If you’ve ever conducted a survey you know how important it is to ask the right questions. Did you know there are actually types of questions to include and to avoid? Frequently someone will make a list of questions without giving it much thought but there is a better way which can yield more information from the participants.
The first priority with any survey is to determine the main focus for this specific survey. The survey should stay on topic. If a wide range of information is hoped for than consider conducting more than one survey or splitting the questions between two different groups of people.
Things to Avoid
- Free surveys should be relatively short; not exceeding 10 questions. If your survey is longer consider breaking it into multiple shorter surveys or paying participants to take the survey. Generally, the person taking the survey gains nothing to help you collect data.
- Avoid filling the survey with the same types of questions. All Yes/No responses or all open-ended questions, for example, are boring and wear on the survey taker.
- Avoid haphazardly listing questions in no particular order. If you use one response to lead into the next question that second question has the potential to yield a more in-depth response.
What to Include and Do Better
- Plan ahead and be strategic with which questions you really want to know more. Save your open-ended questions for these points. While the tendency may be to ask open-ended questions for each question studies have suggested that mixing the types of questions between open-ended, multiple choice, and rate this will give you a better analysis of data than asking only one type of question. Participants tire of the same types of questions. It’s been noted that surveys of all open-ended will have lengthier and more detailed responses for the first questions but as the survey continues the length and detail dwindles.
- In place of asking a yes or no response question consider having the person use a rating. On a scale of 1 to 10 with one being Not At All (aka No) and ten being Extremely Pleased (aka Yes) you can quickly see how the scale may reveal more depth than a mere yes or no. If there are a majority of fours that is more telling than seeing a no. A four is close enough to neutral that you could do some tweaking to the topic being asked about and perhaps a survey in a couple of months would yield more Extremely Pleased or numbers closer to this side of the rating scale.
- When asking multiple choice questions keep the choices distinctive, descriptive, and no more than five. Again, using a range of five will give you a high, a low, and a middle response with some in-betweens to gauge how you’re doing.
- The best surveys will be up to ten questions and mix up the types of questions being used. If you try this approach I think you’ll be pleased with how responsive the participants will be as well as appreciate the improved quality of the data collected.
I have noticed an increase in surveys around product development where the person clearly doesn’t have a direction to their questions and is grasping at trying to assess what her market wants. You may have participated in one of these surveys or even used it yourself. It’s filled with multiple choice questions similar to Which of these services would you use? Or, asks the open ended question What causes you pain or frustration in your business?
A better way for these types of survey questions to be worded is to ask someone to rate in order of importance. You will start to see some patterns developing in the answers that come in. Additionally, when you provide the list that should be rated you are centering in on your own products and services. What good does an answer do you if it is outside of your service offerings? Yes, you have new data but that data won’t directly link to something in your business offerings that you can market more. Be aware that a person may not be experiencing frustrations but should always be able to rate a list in order of importance.
Remember each survey should have a main focus and the questions should support that focus so each question builds and digs a bit deeper. Survey data is most valuable to you if it helps you gather data you can act on.