Why Are Surveys SOOOOOO Boring!!!


Free twitter badge

Image via Wikipedia

Let me start by reminding people that I LOVE surveys. Sounds crazy, but it’s true. I love the amazing amount of data you can get in such a small space, the unending types of statistics you can run that reveal an unending number of ways to interpret the results. Bring on the reliability tests, factor and cluster analysis, regression, it’s all great stuff. But, many people, including myself, complain about the design of surveys. In many cases, they look horrible. Among other things, the grids make it difficult and annoying for participants to complete.
.
So WHY do we set up surveys like this anyways? There’s one huge reason – that’s the way we’ve always done it. In the good old days, surveys were written on paper. You had only so much space to ask so many questions. You used your space wisely because every additional sheet of paper cost money to print and mail out and mail back and do data entry. Every single question was money. Enter the online world. It was soooo simple to transfer your paper survey over to an online survey. Twelve years ago, online page design worked about the same way as paper design. Basic HTML made it possible to set up grids. Click and drag images, and other fancy things, were pretty much out of the question, but no one had even contemplated them at that point. Even better, if you wanted to add a question to your survey now, you just did it. You didn’t have to print another page or pay more postage or pay for more data entry. Online was the be all and end all answer. If you want to be terribly bored about comparing the quality of offline and online data from ten years ago, just read my dissertation here. I did the HTML programming of all the horribly long grids myself.😀
.
Part 2 of story. So ten years have passed. Surveys have grown to be ridiculously longer, and they still look the same as they did ten years ago. Once again, though, there is a very important reason for doing this. For some strange reason, the way you present or design a survey question has a scientifically substantiated impact on survey results. Let’s consider a grid question. If you design a grid question with the answer options going horizontally, you WILL get different results than if you ran the answers vertically. For example, if you always used to have top 2 box scores of 30%, you might get 20% or 40% using the other design. Now, change up the format of your grid question even more, perhaps bring it into 2009 and use some images, some click and drag, some easier to read fonts. Surprise, surprise, your results WILL be different than if you had run your survey using the tried and true grid format. It’s just fact.
.
How do you get around this? Easy. Well, not easy, but certainly doable. Just refer to my previous post called How to Transition a Survey. Different application, exact same concept. We simply have to take the plunge. Decide on a new method that gives the quality of data needed, run the parallel tests, and leap into transition.
.

Related Articles

 

  • Pie Charts – Our Evil Friend
  • Why market researchers can never be marketers
  • Survey Design Tip #3: Do You Encourage Straightlining?
  • Laugh at yourself and then cry at our flailing industry
  • In Honor of Infographics. #MRX This is why Twitter will die
  • Only crazy people are on twitter
  • 2 responses

    1. hi great you get superb data through surveys but yes a “conditions apply thing” in these mobile companies they make them is horrible.

    2. Maybe the solution to that are surveys that are tailored to individual respondents (when the form, content, mode of the survey are tailored to demographics, attitudes and lifestyle of the respondent)?

    %d bloggers like this: