Research MethodologyBy Tom Dougherty
16 July 2014
Marketing Research Methodology
The marketing research methodology you use to collect data is as important as the questions you ask. Because market research must be projectable or you to take the results of quantitative research at face value. That means your sample must be mathematically projectable to the whole audience within an acceptable margin of error.
That kind of research forms the basis for the most effective brand strategy and should be used as a benchmark research study to measure your effectiveness over time. Plus it is important to pay attention to your marketing research methodology.
Marketing research methodology is important. How your market research is conducted is crucial. Because we try to improve research, we all continue to look for more cost-effective ways to conduct projectable research, there is simply no acceptable substitute for the traditional telephone (and cell phone) survey method.
Online studies, expert panels and studies conducted at the urge of an email have some uses, but they also have limitations.
So at Resultant Research, we insist that the findings the strategists at Stealing Share base their brand strategy on are projectible.
Types of Marketing Research Methodology
To accomplish projectable research, Resultant conducts double-blinded quantitative studies, which means that neither the respondent nor the questioner
knows who is sponsoring the study that might otherwise color the outcome.
This also means the lists of possible respondents is generated and accessed randomly utilizing a RDD sample methodology. This ensures limits bias in the responses and allows Resultant Research to measure incidence. Incidence, in simple terms, is a record of the percentage of respondents that finished the study.
For example, if 90 out of 100 respondents that were randomly contacted finished the study you would say you had an incidence of 90%. You know that if the incidence were too low, the confidence of the study would be called into question.
The issue of bias in marketing research methodology
So it is possible that the study creates a bias and rather than being random, the market study itself may be self-selecting. Another way to increase the incidence is to pay an incentive for finishing a study. Incentives increase incidence and allow the researcher to control randomness.
As a general rule, we prefer to use CATI (Computer Assisted Telephone Interviewing) systems, which automate the activities of the interviewing facility as our preferred marketing research methodology.
This allows the system to control the questionnaire, ensures random calling, rotates the questions to account for fatigue, and collects the data to ensure projectable results.
Many research companies and brands make a litany of mistakes that only ensure the data will not be projectable or valuable in developing a brand.
To avoid higher costs, many avoid random sampling, choosing online, email and mail surveys that can be self-selecting. And therefore, are not as projectable to the larger target audience.
Online surveys, for example, are self-selecting because you only get results from those who would be most likely to fill out a questionnaire from an email request and go online to fill it out. You only get the extremes when most of us think in the middle.
Contact us at email@example.com
Remember that great brand research asks more than a traditional usage and attitude study does. And supplemental research, like focus groups and name testing, is misleading and a waste of money.Click here for more information.
Click for the Marketing Research Web Story