In 1962, the year of publication of our previous book on multivariate procedures, John Tukey published a long and marvelous discussion on "the Future of Data Analysis." His gift to us was our professional identity. We knew we were not statisticians. That was too painfully obvious. We were young educated researchers who had been thoroughly persuaded by Professor Phillip J. Rulon that multivariate quantitative procedures had to be the keys that would unlock the secrets of human social behaviors. Rulon transmitted primarily the vision of Truman L. Kelley, although other pioneers in behavioral research, including Cyril Burt, Louis L. Thurstone, and Raymond B. Cattell, had similar visions. The multivariate heritage we acquired was the work of great statisticians, such as Pearson, Fisher, Hotelling, Bartlett, and Wilks. Still, we were not statisticians. Tukey told us we were data analysts, and we have since learned to accept and even enjoy the identification. Tukey argued that there have to be people in the various sciences who concentrate much of their attention on methods of analyzing data and of interpreting the results of statistical analysis. These have to be people who are more interested in the sciences than in mathematics, who are temperamentally able to "seek for scope and usefulness rather than security," and who are "willing to err moderately often in order that adequate evidence shall more often suggest the right answer." From the Preface to Multivariate Data Analysis, by William W. Cooley and Paul R. Lohnes, 1971.

In a November, 1968 Journal of Marketing Research editorial, Dr. Paul E. Green described what he was calling the "Research Generalist" as follows

He will be interested in operational significance as much as statistical significance, "quick and dirty" tests as well as precise experimental designs, relevant criterion functions rather than elaborate analytical procedures used to optimize some naïve figure of merit. On the other hand, he will be interested in structure as well as content, generalization as well as specific problem resolution, future implications of actions as well as what may happen tomorrow. As such, his orientation will reflect at least some of the makeup of the scientist. 1

Over the years, we have called them research analysts, statisticians, and various other names, mentionable and unmentionable, never quite feeling we really have adequately described those individuals who toss around multivariate statistics, principles of buyer behavior, sample size rules, measurement mandates, and the weird language of computer statistical packages with equal ease.

They come from no particular field of study - some come out of psychology, others from sociology and marketing. Some really do have degrees in statistics, but others may come from advanced training in agriculture, architecture, or accounting, or even, heaven forbid, physics, chemistry or engineering.

Just who are these people and why do we value them so highly?

More and more, we, as a profession, call these specialists Marketing Research Methodologists.

This article advocates the recognition of the marketing research methodologist as a specialty and specific job title in the marketing research profession. I attempt to define the title and the qualifications for that position. I also attempt to give the reader a sense of the current state of the various roles that marketing research methodologists are playing in our profession and argue for the continued expansion of that role. I would suggest that a further refinement of both the definition and the qualifications of the research methodologist is a worthy subject for additional discussion and study.²

A Definition

We may define the marketing research methodologist as an individual who has a balanced and in-depth knowledge of the fields of statistics, psychometrics, marketing, and buyer behavior and applies that knowledge to describe and infer causal relationships from marketing data.

As the definition states, the marketing research methodologist requires specific education in, and knowledge of, a variety of converging disciplines. Typically, today's marketing research methodologist has specialized in one of the fields mentioned at the masters or doctorate level, then acquires considerable knowledge in the related fields through additional course work, targeted in-depth readings, and ongoing communication and interaction with peer-level specialists.

However, let us recognize that there are many very successful marketing research methodologists who have come from far different backgrounds and have taken a different approach to finding those inferred casual relationships. There are well known methodologists who have come out of the engineering and hard science disciplines and others who have come out of political science, sociology, agricultural science, geology, and anthropology.

We may conclude that there is some other key characteristic that sets apart the marketing research methodologist, regardless of whether formally trained or self-trained. I would suggest that this key characteristic is a passion for trying to understand why buyers (consumers and organizations) do what they do. And, I place this passion above that of the hard scientist and the engineer, who try to determine how and why physical relationships work, or don't work. Understanding and modeling human and organizational buyers is a far more complex endeavor, requiring a wider range of technical and methodological skills, moderated with a good dose of logic and art.

