€¦  · Web viewThe Alphas and Omegas of ... It was not uncommon for Rajan to deliberate and ponder whether it was more appropriate to use the word ‘insure’ or ‘ensure ...

The Alphas and Omegas of Research and Publishing: A Primer on Rajan Varadarajan’s Perspectives

INVITED COMMENTARY FOR PUBLICATION IN

LEGENDS IN STRATEGIC MARKETING: RAJAN VARADARAJAN

Kartik KalaignanamAssociate Professor of Marketing

Moore School of BusinessUniversity of South Carolina

1014 Greene StreetColumbia, SC 29208

Phone: (803) 777-4440E-mail: [email protected]

http://sc.edu/study/colleges_schools/moore/directory/kalaignanam_kartik.php

It is an honor and privilege to be invited to serve as a commentator for “The Legends of

Marketing,” a program that is designed to honor exceptional marketing scholars within that

community and create a repository of educational content to inspire students, scholars, and

managers. To be invited to do so for a distinguished scholar such as Professor Varadarajan is an

even bigger honor. Reflecting on Professor Varadarajan’s decades of scholarship on marketing

strategy within the confines of an essay or a series is daunting to say the least. While Professor

Varadarajan’s contributions to marketing strategy are numerous and well documented, his

mentoring of doctoral students at Texas A&M University over three decades deserves special

mention.

I consider myself fortunate to have had the opportunity of being exposed to Professor

Varadarajan’s perspectives on research and publishing at an early stage. I enrolled in Professor

Varadarajan’s doctoral seminar on marketing strategy while I was still pursuing a Master’s of

1

Science degree in Marketing. While the format of this seminar was similar to other doctoral

seminars, the extent to and depth with which assigned readings were discussed was remarkable.

He had exacting standards for evaluating the merits of research papers. Professor Varadarajan

would spend considerable time discussing the precision with which constructs were defined, the

logic and precision of the hypotheses, the soundness of proposed relationships and commonly

observed mistakes in the description of methods and findings. These requirements which he

refers to as the “litmus test” of good research seemed overwhelming to seminar participants at

the time. I have, however, no doubt that this training early on helped me to understand the

research and publication process. My goal in this commentary is to share his perspectives on

some of these fundamental issues in the hope that other students of marketing strategy would

also find Professor Varadarajan’s guideposts useful for their careers.

The commentary is organized as follows. First, Rajan’s guideposts on framing research

questions, construct definition/conceptualization and building conceptual/organizing frameworks

is presented. Next, his guideposts on how to craft hypotheses and guard against pitfalls in the

hypotheses development stage are discussed. Finally, Rajan’s perspectives on how to write

effectively for a scholarly audience are discussed.

RICHNESS AND RIGOR IN CONCEPTUALIZATION

It is hard to overemphasize how gifted and adept Rajan is at synthesizing and organizing vast and

complex topics into parsimonious conceptual frameworks. He sought to build integrative

conceptual models by leveraging theoretical perspectives from several allied and ancillary

disciplines as opposed to being wedded to a single theoretical paradigm. He would describe his

integrative frameworks as “efforts that sought to bring structure to chaos.” In a commentary

written in honor of Rajan receiving the Converse Award, Sundar Bharadwaj notes that “One of

2

his key strengths is his ability to weave diverse viewpoints and theoretical perspectives into a

single consistent and effective story. In some ways, he is like a master weaver that can somehow

combine multiple streams of research into a coherent story.” My early interactions with him left

me awestruck by his ability to link business articles in magazines such as Forbes and Fortune to

theoretical principles. For instance, in one session of his seminar, Rajan queried doctoral students

as to what explained the co-existence of “do it yourself” tax preparation software (i.e., good)

and tax consultants in the marketplace (i.e., service)? After presenting several examples and

cautioning that top tier journals are only interested in topics of scholarly value, he reframed the

question as what explains the co-existence of internal and external market exchanges at the

household level? The deliberations ranged from analyzing theoretical frameworks related to time

allocation (Becker 1965), embedded markets (Frenzen and Davis 1990) to parallel issues such as

‘make or buy’ or vertical integration at the firm level of aggregation. It was common practice for

him to invite interested doctoral students to his office and hand over manila folders consisting of

documents related to conceptual ideas he was interested in (with the caveat to return the material

to him in due course!). As one browsed through the material Rajan had painstakingly assembled

(i.e., scholarly articles, business reports, hand written notes), one could see how deeply and for

how long he pondered about the merits/demerits of research ideas. I found this approach of

building theory from the ground up to be valuable. I began appreciating the fact that the world of

marketing theory and practice were not that far apart.

