Common Statistical Mistakes to Look Out For in Market Research

Adding to last week’s blog, as it relates to a company’s offline or online marketing strategy, here are some more specific mistakes people make when they analyze data using statistics. You will see the following two mistakes in many marketing segmentation reports. Market researchers should look out for these when analyzing survey data.

Using means to analyze rankings: Suppose you ask your respondent how likely they are to recommend your product to someone else. You ask your respondent to rank their response on a scale of 1 to 10, where 1 equals “not at all likely to recommend” and 10 equals, “extremely likely to recommend”. In questions involving a scale, always analyze the responses using frequencies, as in the percentage of people selected who rank. A common mistake is to report the mean.

Using means to analyze open-ended responses: Say you ask your respondents…how many hours do you spend surfing the net every day? You keep this question open-ended, so people can insert any number in the response box. In the responses, some people say 5, others say 10, some say 1, and so on. In this scenario, take the mean when doing analysis, because really high or really low outliers will increase or decrease your average significantly. The median is a better statistic to use in this scenario.

Jeffrey Henning’s blog talks about some more of these common marketing strategy mistakes.