Myths about research

A recurring question when presenting survey results is whether the sample size was sufficient and whether the sample was representative. Of course, the sample size is sufficient and the sample is representative, it cannot be otherwise for a research company! However, I add that “sample size is not everything and that it is important how the survey is set up.

When asked “And how many respondents did you have? Was it only 500? That’s not enough!” I sometimes start explaining the context of sample accuracy, sample structure weights, the methodology for selecting each respondent and their verification and inclusion in quotas, and other contexts.

And sometimes I just announce that the sample is sufficient and representativeness is not about the number, but about the composition. I prefer 500 well-selected people who correspond to the target population by age, gender, region, or education, rather than 5,000 randomly collected clicks.

A quality sample is like a well-designed dish: it’s not just the quantity that matters, but the structure.