High quality conceptual development is a cornerstone of good research. While doctoral

students realized that this was a threshold they had to meet to discuss ideas with him, how to get

started was a practical concern. It is common that doctoral students at an early stage in the

program struggle to identify meaningful issues/topics for research. This is especially so when

3

students burden themselves with the thought that the goal of research is to challenge and alter

received wisdom. He would put such concerns to rest and suggest that we (researchers) ought to

recognize the incremental nature of scientific progress. In his words, “as researchers, we need to

take baby steps to advance knowledge.” For struggling doctoral students, he had practical tips

on how to get going. One approach he recommended was to identify a relationship (XY) for

which the expected direction was arguably equivocal either from a conceptual or empirical

standpoint. A contribution could be potentially made by identifying boundary conditions that

account for why the direction of this relationship has proven to be inconclusive in past research.

The quality of contributions of this type would depend on the kind of variables identified and the

extent to which these variables account for the variance in the observed relationship. Another

approach would be to identify a phenomenon of interest based on trends in the marketplace (e.g.,

deconglomeration, multimarket competition, outsourcing, sustainability) and develop an

inventory of propositions related to its drivers and performance consequences. This grounded

approach to theory building, he would note, had greater potential for impact.

Further, he had clear advice on how to frame research questions. Instead of posing

questions such as, “does market share impact firm profitability,” students would be advised to

think about when this relationship was likely to hold versus not. Research questions that elicited

simple yes/no responses are often not interesting to pursue. A persuasive way of framing this

research question would be, “under what conditions does market share positively impact firm

profitability?” Alternately, one could pose the question as, “what factors explain the variance in

the market share-firm profitability relationship?” Further, he had clear guideposts for

articulating the motivation and contribution of the paper. The introduction section should clearly

state why studying the problem matters and why a better understanding of the phenomenon

4

matters. A common mistake committed by doctoral students was to have a long and rambling

introduction of several pages on topics that only seem to be remotely related to the research

topic. Proposals that claimed that their contribution lies in the fact that previous studies had not

examined a problem or issue would be summarily dismissed. This skepticism was not because

being the first to investigate a topic was a problem in itself. The concern was that topics/issues

may not have been investigated in past research because they were deemed unimportant or not

addressable because of serious challenges. Students were instead advised to think about their

contributions in terms of whether their work extends extant theory, provides evidence of

moderator variables that hold implications for marketing practice or identifies mediational

processes that resolve theoretical/empirical puzzles in the literature.

Constructs and measures are fundamental building blocks for theoretical development

and advancement. Not surprisingly, Rajan’s comments on do’s and don’ts regarding

conceptualization often exceeded his comments on other sections of the manuscript. A lesson

learnt was the need to define constructs clearly and early on in a manuscript. Without definitions,

he would be loath to evaluate the merits of an argument. Good definitions have two parts. First, it

assigns an object to a group or class whose characteristics are also its own. Second, it specifies

wherein the object differs from other members in its class. Therefore, when introducing new

constructs, one should explain how it is similar to and distinct from extant constructs. This rigor

in defining constructs is necessary not only for improving the exposition of a study but is also

useful for directing meta-analytic efforts and generating empirical generalizations. To emphasize

the importance of construct definitions, he would invoke ideas from the “Story of Philosophy”, a

classic by Will Durant (1961):

5

“There was a hint of this new science in Socrates’ maddening insistence on definitions

and in Plato’s refining of every concept. Aristotle’s little treatise on definitions shows

how his logic found nourishment at this source. “If you wish to converse with me,” said

