As you sit down to begin building out your survey, there are a few best practices you should keep in mind. Follow these tips as you take your initial list of questions and mold them into your survey creation form. The central theme here is ensuring you are presenting your survey to your respondent in the clearest, most concise way possible.
1. Focus on one research goal
Reiterating a point from our last lesson, each survey you create should focus on one research goal. Everything you ask should be related to acquiring knowledge related to that one topic. Avoid unrelated ‘nice to have’ questions. If you are inquiring about consumers preferences related to an iPhone case, don’t lob in a question about a charging stand concept.
2. Ask straightforward questions
Vague questions lead to vague responses. Write in plain English and be pointed about what information you want to collect. Avoid things like double negatives and difficult concepts. If respondents clearly understand your questions, you are more likely to clearly understand their responses.
3. Ask one question at time
Asking for feedback on two variables in one question is a recipe for bad data. Example: Which brand of running shoes provides the best comfort AND support? In this example, a respondent may feel one brand has great comfort but mediocre support.
4. Keep it short
Time magazine claims a human has a shorter attention span than a goldfish. Keep your surveys short and your respondents engaged. The longer the survey, the more likely respondents are to speed and provide less than thoughtful responses.
5. Use closed-ended questions
Respondents have an easier time comprehending questions that have a stated response. While open text (free response) questions are necessary at times, they make it harder to quantify the results. They can also wear a respondent out. Try to limit your survey to two open text questions max.
6. Avoid the Yes bias
Decades of research tells us people are uncomfortable telling you no. In the context of a survey, this is referred to as the acquiescence response bias. Presenting too many yes/no questions can severely compromise your data. If you want to know if someone is likely to purchase a new suitcase in the next 12 months, use a “Probably Not / Definitely Will” 1-5 scale rather than a binary yes/no format. A scaled question also has the added benefit of conveying the intensity of your respondent’s opinion, providing richer data.
7. Use Neutral Language
An online survey is constructed to assess what other people think, not to confirm your preconceived notions. Abstain from language that is likely to persuade your respondent in any way.
8. Provide a “Not Applicable” response
Certain questions may not apply to a respondent. Provide an “N/A” option to avoid forcing a respondent to falsely report on a question that does not apply to them.
9. Include images and video
Visuals can stimulate your respondent’s memory. Use a free online image tool like Canva to get your images sized properly. Canva also lets you easily layer in text overlays and graphics.
10. Have fresh eyes review your survey
Ask a friend to review your survey before you submit it for responses. A fresh set of eyes can help you catch any errors.
If you are interested in putting your new research skills to work, you can head over to Centiment to start creating a survey.