Localized vs. Globalized Strategies
Competitive market structure Rationale for sales promotions
Memory representation Expectations and Perceived Service Quality
Brand loyalty vs. spurious loyalty Experimental vs. quasi-experimental design
Localized vs. global strategies Regression vs. ANOVA
Affect vs. cognition Multidimensional techniques vs. factor analysis
Prisoner’s dilemma Stepwise regression
Theory of reasoned action Stochastic models
Heteroskedasticity Conjoint analysis
The marketing concept is generally considered one of the most important foundations for the practice of good marketing. As a consequence, the marketing concept is often “taken for
granted” as a sound philosophy by which to operate a business.
a. What limitations, if any, do you see in the application and implementation of the
marketing concept? Be sure to describe and evaluate recent marketing literature that
has addressed both the appropriateness and the implementation issues associated with
the marketing concept.
b. Assume that you are a director for The Marketing Science Institute and you have
named research on the use and implementation of the marketing concept as an MSI
priority. Develop a general list of research questions that you believe would be
worthy of funding.
What is the marketing concept? Trace the origins of the concept and then relate your version of the current from of the concept. Is the marketing concept a theory? Why or why not? How is a market orientation related to the marketing concept? When you teach marketing, will you include the marketing concept in the topics you cover; why or why not?
If there is one concept that has continuously captured the fancy of marketing managers and scholars over the twenty years, it is market segmentation.
a. Explicitly discuss the meaning of market segmentation and the rationale underlying
b. Identify and discuss the variables that have over the last 10 to 15 years received
considerable attention as basis for segmenting the market. (Rather that an arbitrary
ordering of these variables, place them, during your discussion, in a framework useful
for categorical purposes.
c. If a marketing manager today (1990) asked you to tell him/her which of the above
variables is (are) the superior segmentation variable(s), what would you tell him/her?
Marketing modelers often claim that "optimal site selection for a new retail outlet in geographic space is similar to optimal positioning of a new product in perceptual space." Briefly review modeling approaches in these 2 areas and identify (or develop / propose) a model that can be used (with few modifications) for retail site selection and optimal product positioning.
In July, J.C. Penney opened seven new stores in the area. In an unusually opportunistic move, the stores were purchased from a bankrupt retailer at auction. Penney’s main competitor in all locations is Hecht’s. Two of Penney’s main objectives are: 1) to overcome its image as a “plain Jane retail chain” and 2) to establish a secure niche in the middle of the market between upscale competitors and discount mass merchandisers. Over the next five years, the marketing research division will be monitoring consumer perceptions to evaluate the company’s progress in repositioning itself. Plan a research program for J. C. Penney to use in evaluating its position in the market, now and during the next five years. Proceed as you would in planning any study, by doing the following:
a. Briefly summarize the relevant literature, and derive at least 2 testable hypotheses.
b. Draw a model showing the important concepts and how they are related.
c. Present all stages in the methodology, including the statistics you will use to test your
d. Discuss the implications of your research for the company’s marketing strategy
DIFFUSION OF INNOVATION stSome researchers have suggested that as marketers move into the 21 century, better
understanding of the diffusion and adoption patterns of technology-based innovations will be a critical piece of the complex puzzle of how to remain profitable in a high-tech information world. Discuss each of the following statements or questions:
？ What are the general theoretical underpinnings of diffusion on innovation theory?
Identify and assess the measurement issues associated with the theory (i.e., reliability,
validity, theory assumptions)?
？ What are the critical issues surrounding the employment of diffusion theory to
understanding consumers adoption rates of advanced technology-based innovations?
？ Discuss the constructs of “perceived risk” and “forced adoption” and how they impact the
diffusion of innovation process?
？ How might consumers innovativeness and adoption processes impact the marketing
opportunities and problems associated with the “information superhighway”?
What are the important assumptions (or limitations) of the Diffusion of Innovations Models in marketing? Since the Bass diffusion model was proposed in 1969, what have been the major research streams in the area? Briefly explain each of these streams with appropriate cites.