Training the Research Methodologist

If we were to design an ideal post graduate program of study for the well-rounded research methodologist, what would it look like?

In my opinion, it would include the following course work:

• Probability and Statistics, (including a heavy dose of non-parametric statistics)
• Multivariate Statistics and Statistical Modeling
• Human Psychology
• Organizational Psychology
• Psychometric Measurement
• Marketing Research, Basic and Advanced
• Sociology
• Buyer Behavior
• Micro Economics
• Marketing Management
• Business Communications

Most of these courses should also be heavily computer-driven using one of the major statistical packages, such as SAS or SPSS. By the end of the program of study the candidate should be an expert in the chosen statistical package.

But, as most of us know, course work alone, no matter how well taught, does not a methodologist make. Regardless of the rigor of the program, there is no substitute for experience. I would suggest that a minimum of five years of varied on-the-job experience is necessary to make a well-trained research methodologist qualified to lead a department or research practice. During this experiential period, the marketing research methodologist should become intimately familiar with research and statistical applications in market segmentation, product/brand positioning, product/service optimization, pricing, brand evaluation, customer satisfaction/loyalty measurement and modeling, concept evaluation, simulated test market modeling, and communications/media research.

In addition, given the speed at which new methods and applications are being developed and proliferated, especially by the academic community, the research methodologist should be provided the opportunity to interact with peers and academe on a regular basis, simply to stay current. One of the great laments of many research methodologists is the isolation from peers and colleagues and the rarity of opportunities to discuss new approaches and new techniques. On-going investments in professional development and training are essential.

Successful marketing research methodologists also need to be excellent communicators. They are often faced with the challenge of explaining complex research methodologies, advanced statistical analyses, and esoteric research design considerations to business managers and research users who are not trained in the field. The methodologist's ability to do this well builds trust and confidence with clients and colleagues and enhances the profession of marketing research.

The Reality

In fact, very few recognized marketing research methodologists have taken the breadth of course work previously suggested. Typically, they have bridged this gap in formal training in four ways - longer experience, independent study and reading, professional training courses, and heavier involvement in professional development activities.

At this time, those in the profession that are generally recognized as "senior marketing research methodologists" have well over 10 years of practical, applied experience. Over 20 years of experience is not unusual. During that period they have typically been employed by three or more firms. An exception to the employment switching are those methodologists who are employed by firms with large marketing science groups or departments, in which case they tend to stay with one firm much longer. Firms with larger departments tend to offer a higher level of peer interaction, on-site learning opportunities, and advancement opportunities.

Most senior marketing research methodologists are also avid readers and engage in spurts of self-directed study in a particular field. Additionally, these practicing methodologists subscribe to and read a wide array of primarily academic journals and periodicals - JMR, JCR, JASA, JAMS, Marketing Letters, and so forth. As part of their self-directed study, practicing methodologists tend to follow the writings of a select set of well-know academic researchers to keep abreast of the latest developments in a given field. These senior methodologists are also famous for the extensive number of books on research methods and applications that they accumulate for both reference and detailed reading.

The third avenue for professional development is the myriad technical training courses offered by professional associations, major research-oriented universities, and a few for-profit organizations. Though these programs tend to be quite expensive, they often teach the latest developments in emerging techniques and research methods and are well-attended by practicing methodologists.

Finally, the professional marketing research methodologist is likely to attend some of the very few professional development and academic conferences directed specifically to their field. Notable among these are AMA's Advanced Research Techniques (ART) Forum, the annual American Psychological Association Conference, a few of the targeted conferences of the Advertising Research Foundation, the annual Institute for Operations Research and Management Sciences (INFORMS) conference, and AMA's annual Frontiers in Services Marketing Conference.

The Current Situation

Currently, the demand for qualified marketing research methodologists far exceeds the supply. The demand for advanced analytical techniques has been accelerating for the last several years as both research companies and corporate research departments become more familiar with the power and utility of these newer methodologies. Furthermore, corporate research departments are being re-built and re-centralized following the down-sizing frenzy of the early 1990's. Consequently, salaries for methodologists that are even marginally qualified have soared.

Given the current long lead times necessary to train a qualified marketing research methodologist, it appears that this shortage will exist for the foreseeable future. This sharp dissonance between supply and demand is fostering two major negative effects on the profession of marketing research.

