When designing a survey, you can greatly improve the quality of the survey by checking the following things:
- Use simple English.
- Think about whether the question can be interpreted in multiple ways. When designing questions you need to make sure that there is only one interpretation of the meaning of the question otherwise respondents will become confused and your results won’t be as accurate.
- Try not to include multiple concepts within a single question. For instance, rather than ask people to rate the ‘appearance and reliability’ of public transport, ask it as two separate questions (one for appearance and one for reliability), as experiences may differ across the two concepts.
- Check to see if any of the questions are leading. It is important that you do not lead the respondent into a particular opinion by the way you have framed the question.
- Check the question order, is there a risk that earlier questions may influence the answers given?
- Never have a grid question with more than 10 statements/rows.
- Try to keep surveys to less than 15 questions.
- Do a run through of the questions and for each ask if it is ‘useful’ or ‘interesting’. If it is useful (that is, you can think of an example of how the result will be used) then keep it. If it is just interesting, and you can’t think of any way in which it might be able to be used, then delete it as you don’t need it. The only exception to this rule is with demographics; they may not be useful in their own right, but when you use them to cross-tabulate other data they can be highly useful (for instance, identifying differences in opinions or behaviours when comparing younger people to elderly people).
If you would like us to review your survey to make sure it is going to get you the most reliable findings, whilst ensuring respondent satisfaction with the survey experience, please contact us