Whenever you undertake surveying a population, whether employees, residents, or other stakeholders, you should do so with the end goal in mind. Commonly, people wonder what kinds of questions they should be asking.
The short answer is that you should ask simple, direct questions that will provide the information you need.
The longer, more elaborate answer includes a few more considerations. Ultimately, it depends what you want to know, who you want to learn it from, and what you intend to do with that information.
When crafting a survey, there are a few basic things to consider:
- Purpose – Take a moment and really think about why you are surveying. Is it because you have a specific problem to solve? Or because you want a broad view of how things are going in your organization? Does the survey need to give you precise measurements, or just general feedback? You will want to keep the survey questions in line with this overall purpose to help streamline things for both the respondents and those who will be working with the results.
- Audience – Who you’re getting the feedback from makes some difference in the question selection as well. You may want to keep the questions more casual and universal for a broad, mixed group. For a defined set of respondents (event attendees, for example), you can design the questions more specific to your topic at hand.
- Methodology – The way you execute a survey plays a part in the question design. For example, if you are administering an online survey, you probably have the option to use something like a sliding scale or skip logic to make your questions more dynamic. If you are using paper surveys, neither of those things will work well, so sticking to the basics is best.
- Timing – Related to audience and methodology, think about how long it might take for a respondent to take your survey. If you want busy professionals to complete your survey in a few short minutes, then keep the questions simple and to the point with easy rating scales and minimal need for typing or writing. If you are looking for a more substantial set of data or comments, understand that it will take more time and plan accordingly.
Once all of those things are understood, there are some things to consider when writing or selecting the actual questions. Keep in mind survey questions should be:
- Straightforward – Always make sure the phrasing is not leading, and the question is not “double-barreled” (meaning the question could be taken from multiple angles or is too subjective.) In other words, be careful not to bias respondents into a forced response and don’t ask about two or more things in one question
- Easy to read – With few exceptions, surveys should be understandable at a basic literacy level (some research suggests a 7th grade level, but opinions vary). They should also be carefully proofread for spelling and grammar accuracy; possibly nothing undermines your study faster than a poorly worded survey suggesting a lack of professionalism.
- Actionable – Think about this when selecting questions—will you be able to respond to the answers you get, and can you really do anything about it? For example, if you are considering asking residents to rate “availability of parking,” but you already know it’s not ideal and have no intention or ability to change the availability of parking, then you might want to consider not asking the question. This doesn’t mean you should only include questions that get you favorable feedback; it just means you should be realistic about what you’re asking and what data will be really significant for you.
A good survey partner will take the above considerations in mind and work with your organization to craft a survey tool that aligns with your mission and goals. This relationship should be dynamic, where the survey provider ensures a valid and actionable survey, but at the same time, keeping your priorities at the center of the survey design process.