Questionnaire deign is both an art and skill requiring much work and time. Poorly designed survey questions will result in “rubbish in” and “rubbish out.”
- Ensure your questions are specific. How would you rate your general practice? This is very unclear what the question is asking. Is it about accessing your GP services, ease of making appointments, waiting times? Be specific in what you are asking. How would you rate the time you have to wait for an appointment to see a doctor?
- Avoid loaded and leading words. Do you agree the NHS is severely underfunded? Should smoking in your car be prohibited? Biased or strong words can make a lot of difference in how informants respond to a question.
- Confusing or unfamiliar words. At what time do you normally eat dinner? Understanding of the word dinner can vary across different parts of the country. Dinner can be understood as either the meal at lunchtime or the meal eaten later in the evening.
- Avoid double-barreled questions. How would you rate the friendliness of your doctor and practice nurse? The question should be split into two.
- Ensure response options are mutually exclusive. Responses options should not overlap making the choice difficult for the respondent. E.g. Age: 15-20, 21-26, 27-32 years.
- Avoid long questions. Questions should not have more than around 20 words.
- Balanced and unbalanced scales. Unbalanced scales (where the majority of the the response options are either positive or negative) are suitable in some cases where for example, respondents are more likely to have a particular view. However, balanced scales (where equal and negative responses are provided) should be used in general.
- Misplaced questions. Whenever possible group questions together when they relate to a specific topic e.g smoking behaviour.
- General questions before specific questions. How would you rate overall, the impact of the treatment you are currently receiving on your life style? How would you rate your ability to eat what you like due to the treatment you are currently receiving?
- Ensure your response options are exhaustive. Are you providing all the response options to answer the question correctly? If not you will need to do a pretest using the “Other” category to identify those options which need to be included.
Look out for our next postings: (1) 10 Easy ways how you can increase response rates to your online survey.
(2) Comparing the benefits of paper and online questionnaires
(3) Understanding the different types of questions for online surveys
If you would like to know more about designing online survey questionnaires and how we can help you please provide your details below.
You can also visit our website: www.healthsurveysolutions.com
Categories: Questionnaire design