Scientific Method in MR

Marketing is more an art than a science. Successful decisions making in marketing depends on the skill and judgment of the individuals involved and cannot be reduced to an organized body of principles. Most decisions are made quickly and are based on experience, but sometimes new information is collected to help.

When marketing managers make decisions, they normally use all information that they have accumulated as part of their general knowledge or that comes to them as part of the normal flow of day-to-day operating data. If a particular decision is important or if a manager feels more certain than usual in making the decision, he may decide to gather additional data pertinent to the problem. Any such data gathering is ‘research’ but that term is usually used only when the data collection is organized in a formal manner. It is the function of marketing research to reduce they are applying the methods of science to the art marketing.

No method known to man can entirely laminate uncertainty. But scientific method, more than any other procedure, can minimize those elements of uncertainty which result from lack of information. By so doing, it reduces the danger of making a wrong choice between alternative courses of action.

The term research might be substituted for scientific method in this quotation without changing the general meaning. No research is perfect, but good research is better than research because it is done more scientifically. Accordingly, it is worthwhile to consider what the scientific method is and makes some research more scientific than another research.

What is Scientific Method?

>I don’t know what explains this. I wish to hell I did.
>Burs Roper pondering his own Roper Poll on Election Day, had rechecked his methodology and found it sound. So why was the Roper Poll predicting only a 10 point Reagan victory while other polls were predicting a gap as wide as 25 points?
>I’m very concerned, he conceded. This raises real questions of whether this business is anywhere near a science.

The fact that various election polling services predicted quite different results indicated that the process of polling was not scientific. Many of those polls are conducted by experienced, capable pollsters who worked with care yet their results were different. All though they were using the most scientific methodology available, but truly scientific methods should produce the same results even used by different in individuals. Marketing research relies on many of the same techniques as election polling and has some of the same problems is being scientific. Why this is so and what can be done to make marketing research more scientific are the basis for the following discussion.

Two general traits characterize the scientific method: validity and reliability. Validity is the characteristic used to describe research that measures what it claims to measure. For example to measure television viewing audiences, mechanical devices are put on sets to determine when they are turned on. This method is used to measure audiences, but in fact, it measures only sets turned on. On the surface, validity often seems so obvious as to be unimportant but in practice subtle variations such as that in the TV viewing example above make it necessary for researchers to consider validity carefully in each project.

Reliability is the characteristic as research methodology that allows it to be repeated again and again by any researcher always with the same results. Scientists working in their laboratories carefully control all aspects of their experiments and report them in detail so that other scientists can attempt the same study and confirm the results. Yet even in the highest realms of science, reliability may be hard to achieve.

Most marketing research is done as proprietary research by individual firms that have no intention of publishing their results for others to see let alone to verify. If research results are to be useful, researchers must use methodology that is reliable that is methodology that they or others can reproduce with the same results.