Suppose that you tell a fellow AMA doctoral consortium candidate that you have taken a course that examined methods of collecting and analyzing qualitative data. She responds: “Why are you bothering with that weird science? Those kind of people don’t do real science. They don’t use scientific sampling; they collect unreliable, anecdotal data. There’s no rigor involved; its
too touchy-feely. You can’t establish the validity of their claims. Besides anybody can do it.” Formulate a response to the colleague grounded in the philosophy of science and methodological principles to which you have been exposed.
Describe and discuss the philosophy and method of Existential-Phenomenology. Using examples from recent consumer research, illustrate the advantages and potential limitations of the phenomenological interview. Compare and contrast the existential-phenomenological approach to consumer understanding with ethnography as an approach to consumer understanding.
There are numerous ways to classify the research methodologies used by researchers and practitioners alike, to collect data for investigating today’s marketplace behaviors, consumer
behavior issues, or to answer marketing managers questions. There is a growing acceptance to classify research methods as being either qualitative or quantitative in nature. While most researchers believe that good solid research endeavors will incorporate both qualitative and
quantitative methods into their practices, there are those researchers who believe that each “family methods” while different, can provide similar results and information to researchers and
practitioners alike. Using your knowledge and understanding of both qualitative and quantitative research methods, provide a supportive discussion to each of the following sub-questions. [Make certain you support your answers by including appropriate literature cites where necessary]:
a. What are the general differences between qualitative and quantitative research
methods? Make sure your discussion includes the objectives/goals that underline
the two types of research methodologies, their strengths and weaknesses, and
potential information applications for practitioners and researchers.
b. Select a particular research method that is truly a member of the family of
qualitative research methods and a method that is a member of the family of
quantitative research methods and briefly discuss how each method might be used
to investigate the following research question:
Do mature consumers (ages 65 and older) use significantly different evaluation
processes and constructs than do established consumers (ages 30 to 50) to
determine their “level of materialism” and “satisfaction” of products and/or
services in the disposition stage of a post-purchase consumption process?
c. Regardless of the research method used to investigate a marketing phenomenon,
issues of “reliability” and “validity” are critical t any researcher. Assessment
techniques to address these critical issues are well established within qualitative
research methods, but are difficult to implement within qualitative research methods.
Discuss some of the techniques and criteria used by researchers to address the
issues of reliability and validity within qualitative research methods.
Today, many marketing researchers strive to investigate many marketing phenomena by asking important questions to a variety of respondents. When investigating consumers attitudes and behaviors towards products, services and marketplace behaviors, it is a very common practice to design and use sets of predetermined scale measurements to aid in guiding the respondents’
thinking and answers t the questions presented in a standardized questionnaire (survey instrument) format. In marketing research, all question/scale formats can be logically classified as being either nominal, ordinal, interval or ratio in nature. Given your in-depth
understanding of scale development issues, provide a detailed, supportive discussion of each of the following sub-questions [Include examples where appropriate]:
a. What are the scaling assumptions (properties) activated with each of the above four
categorical types of scale measurements? Within your discussion, make sure you point
out (or address) any similarities and differences.
b. For each of the four types of scale measurement formats, what are the hierarchical
implications regarding the appropriateness of “measures of central tendencies” and
“levels of dispersion” of the information being provided?
c. From an informational interpretation perspective, what are the differences in the
information provided to the research/practitioner by these four scale measurement
d. What data analysis problems (if any) are created by researchers who abuse (or misuse)
the inherent scaling properties associated with these four scale measurement formats?
[Make certain you provide concrete examples to support your thoughts.
Survey methods and depth interviewing are widely used data-gathering techniques in both commercial and academic marketing research. Discuss the strengths and limitations of these techniques form a philosophy perspective (i.e. paradigmatic presumptions, strengths, weaknesses, blind spots, lacuna, etc.). Draw on literature that identifies the methodological weaknesses and/or attempts to address them in formulating your answer. Identify appropriate uses and common misuses of these techniques in either applied or academic contexts. Relate this literature to your own experiences where appropriate.
