By Salome Pachkoria, Anna Phillips and Jasmine Wang
How many of you have either created a survey or filled one out? Everyone? That’s what we thought.
This is why today, we will be talking about survey design! (we’ll just pretend you didn’t notice the image at the beginning of this post)
Survey design is the process of creating surveys with the goal of receiving maximum insights from survey research. In this process, it’s important to keep in mind the famous 7 Ps applied to survey design: Prior preparation and planning prevents piss-poor performance.
Here are 10 tips for designing a survey:
- Design the questions according to your plan for using the data. If you’re planning to make graphs, charts, or infographics with the information for the public to see and absorb, then closed questions are best. If you want to learn more about people’s opinions, experiences, and beliefs, then open-ended survey questions will be more appropriate.
- Questions should be as clear and specific as possible so that respondents will interpret their meaning in a consistent manner. For example, if you ask a broad question such as “do you have the means to live comfortably”, each respondent may have a different interpretation of what “living comfortably” means.
- Construct questions that people will be willing to answer. If you don’t, you might end up with questions that are too personal for respondents to feel comfortable answering.
- Formulate questions that encourage respondents to answer truthfully. Sometimes, respondents feel compelled to answer the way that is considered socially acceptable, such as saying they own a car even if that isn’t the case. You might instead ask how they travel from one place to the other, which would allow them to answer more truthfully – perhaps they take the bus.
- Make sure to ask questions about something people are likely to have the knowledge or information to answer. If, for example, you ask how many times they’ve had a headache in the last three months, they may not know.
- Avoid asking double-barrelled questions – a.k.a. questions with two different parts. For example, do you believe that businesses must reopen as soon as possible and that the U.S should continue blocking flights with Europe and Asia? This should be converted into two separate questions, as a respondent may have different answers for each part.
- Pretest your question. Make sure that it follows the tips we have been talking about before starting surveying actual respondents. Remember the 7 Ps? Consider this a special addition to them from the META Lab!
- Avoid asking loaded, assumptive, or leading questions. For example, if your question is “What is the biggest problem you have with your boss?”, it implies that the respondent has had many problems with their boss, even if the respondent actually has an excellent relationship with their boss. An alternate question might be “Tell me about your relationship with your boss.”
- Keep the language accessible – you don’t want the question to confuse the survey respondents. If you want to know, for example, how your respondents feel about higher education, it’s better to ask “Do you think higher education is important” – not “Do you prescribe to the notion that higher education is essential to the betterment of oneself”.
- Now that you have designed the best questions for your survey, think about the order of your questions. Try to start with more general questions, and go into specific ones later.
Luckily for us, we can now conduct survey design easily through some online survey design tools. Most of them are very user-friendly and easy to operate. Here we have listed some of the most popular ones:
- SoGoSurvey https://www.sogosurvey.com/
- Survey Monkey https://www.surveymonkey.com/
- Typeform https://www.typeform.com/surveys/
- Google Forms https://www.google.com/forms/about/
- Client Heartbeat https://www.google.com/forms/about/
- Zoho Survey https://www.zoho.com/survey/
- Survey Gizmo https://www.surveygizmo.com/
- Survey Planet https://surveyplanet.com/