1. With the advent of readily available and easy-to-use statistical software packages, many unqualified research analysts are executing and delivering advanced analyses that are clearly erroneous. Unfortunately, recipients of such work cannot easily assess the correctness of the delivered results, but often know they fail the test of face validity and logic. The result is that the user loses faith in the advanced methodologies and, ultimately, in the marketing research profession.

2. A second outcome is that independent research organizations and consultants often fail to recommend or use these advanced methodologies due to either unfamiliarity or limited access to qualified specialists, even though the advanced methods may provide a much higher level of utility and accuracy to the end user. In this case the user often gets results that are not operationally actionable or fail to provide the expected level of insight. Again, the result is that the user loses faith in the marketing research results and, ultimately, in the marketing research profession.

Supporting the Marketing Research Methodologist

It seems obvious that the marketing research profession and the associations that represent that profession should work toward remedying the situation just described. I would recommend several avenues of endeavor.

1. The academic community should seriously consider crafting courses of study at the masters and doctorate level to increase the supply of well-trained methodologists for the practicing marketing research community. Currently, almost all doctorate-level training in this field is oriented toward a career in academia. There is little reason not to consider training practitioners at this level much like the schools of medicine, engineering, and education. Yet, I am not aware of any school of business that specifically trains advanced research methodologists for a career in the practice of marketing research.

2. The academic community can help in another way - in the dissemination of new research methods. Currently, dissemination is assumed by publication in a single academically-oriented journal. However, I would suggest that this is insufficient. A typical methodological article reveals the technique, but seldom applies that technique to a variety of situations, leaving the practitioner little choice but to undertake considerable experimentation and simulation on their own, time permitting, or waiting to see if others in the profession are applying the method to real problems. Additionally, even though the articles are refereed, it seems to take years for the rest of the academic community to make a judgment on whether the method is sufficiently sound and robust for practical application. This is a haphazard process at best.

   It would seem that the dissemination issue could be easily addressed by a journal that concentrated on research methods alone, and would actively solicit comment and tests of these methods in a variety of situations by both other academics and practitioners - a Journal of Marketing Research Methods.

3. The professional associations should increase their training and education opportunities on two levels. First, research users and sellers of research services need to be better informed on using the newer research methodologies and technologies that are available and under development. These programs need to address the proper applications and clearly discuss assumptions and limitations. Second, ongoing training opportunities for emerging methodologists, addressing the proper application and use of advanced research methods and techniques, need to be expanded and targeted specifically toward the practitioner. AMA's Advanced Research Techniques Forum, AMA's Applied Research Methods conferences, and Sawtooth Software's revived annual conference are three very good models. We simply need more of those, plus additional programs targeted to in-depth understanding of more narrowly defined subject areas. It is my observation that the aforementioned conferences have become too diverse to provide adequate, in-depth exploration of specific tools and techniques needed by the research methodologists. An exception is AMA's tutorial series at the ART Forum where specific methodologies are given in-depth treatment.

4. Research firms and corporate research departments need to increase their support of training and professional development for current and emerging research methodologists. They must recognize that peer interaction and communications between practicing methodologists and academic researchers are essential for the veracity and health of the marketing research function and the profession. Furthermore, they must recognize that the costs of mis-application in advanced research methods far outweigh the cost of those training investments.

5. The senior marketing research methodologists themselves must invest some of their time and energy in freely sharing their own research developments, experiences, and findings with colleagues through conference participation, non-promotional writing for the trade press, and peer-to-peer communications. They need to actively seek out and welcome peer review of their work and learn from those interactions.

It's time we recognized that the increasing complexity of marketing research has spawned increased specialization in our profession. The specialty of the marketing research methodologist will assume an ever increasing role in the business of marketing research and in the marketing research profession. If we are to continue to grow as a business and a profession, we should recognize this specialty and nurture its growth.

END NOTES:

1. Green, Paul E., Journal of Marketing Research, Vol. V (November 1968), p 442

2. I wish to thank Dr. Paul E. Green for reviewing this article and providing many insightful comments and suggestions.