Here’s a quiz. Try to write a survey. No, really, I mean it.
You’ve probably taken dozens of surveys in your life. Some have been quick, four question preference surveys (do you like coffee, tea, beer or water with dinner?). Some have been more complex (How do you think Congress is doing? The president? The presidential challenger? Excellent, Good, OK, Boring, Loser, Miserable, Don’t know). But we’ve all taken them.
Surveys gauge your current knowledge and your current viewpoint. They can be used to educate you or to lead you in a direction. Good surveys are written by people who attempt to leave the outcome up to the survey-takers. Bad surveys confuse or irritate the takers. Until you’ve tried to write a survey, you cannot imagine what a work of linear thinking it is.
- Do you want to test knowledge, attitudes and practices (that’s thorough!)
- Do you want to learn what people like and dislike?
- Do you want to learn the benefits and barriers to implementing a change?
- Do you want to see if your idea(s) are good enough to start?
Write a survey.
Give it a try. And here’s a clue: a survey doesn’t give you the results you and your survey takers want unless it’s clear and makes internal sense.
Things to know about designing a survey:
Don’t presume what the right answers are (this isn’t a test!) but do allow for more than one viewpoint.
Know why you are asking the question. In a survey about political preferences, don’t throw in a question about daily exercise (yes, I hope that’s an exaggeration!). No matter how interesting the answer might be to you, it doesn’t move the survey forward.
Have a purpose for the survey. Are you testing people’s knowledge? Are you interested in finding out the survey respondents’ willingness to change? Are you trying to learn whether there’s support for an idea? Design your questions around the outcome. (This is different from having the right answer. If you want to know whether most people in Ann Arbor would eat tofu for breakfast more often than they’d eat bacon, don’t design the survey to force the outcome you prefer.)
Share your survey designs with me. They won’t be any worse that some of the stuff I read.