Observing these simple quidelines will help ensure that your respondents understand your survey requirements and complete it as accurately and fully as possible.
Do define your objective
- Before you write the survey, ask yourself why you are conducting the research. Try to narrow your objective down to one main point. Write one line explaining this objective and use it in your intro. In your intro you should also provide simple information about how and where your research will be used and presented. As you write the survey, be sure that each question relates to your objective.
Do know your target
You may not be able to control exactly who takes your online survey, but if your objective is defined you should have a target group in mind. For best results, notify your target respondents ahead of time about the upcoming survey.
Do give your survey a good title
Use strong keywords and simple language. A good title will attract respondents with strong ties to your subject.
Do provide simple instructions
In one or two lines, respondents should be able to know exactly how to complete your survey. Include a cut-off date.
Do write objective questions
Make sure you remain neutral as you write the questions.
Do use simple language
Use the simplest wording possible for the instructions, questions, statements and choices in your survey.
Do use mainly closed-ended questions
- Closed-ended questions provide options for respondents to choose from. People are more likely to complete a survey in full if it contains mainly closed-ended questions. These questions require less time to complete. They also make it easier for you to analyse and present your findings.
- Open-ended questions are useful for gathering information about something that you have very little background on. If you're looking for good ideas rather than statistical data, use more open-ended questions.
- You can always provide an extra text box or make your contact information available for those who want to share more information with you.
Do be consistent
- Numbers should be in ranges.
- Verbs should be parallel in form.
- Punctuation and capitalization should be uniform.
- All choices in one question should be similar in length.
Do consider how long respondents should spend on your survey
- How much time would you be willing to spend filling out a survey questionnaire for a colleague, friend or researcher? Consider this when you decide how many questions to include.
Do think about the order of your questions.
- Group all questions on similar topics.
- Put the most interesting questions first.
Do include appropriate ranges and rankings in a logical order
- Strongly Disagree, Disagree, Agree, Strongly Agree
- Poor, Satisfactory, Good, Excellent
- Never, Rarely, Sometimes, Often, Always
- Less then once a week, Once a week, Several times a week, Once a day
Do include enough choices
Respondents should always feel that at least one choice is applicable.
Do consider the look and feel of your survey
- The instructions should be easy to find.
- The basic font should be large enough to read.
- There should be a nice balance of text and white space.
- Ads or other distractions should be minimal.
Do test your survey before you publish
Find a few people who are willing to take your survey. Ask for feedback.
- Do have someone proofread your survey.
Typos will turn off your respondents.
Survey Don'ts →