Writing a survey for the web is hard work. I’m always looking for errors because one good one could send the whole project down the tubes.
These days on Monday and Wednesday (my designated “survey days”), you’ll often find me in the University of Washington library scribbling notes in my thesis project yellow composition notebook while reading books with titles like: “Mail and Internet Surveys: The Tailored Design Method,” and “Conducting Research Surveys Via E-mail and the Web.” They provide me with advice along the lines of: “research indicates that answers that appear at the top of a scroll box are more likely to be answered positively.” Of course we don’t want that; it would introduce bias. So I won’t do that…heck, I eliminated those dreaded scroll boxes entirely.
If you write for a living or as part of your work, you know how it is. Every time you look at your current project, it needs a little tweak here or there…this is only compounded with a survey. You don’t want to overdo change, but you do want a really streamlined, interesting document. A survey that people will enjoy responding to (I may be pushing it with that, but I do believe it is possible).
To edit your survey, you need to look at it from far away and close up, and generally you can’t wrap your mind around both processes on the same day. So you officially become a geek. You ask your friends to take your survey and let you know if certain questions make sense. Would you find this survey unworkable or offensive in any way if you were a Pacific Islander American mother of two, in the process of divorcing, and living with your parents? Can you take my survey again and again, with little thought to the answers you are providing, just to make sure the program does not crash.
That said, if you are here, it is because you have already taken my survey. Please do not take it again. Once is fine and I hope it worked well for you.
Up next, some thoughts on the differences between online surveys versus the more traditional approaches (ie, mail- or telephone-based).
