12

I'm conducting a survey, and will ask for demographic information, e.g., gender, age, level of education, frequency of computer use. Should I ask users for this information at the beginning of the survey, or at the end? Why is it better to ask at the beginning or end?

Any sources of additional information would be helpful.

Note: I will not be using any of the demographic data for screening purposes.

1

3 Answers 3

21

Place demographic questions at the end of the survey. If you place them at the beginning, you will induce a phenomenon called stereotype threat.

Stereotype threat says that if you remind someone of a stereotyped attribute of themselves, it will impact their performance even if they don't believe that stereotype. For example, suppose a girl has heard a stereotype that girls are worse at math; if you give her a test that involves math, she will perform worse on the test if you ask her for her gender ahead of time, even if she generally believes herself to be good at math.

For more information about stereotype threat, see "In the Air Between Us: Stereotypes, Identity, and Achievement" by Claude M. Steele, an essay published as Chapter 14 of the book Doing Race.

1
  • Cool. Psychology is a major cornerstone of the field and should be paid more attention to. Thanks for the answer. Commented Apr 15, 2013 at 13:05
14

This article cites some studies about it:

Some recent research in the Journal of Business and Psychology reveals that placing demographic items at the beginning of a survey increases the response rate to those items in comparison to demographic items placed at the end. And more importantly, it did not affect scores on the three noncognitive measures that came afterward

Depending on the type of questions you're asking in the rest of the survey though, putting demographics at the beginning could still possibly affect participants' answers; so your decision might depend on which is your biggest concern: making sure that they enter the demographic information, or making sure that it doesn't affect their other answers.

That being said, it would be pretty easy to split test this with your particular survey, so that might be a good option if you're unsure.

2
  • 3
    Hi Angela. Welcome to the UX SE! Interesting first post. +1 for citing research. Commented Apr 15, 2013 at 1:44
  • Great answer - trade-off between completion rate and accuracy. Commented Apr 15, 2013 at 8:58
0

I think you have to think about the type of survey and how it is conducted to start with. Generally people would regard demographic information as a barrier to completing the survey (they want it to be as short as possible for obvious reasons), so if you put it in front then it reduces a person's ability to be focused for the rest of the survey. On the other hand, putting it at the end might provide a surprise to people who was not expecting this information to be collected, and they may end up not submitting the survey as a result. It also depends on how relevant the demographics information is to the survey because people normally don't want to offer more information than necessary, especially if they are only motivated by incentives (e.g. go into entry for a prize) rather than genuine benefits by completing the survey (e.g. surveys that help improve a service).

Looking at all these variables, hopefully it will help you either make a decision or make adjustments so that your response rate is higher.

Not the answer you're looking for? Browse other questions tagged or ask your own question.