Holger Nowak, our Director Research Services, has visited the GESIS Summer School in Cologne and listened to an exciting talk. Here you find his review:
I will add the “Cool to do drugs” pencil to my list of vivid examples of a survey design fail. Over the last years, there has been a lot of talk, in the MR sector, about the declining willingness to participate in surveys and the challenge of being predictive and representative. The actual visual design of impersonal web surveys has, up to now, only been a tiny part of the discussion.
For this reason, I was delighted to get the chance to see Emily Geisen from RTI talking about this issue in Cologne. After a three week Gesis summer school on a beautiful evening it was not too crowded, which meant there were plenty of opportunities to discuss several aspects from a practical view with a mostly academic audience.
Although I have been following the discussion about survey design over the last few years, it has been a pleasure to hear Emily speak. On the one hand, more and more important aspects are emerging, but above all her presentation was very hands-on and summarized the state of research very well.
As far as I am concerned, the discussion can be condensed down to a single aspect: We need to do everything to motivate participants to do surveys and ensure they can do them at any time, under any circumstances and on what devices they choose. When designing surveys, we always have the undistracted participant in mind, the one who reads everything carefully on a large screen with a fully-equipped desktop computer. However, in fact it should work just as well for the distracted smartphone user doing a survey add-on with speaker turned off.
I agree wholeheartedly with Emily’s advice that we should design for “scanners” who take part on mobile devices. Copying personal interviews on to the web doesn’t make sense. And even though we don’t have a personal interviewer, we have so many more tools which we can use to create a perfect survey experience.
Usability tests even with a small number of participants help a lot and are, in the end, the better and cheaper option.
Keep in mind: In the end, we do it for our clients and for ourselves. From badly designed surveys that are not usable, you won’t get either valid or reliable data. Long term there will no longer be a substantial group of participants. So we come full circle: Good visual design and usability are an important part of the solution for getting predictive and representative surveys.
By the way: Emily published a book on ‘Usability testing for Survey Research’. I already put it on my wish list.