Here are our 10 tips for designing effective surveys, assessment tests and polls that will help you collect the best data possible.
1. Make each question count
If it doesn’t need to be there, remove it.
Decide on the purpose of your survey, then write the questions that match.
2. Keep it short
No one likes filling out long surveys.
The shorter it is, the higher your completion rates will be.
3. Use plain & simple language
Use clear and precise language and don’t use double-negatives
e.g. Don’t ask this: “The instructor wasn’t knowledgeable on the subject: agree or disagree“.
This is better written as “The instructor was knowledgeable on the subject: disagree or agree“.
4. Only ask one question at a time
Don’t ask questions about two things. Separate them into separate questions.
Don’t ask this: “The instructor was knowledgeable and friendly: agree or disagree”
How should someone answer this question if the instructor was knowledgeable but unfriendly?
5. Don’t use leading or biased questions
Your questions shouldn’t point an attendee to a particular answer or use emotional or descriptive language.
Don’t ask this: “Most people who attend this course enjoy it so much, they recommend it to all their colleagues. How did you enjoy the course?”
6. Avoid technical language and jargon
Don’t assume your attendee knows all the acronyms and jargon you do.
Use simple words and sentences wherever possible.
This is particularly relevant if you have attendees that don’t speak English as their first language.
7. Use Ratings wherever possible
Numerical ratings make it very easy for attendees to indicate their response, they provide much more information than just a Yes/No response, plus they’re very easy to analyze.
In addition to the Ratings, provide attendees with a comment box so they can explain their answers if they wish to.
8. Don’t over-user grids of Ratings
If there are too many numerical ratings, your attendees will grow weary of them, stop reading the questions, and just click the same rating for all questions.
9. Re-phrase yes/no questions where possible
If you change yes/no questions to Ratings, you’ll gather much richer data.
10. Test your survey
Test it out on a few friends and colleagues.
What makes sense in your head could be misinterpreted or misread in ways you didn’t contemplate by someone else.
Did you like our “10 tips for Designing Effective Surveys” article?
What did you think of our “10 tips for Designing Effective Surveys”? Let us know if you have any others!