If you want to create great survey and poll questions without the hassle of writing your own, you can use surveys written by the experts at National Research Center in the Polco Library.
If you’d like to write your own questions, begin here:
Define the objectives or goals for the question to ensure the information you collect will provide meaningful feedback. Collected data should help inform some action on the part of the city or organization.
Use clear, precise wording and sentence structure everyone will understand. Avoid using jargon, acronyms, and double negatives. A good question means the same thing to all respondents. Make sure the average person who doesn’t work in your field also understands all the terms in your survey.
If necessary, present any relevant background information in a neutral manner. Present enough information to ensure that respondents understand the topic, but not so much that some respondents won’t consider all of it. If respondents consider different amounts of background information, their responses will be based on different stimuli, negating the benefits of a standardized survey.
A good question is worded neutrally and does not influence respondents to select one response over another. Use balanced wording in the question and response options so that all possible opinions are neutrally represented. When you write a question, think about whether it is as easy for your respondent to agree as it is to disagree, and make sure that you are not signaling a correct or preferred answer. We want to welcome all opinions whether they are good or bad. Using neutral questions helps the survey to feel less like marketing and more like a tool that can provide actionable data. For example:
Biased: How often do you attend (city name) City Council meetings? (Implies that they should be participating)
Neutral: How often, if at all, do you attend (city name) City Council meetings?
Make things easy for the person taking the survey. Survey questions that are too complicated will lower response rates and decrease community representation in the results. To reduce respondent burden, keep literacy levels lower, use easy questions at the beginning, and put sensitive or more difficult questions at the end of survey.
Focus on one topic per question. You might get an inaccurate result or confuse the respondent if you ask about multiple topics in one question. Ask about one specific topic at a time so that a single response is not conflated with two or more possibly differing opinions.
Pilot test! Great survey questions are straightforward and easily understood by the survey taker. It’s important to always pilot test questions before using them. Get feedback on how user-friendly the survey questions are.
Avoid false assumptions. Think your questions through to ensure you haven’t missed a necessary response option.
Consider social desirability bias. People want to tell you things they think you want to hear. They will err on the side of trying to make themselves look better. Give them permission to give a negative answer by structuring survey questions that encourage them to feel okay if they answer negatively.
Limit open-ended questions. We recommend using fixed-choice questions as much as possible for surveys since open-ended questions can be more challenging for the respondent and harder to code later. With that said, open-ended questions work well in small amounts when you can’t think of specific response options.
Carefully consider your response options. There are many considerations when choosing your response options. Experts often use five scale points, including a “Don’t Know” option. It’s also best to label options with words rather than numbers.
If you want to draft a semi-custom survey:
Carefully investigate previously-created surveys. Make sure you are modifying only surveys that follow the best practices listed above. They should be written by someone with both experience in the topic area and expertise in creating surveys.
Watch for scope creep. Sometimes when you use someone else’s survey, you may find questions you hadn’t planned to ask, but that would provide interesting data. Keep yourself focused on the task at hand, but keep those other questions in mind for a follow-up survey.
Consider if the questions are really suitable for your situation or environment. Many times surveys written for the private sector may need to be modified substantially to be appropriate for the public or nonprofit sectors.