Constructs occupy a great deal of social scientists’ as well as marketing scholars research attention. As a prospective Ph.D. in Marketing, you too will spend much of your research attention directed at developing, measuring, and investigating constructs. Make certain that you support your answers by including appropriate citations form the literature.
a. What is a construct?
b. Reliability and validity are two characteristics of construct measurement. What are each
of these characteristics? Why are they given so much importance?
c. How do marketing researchers assess these characteristics of construct measurement?
Compare and critique the techniques used in assessing these characteristics of a construct.
d. What is the difference between a “mediating” construct and a “moderating” construct?
How does a researcher determine whether a construct is a mediator or a moderator?
Explain how you would go about developing a measure of job satisfaction of industrial salespeople. Include in your discussion the procedures) you would use to assess reliability and validity of the proposed measure. If you are familiar with more than one approach to developing such a measure, make sure to include in your discussion the method you would use and why you would use it.
Describe the procedure you would employ in developing a new measure by which the job satisfaction of a retail sales representative could be assessed. Indicate the type of scale you would use and why, and detail the specific steps you would undertake to assure the validity and reliability of this measure.
Describe the procedure you would employ in developing a measure by which the job satisfaction of a retail sales representative of a hardware chain could be assessed. Indicate the type of scale you would use and why, and detail the specific steps you would undertake to assure the validity and reliability of this measure.
Reliability and validity are areas of concern to researchers. Briefly explain the two concepts and discuss their importance in research (both positivistic and interpretive). What methods) would you use to assess reliability and validity?
You have a set of data from consumers comprised of the following:
？ Likert-type responses to questions concerning attitudes toward automobiles
？ Past purchase information concerning automobiles
？ Responses to life style questions
？ Demographic information on gender, age, education, income, marital status, and family
Your overall research objective is to better understand how attitudes, lifestyles, and demographic information might be related to automobile purchase. Describe how you would use the following procedures to extract that information: (a) factor analysis; (b) multiple regression analysis; (c) discriminant analysis; and (d) analysis of variance (ANOVA and MANOVA). List any necessary assumptions that you would have to make relevant to these analyses.
Many experimental design studies are subject to the following criticisms:
a. They use student subjects.
b. Alternative explanations are not considered in interpreting the results.
c. The task involves manipulations that are extreme.
d. They are subject to demand artifacts.
e. Sample sizes tend to be very small.
f. Pen and paper scenarios cannot approximate the real world activities they purport to
Discuss each of the criticisms, citing relevant literature and examples. Under what conditions is each criticism valid and appropriate? How can researchers minimize such criticisms in their research designs?
Conjoint Analysis and Multidimensional Scaling have been widely used in marketing to design new products. What are the advantages and limitations of these two methods? Explain how you can (or can not) use these two methods to design radically new products (new-to-the-world products) when consumers do not even know what to expect in the new product.
Conjoint analysis and multidimensional scaling are two popular multivariate techniques used in the consumer goods industry. Compare and contrast these two methods and list their strengths and weaknesses. Explain situations in which one method is preferable over the other.
Structural equation modeling is an increasing popular statistical approach in marketing research, even though the methodology does have obvious weaknesses and shortcomings. Describe the major limitations of structural equation modeling--what kinds of problems are not good candidates for SEM analysis? For each kind of problem or each limitation that you identify, indicate what other statistical approaches are available which avoid or minimize that problem?
