An important issue to address in market research is the matter of bias. Bias is the tendency of data to skew toward a particular direction. It is, normal for any research to have a certain amount of bias, but it is the job of researchers to minimize bias to the best of their abilities.

Bias is normal for researches because, by definition, a sample is a small portion of a population that tries to explain the entire population but this sample may not perfectly represent the population. Aside from this, the process of research itself can introduce biases as well.

Example

A survey is conducted and the question is “Do you actually think that Brand X is better than Brand Y?” In such a question, the word actually becomes a “loaded” word that may bias or influence respondents toward saying “No” due to its subtle intimidation.

Other examples of biases:

• Phone or online interviews. Respondents may not take these interviews too seriously because of the lack of actual contact.
• Mail or email surveys. Only a particular kind of respondent may be motivated to actually mail the surveys back.
• Questions regarding income. Respondents may either not actually know what their incomes are or, post a different figure due to fear of divulging such a personal bit of information.