When to Collect Demographic Data

Over the past few months we've had several articles about the perils of demographics data and why it is not necessarily useful to collect. Still, there is no denying that researchers like demographics data – especially financial demographics data – because it gives them an idea of what their customers can spend, who they are, and how that might affect business decisions. Yet of course, we've also talked about how unless you collect a large enough sample within each of your demographics, you may not be able to make judgments about what the demographics data means.

There is also the potential cost to your survey response rate, since people drop out if a survey is too long, and demographics make a survey longer. Still, people like demographics data, so here are some thoughts on the options for collecting it.

Within the Survey The easiest way is to include it in the survey, but this is what causes most of the problems we've discussed. Still, if you place them in the beginning or end of your survey, with a "before you begin" or "you're almost done!" tag, you might be able to get around the problems with survey length. When You Collect Contact Data Many companies get their customer samples through signup sheets.

Here is perhaps the best time to get the demographics data, because you can include all of the demographics questions in the beginning. Then find a way to link those demographics to the survey response data, so that you are able to know exactly what the demographics of your sample are without asking them again. From the Panel Company The only other way to gather demographics data is to get it directly from the company you use (if you use a company) that gave you a panel.

Obviously this is not ideal for most research because it involves paying another company to collect the same for you, and chances are they are going to perform one fo the above two techniques. Your final option to consider is to simply poll for demographics once you have a specific question that involves demographics.

For example, if your research has brought you to a conclusion, you can ask your sample one question (do you support X idea), and then ask them demographics questions, and end the survey. That will keep the survey short, and will allow you to collect demographics data only when it is necessary.