Top Ten Survey Writing Tips

top_10_pedistal_1600_clr_6486

You’ve been tasked with writing a survey. The only thing is…you’ve never written a survey. While writing a survey isn’t too hard to do, here are ten best practice tips you should know before you start.

Survey Writing Best Practices

  1. The first rule of survey writing is to have a clear purpose for the survey. A survey without a purpose is like a movie without a plot. You can still write it, but it will leave your respondents wondering why they spent their time on it, or worse, they’ll drop out early and you won’t get many completes.
  2. Make sure the questions match the purpose. See above analogy for why you want to be sure the questions match the purpose of the survey.
  3. Consider demographic questions you’ll need to include for analysis. For example, do you think respondents of different age groups might answer differently? Do you think respondents in different professions might answer differently? Include the appropriate demographic questions.
  4. Make sure your survey flows naturally. A basic layout: ask any filtering questions first (questions that might determine whether respondents meet the requirements you may need, such as being familiar with the topic of your survey). Next comes the main body of your survey, or the questions that you’re using to gather the information desired. Close with the demographic questions; if you need demographic questions to filter your respondents at the beginning, save the rest of the demographic questions for the end.
  5. Have enough answer options to cover the possible answers your respondents might have for each of your questions. Sometimes, using the answer option “other” or “none of the above” is enough to cover this need. I have taken surveys where neither “other” nor “none of the above” were an option, though, and I had to quit the survey because there was no way I could continue answering it without simply lying (how exactly you’re supposed to answer questions about what your pet eats when you have no pets is still beyond me).
  6. Don’t ask your respondents questions that don’t pertain to them. Using logic options such as branching, skipping questions, and extracting answers to focus the survey are all fantastic ways to make sure you’re getting the answers you need from those taking your survey.
  7. Don’t ask double-barreled questions. What is a double-barreled question? Here’s an example: “Please rate your level of satisfaction with the speed and clarity of communication from the sales team.” For the respondent, how are they supposed to answer if the speed was fine, but the clarity was not (or vice-versa)?
  8. Make sure rating scales make sense. I’ve taken more than one survey where the rating scale made no sense to me. In one case, I was asked to rate satisfaction, but the scale was for agreement. Another survey included a 1-5 scale but forgot to indicate which was positive or negative.
  9. Limit the numbers of different types of scales you include in your survey, especially one right after the other. If you need to ask multiple sets of questions measuring various aspects, be very clear in your question text and in the answer options about what you’re asking. I’m guilty of skimming text and then being confused when the scale changed between questions.
  10. Test your survey rigorously. Get lots of people to take it before you go live with it. If you’ve used logic in your survey, ask some people to specifically test the logic. Ask for feedback from your testers, including: did it make sense; did it flow well; did the questions all relate to the main purpose; did any questions seem odd; did the logic flow correctly; how long did it take; etc.

You’re ready to start!

While there is still plenty to learn about things like how to avoid bias in a survey question, you’re ready to go with the basics! If you only remember two things from this, I hope they are to clarify your survey’s purpose and test the survey rigorously. No matter how seasoned you are as a survey writer, you can find yourself forgetting some of these survey writing best practices. Asking others to test your survey before you go live can reveal things you’ve forgotten, and it’s always a best practice to catch errors before you go live with anything, including a survey.

Happy survey writing! Let us know in the comments below if there are other basic survey writing tips or survey writing best practices you’ve learned, too!