Best practice for survey design
A well-structured survey can assist with overall response rates and can help you receive meaningful data. Remember that a survey doesn’t have to be long to be effective. Asking only the information you need will make reviewing your data more efficient, and your respondents will be grateful for a simple and effective survey.
We recommend using the following order of questions when designing your survey:
- Introduction questions
- Outcomes measurement / Dimensions statements
- Experience questions
- Demographic questions
Starting with a few quick and easy questions such as ‘Have you engaged with our online content before?’ or ‘Are you a member of (org name)?’ is a helpful way to ease respondents into the survey.
Think about what additional information could be helpful for you to segment your data (i.e. Which performances of ours have you attended before?)
Outcomes measurement / Dimension statements
Starting with a message about what these questions relate to and why you are asking helps frame the respondent to reflect on their experience before answering the next set of questions.
An example introduction for the dimension questions: The following statements are about your experience of (program name). Please indicate how much you agree or disagree. It’s okay to leave negative feedback – it helps us improve.
We recommend using between 6-8 Dimension Statements per survey. Any more than this, and respondents may tend to get ‘trigger happy’ with the response slider and answer based on the repetitive action rather than their actual reflection.
Net-Promoter-Score (NPS) and Overall-Enjoyment are pre-built question types under Experience in the Add Content list of the survey builder – placing these after the outcomes / impact questions helps to round off a respondent’s reflection process.
Any long text feedback questions and general comments are generally best placed here at the end of the ‘reflection’ part of your survey.
Finish your survey with quick and easy to complete demographic questions (age, gender, postcode).
If it has been a longer survey, introducing this last part of the survey with a message ‘Thank you for your time. Finally, a few quick questions about yourself’ can be helpful.
‘Do you identify as any of the following’ – in this question you can ask demographic questions useful to your organisation reach such as cultural diversity, living with disability, or if they identify as LGBTQIA+. Good practice includes options for ‘Prefer not to say’ and ‘None of the above’.
- Keep your surveys relatively short to reduce any chance of survey fatigue or drop out. Using survey logic (link) is a handy way to target questions and reduce survey length.
- Consider the level of engagement your respondent has had with the activity you are evaluating to help guide your survey length. Artists/Collaborators/Participants in a workshop or forum may have a higher likelihood of engaging with more in-depth feedback surveys due to deeper level of engagement with the activity. General audience members may be more receptive to shorter, succinct surveys.