You are conducting an empirical comparison of two competing theories using a structural equation modeling analysis, based on a sample size of 160. Both theoretical models involve the same latent constructs, and the same measures, and both are supported by theoretical argument. In fact, the two models are identical except that Model B includes three additional paths in the
structural model. Here are the values for some overall fit indices:
212chi (df) CFI pCFI RMSEA
Model A 257.7 (117) .89 .74 .087
Model B 205.4 (114) .93 .76 .071
ldf = degrees of freedom 2parsimony adjusted CFI
Looking closely at the diagnostic information from these two analyses, you notice that the stem-leaf plot of the standardized residuals for Model B is essentially bell-shaped, while the plot for Model A includes a few extreme (outlier) values on the negative end. Looking at the Modification Indices for Model A, you see some extremely large values associated with covariances among the measurement error terms. You create a modified Model A--called Model A'--by freeing two of these paths, and get these results:
2Chi; (df) CFI pCFI RMSEA
Model A' 197.1 (11 S) .94 .77 .067
You also note that the squared multiple correlation for the structural equation predicting the key dependent construct is equal to 1.05 in Model A?. This is a substantially higher value than you obtained for either Model A or Model B.
Sketch the "Interpretation" section of a paper reporting on this research. How well do these models "fit?" If you had to favor one of these three models, which one would you prefer? Why? What other key pieces of information might help you make this decision?
Karl Joreskog developed an overall chi-square test for assessing the fit of a structural equation model (SEM). Today, however, researchers typically assess fit by evaluating a number of indices, and often give little weight to the chi-square. What is the rationale for disregarding the chi-square, or for treating this value as only approximate? Why might researchers conclude that some alternate fit indices are more useful or more appropriate for assessing model fit? Under what circumstances would researchers rely on the chi-square as an exact test of overall model fit?
In a recent Journal of Marketing article, Brown, Cron and Slocum (1997) estimated a structural equation model intended to explain links between salesperson emotions and behavior. Their Figure 2, showing their final model, and their Table 2, are reproduced below.
(a) Based on the evidence presented in the figure, and the footnote, evaluate the fit of the
model. Does it fit well, or poorly? What additional information would be important in
shaping your evaluation?
(b) Their initial model did not include the direct path from "Goal-Directed Behavior" to
"Positive Outcome Emotion." In the original analysis, the modification index for that
path was 13.?_. Brown, Cron and Slocum report adding the path because of the
modification index, and because the resulting parameter estimate was interpretable.
Comment on their decision. Identify both positive and negative aspects, if any.
In defining the term "causation," in connection with the methodology of structural equation modeling, Hair, Anderson Tatham, and Black (1992, p.427) write: "[causation] requires that
there be sufficient degree of association (correlation) between the two variables, that one variable occur before the other (i.e., one variable is clearly the outcome of the other), and that there be no other reasonable cause of the outcome. Although in its strictest terms causation is rarely found, in practice strong theoretical support can make empirical estimation of causation possible." To what
extent do you agree with this statement, and to what extent do you disagree? Be sure to address the argument that we can sometimes estimate causation empirically, particularly in the context of structural equation modeling methods. Be specific. If we can do this, then how do we do it? If we can't do this, then why not.
An academic researcher desires to test a simple causal model of the form:
The two beta weights in this structural equation model are sample statistic estimators of population parameters. Assume that the research design was a one-shot cross-sectional mail survey of the membership of a professional association, such as the National Association of Purchasing Managers. An overall usable response rate of 18% was obtained.
a. What are the major types of validity with which this researcher should be concerned?
b. For each type of validity named above, what are the major sources of error
c. For each error source named above, what can or should be done to control for, test for
a measure the error present?
Recent empirical work in relationship marketing suggest the action of suppressor mediating variables in the trust ？ commitment causal path. Several of these variables have been linked with the moderating effect of length of relationship. It is suggested that due to the emergence of these suppressors, trust has greater effect in shorter rather than longer relationships.
a. Describe the action of a suppressor mediating variable and the pattern of causal
path coefficients one would observe if one is present.
b. Discuss the role of a moderator variable in triggering the action of a mediating
variable. Provide a statement of a testable hypothesis that captures this effect.
c. Describe a study capable of testing for the action of both moderators and mediators
in the same causal network.