The science behind surveys can get seriously complex. “Since the survey usually measures several attributes for each population member, it is impossible to find an allocation that is simultaneously optimal for each of those variables. Therefore, in such a case, we use the popular method of allocation which use the same sampling fraction in each stratum.” That was part of a paragraph from a research paper on surveys. We took out the accompanying equations. It’s not rocket science but it can become daunting. Obviously, we can’t offer a thorough explanation of surveys, but we can simplify things a bit, especially when it comes to marketing. It starts with sample size.
Size Matters
We’ve all seen political polling numbers that show a candidate up by say 4% and we all know there is the statistical margin of error, that plus or minus figure. We’re not going to offer an explanation, let’s just agree to the fact that the smaller the margin of error, the more accurate the survey. And that leads us to the sample size. Let’s say your particular marketplace has a population of 50,000 consumers and you want to learn their opinions, judge their awareness of your brand, predict the outcome of a promotion. You don’t have to ask all 50,000. 397 is the sample you need to get a ±5% margin. To get that down to 1% you would need to sample more than 8,000. It’s not only the quantity of the sample, but the quality is also important. A random, representative sample of the original 50,000 is key. See how easy it is to get in the weeds?
Empirical vs Anecdotal
Empirical evidence is information researchers or scientists have generated. In marketing it would be information gathered by authoritative surveys, focus groups, and actual results. Anecdotal on the other hand is based on an individual’s experience. For example, empirical evidence of Big Foot would be clear, crisp, close up photographs and videos. Hair and skin samples. Anecdotal evidence would be a story about spotting the creature. That’s not to say anecdotal evidence has no value. Show five people the headline you want to use in a print ad and if four don’t understand it, or don’t like it, the evidence, anecdotal or not, is clear.
If you have questions about marketing surveys, testing, focus groups or anything else in this arena, give us a call. We give you accurate answers. ±0%.