Voltaire, “define your terms.”. How many a debate would have been deflated into a

paragraph if the disputants had dared to define their terms! This is the alpha and omega

of logic, the heart and soul of it, that every important term in a serious discourse shall be

subjected to strictest scrutiny and definition. It is difficult, and ruthlessly tests the mind;

but once done, it is half of any task”

It is difficult to think about operational measures for constructs that are imprecisely

defined or stated in vague or broad terms. Even if the paper was conceptual in nature, it was the

author’s responsibility, he would argue, to propose ways to operationalize the constructs and lay

the foundation for future empirical tests. Donning his editorial robes, he would caution that

papers in which the operational measures and the construct’s conceptual domain were

disconnected stood little chance of surviving the review process. One of the projects on which I

had the good fortune of collaborating with him was CRM outsourcing. The genesis of this

project was the practice of firms in the United States to offshore CRM activities to countries such

as India and Phillipines, after the Internet bubble burst in 2002. I recall the numerous discussions

on how to conceptualize CRM outsourcing. How was CRM outsourcing similar to and distinct

from outsourcing in general? How was CRM outsourcing different from the outsourcing of other

marketing activities such as advertising and PR? How was front-office CRM outsourcing

different from back-office CRM outsourcing? Finally, how was CRM outsourcing to domestic

vendors different from CRM outsourcing to international vendors?

6

After multiple iterations, CRM outsourcing was defined as “a firm contracting with other

independent firms to perform activities related to the establishment and maintenance of customer

relationships that were previously performed within the boundaries of the firm.” This

collaborative effort, published in JAMS in 2012, was a conceptual piece that explicated the

concept of CRM outsourcing and developed propositions on the drivers and performance

consequences of CRM outsourcing. In this paper, we examined macroeconomic, industry, firm,

product and task related drivers of CRM outsourcing and its impact on CRM and financial

performance. Although the paper was conceptual, we spent considerable time discussing

measures and ways to test the propositions. This project had an impact on deepening my interest

in the outsourcing phenomenon. Some of the propositions in this project (e.g., the negative

relationship between CRM outsourcing and CRM performance) served as the motivation for a

follow-up research paper. I collaborated with Tarun Kushwaha, J.B.E.M. Steenkamp and Kapil

Tuli to test an empirical model that sought to understand why the performance effects of CRM

outsourcing varied so much. In this paper, we tested the moderating influence of a) pre-sales

versus post-sales CRM activities, b) vendor’s cultural distance and c) the firm’s marketing and

operational expertise on the relationship between CRM outsourcing and shareholder value. This

paper was published in Management Science in 2013. These projects sustained my interest in the

topic and led to another project in which an all Carolina team (Tarun Kushwaha, Tracey Swartz

and myself) investigated the impact of outsourcing of product development on short-term and

long-term product quality and factors that moderated these relationships. This paper is

forthcoming at JM. While the empirical setting and methodologies across these projects were

quite distinct, the focus on how outsourcing of business processes impacts market and financial

outcomes remained unchanged.

7

LUCIDITY AND LOGIC IN HYPOTHESES DEVELOPMENT

Rajan’s students would remember the stringent requirements he had about how the hypotheses

were to be stated or worded. His requirements would range from stating the hypotheses precisely

to eliminating logical fallacies in the hypotheses. He would not let hypotheses be imprecisely

stated or their logic be dodgy. His criticism (rightly) was that readers have limited attention

spans and could misunderstand hypotheses that are not clear. It was therefore incumbent upon

authors to state the hypotheses clearly and precisely. His specific concerns pertaining to

hypotheses primarily related to whether a) the proposed relationships are precisely stated, b)

whether the proposed relationships are truisms, intuitively obvious or self-evident and c) whether

the relationship were stated in generalizable terms.

Stating Hypothesis Precisely: Professor Varadarajan would stipulate that authors clearly

state the expected direction of the proposed relationship. Hypotheses that are ambiguously stated

were according to him, poor or inadequate theorizing on the author’s part. Consider for instance

the following hypothesis.

Hypothesis: Firms emphasizing global standardization will view the importance of the

components of the marketing plan differently from firms that do not stress

standardization.

