Two Major Errors That Can Occur During Marketing Research
Marketing research can be a valuable tool for making marketing decisions about your products and your customers. Fortunately, it is within reach of most small-business owners, especially because online research can be done quickly and at nominal cost. However, small-business owners are cautioned that errors in the research process can result in bad information and ill-advised management decisions. You can mitigate research errors by becoming acquainted with the causes for error.
Your typical marketing research project starts with a concise and cogent statement of the research objectives. The next step is to design the research plan, which can be greatly facilitated with the help of a skilled marketing researcher. With the plan in hand, the next step is to "field the project," or collect the data. Then you analyze the data, put it in presentation format and report the findings. Errors can occur in each step of the research process. For instance, the wrong objective will have your team on the hunt for a chimera. However, researchers commonly divide research errors into two major classes: sampling errors and non-sampling errors.
It is impractical, if not impossible, to survey the entire population of individuals in a market. Instead, researchers collect information from a subset -- a "representative" sample -- of the total population. Sampling errors occur when your sample is not representative of the population. Sampling errors are measurable for "probability samples" where all of the relevant characteristics of the population have a mathematical probability of being included in the sample. The margin of error, a phrase frequently used in political surveys, quantifies the chances of the sample not reflecting those characteristics. Sampling errors are not measurable for "non-probability," or convenience samples. Convenience samples are made up of whoever can be cajoled or coerced into participating in the study.
Non-sampling errors is a catchall phrase that includes all other sources for potential research errors. Researchers group non-sampling errors into the following categories for convenience: coverage errors, non-response errors, measurement errors, response errors and tabulating errors. Coverage errors are similar to sampling errors, because they are caused by not having the correct respondents in the study. Convenience samples are particularly prone to coverage errors. Non-response errors are respondents who refuse to take part in the study. If you mailed out 1,000 questionnaires and received two completed questionnaires in return, you have a non-response issue. Measurement errors are caused by confusing questions that create misunderstanding and bias in the responses. Response errors are caused by respondents not being truthful in their answers or when interviewers influence responses. Tabulating errors are errors in recording, transcribing and interpreting.
Convenience surveys are quite popular with small-business owners. They are inexpensive and very convenient to execute. Recognize, however, that convenience surveys are a leading source of bad research, because of coverage errors. It's OK to do convenient research, but use your judgment in making doubly sure that your sample reflects the population universe under inquiry. If you need to learn about dentists, don't build a sample of dental assistants. Test your questionnaire to correct for confusion errors.