Good data begins before a form is published. One word can make a question ambiguous or leading, and option order can change what respondents select. A grammatically correct question is not enough; the intended audience should understand it consistently.
Define the decision before the question
Write down the decision the answer will inform. If the goal is to improve support speed, ask about waiting time and resolution clarity instead of using a vague satisfaction question. The more specific the intended use, the more precise the question and options become.
Remove wording that pushes an answer
Avoid praise or assumptions such as How much did you like our fast service? A neutral version is How would you rate the time it took to receive help? Offer balanced response options from negative to positive rather than weighting the scale toward approval.
Define timeframes and terms
Words such as usually and often mean different things to different people. Use a defined period: During the past 30 days, how many times...? If a term is specialised, explain it briefly without turning help text into an essay.
Test language and translation
In multilingual forms, do not translate words alone. Test whether the concept, tone and response options remain equivalent. Have a fluent speaker review the translation and pilot the form with a small sample of the intended audience before launch.
- Ask about one idea at a time.
- Use language familiar to the audience.
- Define a timeframe for behavioural questions.
- Offer Not applicable when it is a valid response.
- Randomise options when order effects are likely.
- Run a pilot before broad distribution.
Further reading
- Pew Research Center — Writing Survey Questions: pewresearch.org/writing-survey-questions
- Pew Research Center — Questionnaire Design and Translation: pewresearch.org/questionnaire-design-and-translation/