His criticism of a broadly stated hypothesis such as this was that merely hypothesizing

differences exist between firms emphasizing global standardization versus those that do not but

being ambiguous about the directionality of the relationship or the nature of differences is of

limited value from the standpoint of gaining insights into the phenomenon. Further, the fact that

8

authors claim support for a hypothesis that is ambiguous about the expected direction to start

with was problematic. If the arguments for a relationship were indeed too equivocal to propose a

specific direction for the effect, he would advise doctoral students to present arguments for and

against the hypothesis and let the empirical analyses determine the direction of the relationship.

Avoiding Self-Evident or Intuitively Obvious Hypotheses. Top-tier publications in marketing are

interested in reporting findings that are interesting. Any scientific theory is only as good as the

possibility of falsifying it through observational tests. Therefore, a hypothesis or a set of

hypotheses should pass the minimum criteria of being falsifiable. Participants in his seminar

would debate whether hypotheses in assigned readings were truisms and hard to disprove.

Consider, for instance, the following hypothesis:

Hypothesis: The longer and more stable the prior history of the business relationship

between partners in an alliance, the greater the effectiveness of the alliance.

Rajan would characterize this hypothesis as a truism or as one that is hard to disprove. In this

hypothesis, the length of prior history and stability of the relationship are closely tied to the

success of the relationship and therefore this hypothesis was true by definition. I would be remiss

if I fail to mention that he was critical of his own work. He was a believer that the peer review

process worked fine for the most part. Students doubting this would be reminded that criticisms

by reviewers were mostly directed at the (lack of) quality of the scientific effort rather than being

ad-hominem. He would frequently share reviews of his work that were soundly criticized by

reviewers and advise students to reflect objectively even after the paper is published. For

instance, he acknowledged that one of the hypothesis from his published paper had an intuitively

obvious characteristic. He also remarked (in jest) that all commissions and omissions past ten

years should be forgiven.

9

Hypothesis: The greater the cost and/or demand relationship between a particular

business in a firm’s portfolio and other businesses in the firm’s portfolio, the greater is

synergy as the source of competitive advantage of the business.

(Bharadwaj, Varadarajan and Fahey 1993)

Stating Hypotheses in Generalizable Terms. The goal of scholarly research is to produce

generalizable insights that transcend products, markets and firms. Although Rajan was a strong

proponent of this goal, he also recognized that examining or testing relationships within the

confines of an important industry (e.g., healthcare) was worthy of pursuit if the insights

generated were rich and compensated for the lack of breadth of insights. Of course, judging

whether the insights from a single industry context are rich enough or not was subjective.

Regardless of whether the data spanned multiple industries or was sourced from a single

industry, students were expected to state the hypotheses in generalizable terms rather than within

the confines of an empirical setting. For instance, the following hypothesis would be of little

value to theory building because of the nature of the proposed relationship.

Original Hypothesis: Models in cigarette advertisements for mentholated brands will

be perceived as younger than models in cigarette advertisements for regular brands.

Consider the following hypothesis about the relationship between U.S. suppliers and Japanese

distributors.

Original Hypothesis: The exercise of aggressive forms of power by U.S. suppliers

relates negatively to Japanese distributors’ perceptions of relationship quality.

To improve this hypothesis, Rajan would urge doctoral students to abstract to a higher level and

restate the hypothesis more generally. For example, if the main argument for the relationship

10

pertained to differences between individualistic and collectivistic cultures, a hypothesis such as

the following had a better chance of surviving the review process:

Restated Hypothesis: The exercise of aggressive forms of power by suppliers from

individualistic cultures relates negatively to perceptions of relationship quality by

distributors from collectivistic cultures.

Students in his seminar also learned about the uncertainties associated with research and

publishing. How many hypotheses does a paper need to have? What are the criteria for assessing

the interestingness of a hypothesis? Is lack of support for a hypothesis counterintuitive (and

interesting) or a sign of poor theorizing? How many hypotheses need to be supported for a paper

to be considered publication worthy? The takeaway for doctoral students was to be comfortable

with uncertainty in research and publishing rather than seek to have cookie-cutter strategies to

these issues.

