The market research process consists of six discrete stages or steps. They are as follows:
- Step 1 - Articulate the research problem and objectives
- Step 2 - Develop the overall research plan
- Step 3 – Collect the data or information
- Step 4 – Analyze the data or information
- Step 5 – Present or disseminate the findings
- Step 6 – Use the findings to make the decision
The second state of market research - Develop the overall research plan - is one of the more complex steps as it contains several distinct components. One of these components is to decide who can provide information that is most germane to the research questions and the alternate business decisions. This task includes the construction of a sampling plan that will ensure the data collected is representative of the overall target population.
The development of a sampling plan follows the selection of the research approach and the instruments that will be used to gather data. The processes that are involved in identifying and obtaining a sample are known collectively as the sampling plan. A sample unit is the group of potential research participants or respondents from which the sampling frame will be developed and from which the sample will ultimately be selected.
Once a prospective sampling plan has been developed, and the sampling frame has been established, then the market researcher is faced with figuring out how to best contact and communicate with the participant group that appears to be the best match for the research project. For a variety of practical and process reasons, it is inevitable that every member of the group originally part of the sample frame will ultimately not be included in the sample. For instance, random sampling is a process that limits which members of the sampling frame are selected for the sample. A practical reason why some members of the sampling frame will not be selected for the sample are that the potential participants must agree first of all to participate in the study, and second, that agreement must be based on their comfort with the research conditions and their granting of informed consent.
The Three Main Decisions of Establishing a Sample Plan
Sampling Unit and Sampling Frame - The first decision that a market researcher must make is to determine the sampling frame. In order to accomplish this step, the market researcher must define the target population. That is, the question must be answered: Who is to participate in the research?
A sampling frame is developed from the sampling unit, the main purpose of which is to ensure that each member of the target population has an equal chance of being sampled. However, this consideration is not a stipulation for conducting qualitative research because the equal chance parameter applies to evidence-based research in which an hypothesis is being tested. This type of research, which is invariably quantitative, is grounded in positivist scientific tradition. One example of a probabilistic sampling approach is stratified random sampling.
Sample Size - The second decision that the market researcher needs to make is related to the sample size. The question that needs to be answered is: How many people will be participants in the research? in quantitative research, the goal is to achieve a representative sample of the target population, and this can best be attained through consideration of sample size, levels of confidence, and confidence intervals. Generally speaking, the larger the sample, the more reliable the research findings will be, and typically, the more the findings can be generalized to the target population in a quantitative research project. The rule of thumb is that a sample will provide good enough reliability when just less than one percent of the target population participate in a research study. The caveat is that the sampling procedure must be credible and rigorously executed.
Sampling Procedure - There are two fundamental types of sampling procedures: Probabilistic sampling and non-probabilistic sampling. For quantitative research, a probability sample is drawn from the target population in order to calculate certain statistical considerations. The probability sample reveals the confidence levels or confidence limits related to sampling error. Probabilistic sampling can be a drain on resources (including time, money, and expertise), but it does allow a market researcher to measure sampling error. Non-probability sampling procedures can provide very useful data and, ultimately, market insights. As with qualitative research and quantitative research, there are proponents and critics of probability and non-probability sampling procedures.
Different methods and techniques will produce different forms of data, and it is critical to ensure that the sampling procedures match the type of data that is needed to answer the research questions.
Before moving on to step three of research planning, it is important to consider the part of the sampling plan that deals with contacting those individuals selected for the research frame in order to elicit their participation in the research.
Kotler, P. (2003). Marketing Management (11th ed.). Upper Saddle River, NJ: Pearson Education, Inc., Prentice Hall.
Lehmann, D. R. Gupta, S., and Seckel, J. (1997). Market Research. Reading, MA: Addison-Wesley.
Keep up with the site by signing up for my free newsletter!
"Like" Market Research on Facebook
Follow me on Twitter.