WRITING WELL AND BEYOND

Many students are taught that they are scientists that “write up” research. Students often also

assume that doing research and writing a paper are different things (Cochrane 2005). Neither of

these are true. We are primarily story tellers and writing and doing research are closely related

activities. Rajan spends enormous time in writing manuscripts. Beyond rigor and relevance, he

taught students about Readability, the 3rd “R”. He takes great pains at writing and rewriting his

manuscripts. He believed that “it was the duty of the writer to take the reader by the hand and

walk them through the manuscript”. Doctoral students typically find writing for a scholarly

audience to be difficult. Students submitting a short write-up on a research idea or a term-paper

would receive from Rajan extensive comments and suggestions related to writing. He would

11

offer suggestions on rewording and rephrasing and in some cases, would even rewrite entire

paragraphs. As students who were left wondering, how many times should a manuscript be

rewritten, he would refer them to the following quote:

“One should never be completely satisfied with a piece of writing. My first attempt to

write this essay was made in 1987. This is the eleventh version, but such perseverance

pales in comparison to that of the accomplished writer Ernest Hemingway, who wrote

the end of A Farewell to Arms 39 times just to “get the words right”. (Flatt 1996)

The best way to prepare a scientific paper, according to him, was to pose Bradford Hill’s five

classic questions:

What did I do?

How did I do it?

Why did I do it?

What did I find?

What does it mean?

After incorporating the numerous changes, I used to often think “the paper reads so much

better now”. It was not uncommon for Rajan to deliberate and ponder whether it was more

appropriate to use the word ‘insure’ or ‘ensure’ in a particular context. He was also critical of the

use of adjectives in describing one’s research. Sentences such as the following, “I painstakingly

assembled a dataset” or “I rigorously analyzed the effect of customer satisfaction on stock

returns” would not escape Rajan’s red pen. The words “painstakingly” and “rigorously” would

be struck down by him. He would remark that a paper should have nouns and verbs but not

adjectives and adverbs. He believed “as researchers, we need to be dispassionate and indifferent

12

when communicating ideas.” Another advice for students was to strive for precision in writing.

Does each sentence say something and does it mean what it says?

Rajan has had an outstanding record of mentoring doctoral students. He is open to the

research ideas of students and does not offer his input before the student develops the idea fully.

The greatest gift he offers to doctoral students is that he instills in them the idea of independent

scholarship. It is no accident that many of Rajan’s students are self-starters. He is a conscientious

advisor who teaches his students how to craft a manuscript. Another remarkable trait of Rajan is

his professionalism when discussing problems or challenges in the discipline. In my many years

of association with him, I have not heard Rajan making disparaging comments on anyone or any

issue. He often says “For the good of the discipline, let us resolve to air our disagreements in

scholarly forums. Isn’t that why we have conferences, workshops, consortia and the like?” Even

beyond research and publishing, there is much to learn from a scholar such as Rajan. I believe

students of marketing strategy would do well for themselves, if they imbibed even 25% of

Professor Varadarajan’s perspectives on research, publishing and academia. He is one of the

finest scholars in marketing.

References

Becker, Gary S. "A Theory of the Allocation of Time." The Economic Journal 75, no. 299

(1965): 493-517

Bharadwaj, Sundar G., P. Rajan Varadarajan, and John Fahy. "Sustainable Competitive

Advantage in Service Industries: A Conceptual Model and Research Propositions." Journal of

Marketing 57, no. 4 (1993): 83-99

13

Cochrane, John H. Writing Tips for Ph. D. Students. University of Chicago, 2005.

Durant, Will. Story of philosophy. Simon and Schuster, 1961.

Flatt, Adrian E. "Words." The Iowa orthopaedic journal 16 (1996): 1.

Frenzen, Jonathan K., and Harry L. Davis. "Purchasing behavior in embedded markets." Journal

of Consumer Research 17, no. 1 (1990): 1-12.

Varadarajan, P. Rajan. "From the editor: Reflections on research and publishing." Journal of

Marketing 60, no. 4 (1996): 3.

14