Key points
Write questions by working backward from "what you are asking in order to decide."
Ask about only one thing per question, and avoid leading wording.
Asking for answers by multiple choice and following up on reasons with short free text makes the results easier to use.
Updated
Overview
The first thing to decide when writing questions for a resident survey is "what these answers will be used for." If the use is clear, such as deciding how to run the summer festival, setting the date of the disaster drill, or reviewing the location of the garbage collection point, what you need to ask narrows down naturally.
If you line up questions without deciding on their use, "just in case" questions multiply and the burden on respondents grows. The answers you collect also tend to sit unused, with no one sure what decision they are material for.
Another important point is to ask for the "reasons" behind the chosen answers. The numbers of "for" and "against" alone do not tell you what to change to win people's agreement. Knowing the reasons reveals where opinions divide and makes it easier to think about the next move.
Steps
- Write the purpose in one sentence. Write it in the form "To decide X, we want to know what Y think." Remove any question unrelated to this sentence.
- List the material needed for the decision. Bullet the information needed for the purpose. For a summer festival, for example, this includes intention to participate, convenient timing, and what people can help with.
- Cover one issue per question. Do not ask about two things in a single question.
- Ask in neutral language. Do not use a preface that leans toward a particular answer. Avoid jargon and abbreviations.
- Arrange the options. Make sure people can choose without gaps or overlaps, and include "Other" and "Don't know."
- Add questions that ask for reasons. For important questions, add a short free-text question such as "Please tell us why you think so."
- Test with a small group. Have a few officers or acquaintances answer, and fix questions where they hesitated or took a long time.
Example questions
Summer festival
- Would you like to take part in this year's summer festival? (Yes / Yes, if my schedule allows / No / Don't know)
- Which events would you like the summer festival to continue? (Select all that apply)
- Is there anything you can help with in preparation or on the day? (Setup / Reception / Cleanup / Difficult for me)
- Please tell us why you chose the answers above, or any suggestions on how to run it. (Free text)
Disaster drill
- Which time slot makes it easiest for you to take part in the disaster drill? (Weekday evening / Saturday morning / Sunday morning / Other)
- What would you like to experience in the drill? (Checking evacuation routes / How to use a fire extinguisher / How to confirm people's safety / Other)
- If there is a reason you have not taken part so far, or found it hard to take part, please tell us. (Free text)
Garbage collection point
- How do you feel about the current location of the collection point? (Fine as it is / Should be changed / Neither)
- Please tell us why you feel that way. (Free text)
- If the location were changed, is there anything that concerns you? (Free text)
General meeting agenda items
- What do you think of the proposed revision of membership fees for next fiscal year? (For / For, with conditions / Against / Don't know)
- Please tell us why you think so. (Free text)
- If there are points you would like discussed at the general meeting, please tell us. (Free text)
If you collect opinions on agenda items before the general meeting, the discussion on the day can start from the points where opinions divide.
Common mistakes
| Mistake | Example | How to fix it |
|---|---|---|
| Leading | "Regarding the collection point that many people find inconvenient, do you support relocating it?" | Remove the preface and ask "How do you feel about the current location?" |
| Double-barreled question | "Are you satisfied with the date and content of the summer festival?" | Split date and content into separate questions |
| Vague wording | "Do you often take part?" | Ask in terms of number or period, such as "How many times have you taken part in the past year?" |
| Missing options | The time slot options do not include weekday daytime | Add "Other," or review the options |
| Assumptions that are hard to answer | Asking everyone "Why don't you take part in the drill?" | Set up the flow so that only people who did not take part are asked |
| Too many questions | Asking many demographic or impression questions with no decided use | Remove questions unrelated to the one-sentence purpose |
Writing questions with Baisoku Survey
With Plural Reality's Baisoku Survey, when you write "what you want to know" in everyday language, it asks you follow-up questions on screen, and when you answer by choosing, a draft of the questions is created. You can also attach materials such as notes from last year's review meeting, which makes it easier to write questions that reflect last year's issues. Staff can review and edit the draft.
Respondents proceed just by choosing from options. Depending on the answer, it follows up by asking "Why?" Even for the same question, what it follows up with changes depending on the answer. You can collect the reasons for chosen answers without lining up many reason fields in the questionnaire.
The collected answers are counted automatically and turned into an overall report that shows "points of agreement," "points of disagreement," and "next steps." The summary does not steer toward any particular conclusion. The final decision is made by people, such as the board of officers or the general meeting.
Related case studies
Building a City Where Children and Young People Can Fulfill Their Dreams, Together with Citizens: Higashiosaka City's Effort to Uncover Honest Views Through Conversational AI
An example of using conversational AI to get at the real views behind surface-level requests, which connects to the idea of designing questions that follow up on reasons.
Read articleUsing AI to Overcome the Challenge of Translating Residents’ Voices into Policy
An example of turning residents' diverse voices into policy. Useful as a reference for designing questions based on what the collected answers will be used for.
Read articleFrequently asked questions
- Should we include free-text questions?
- With multiple choice alone, you cannot tell why people chose their answers. Including one or two short free-text questions asking for reasons makes it easier to use the results to improve policies and events. With Baisoku Survey, you can follow up by asking "Why?" depending on the answer.
- What is a "leading" question?
- It is wording that makes people want to choose a particular answer. For example, if you ask for approval or disapproval after the preface "Regarding X, which many people find burdensome," answers tend to lean toward approval. The basic rule is to drop the preface and ask in neutral language.
- Does Baisoku Survey create questions automatically?
- When you write "what you want to know" in everyday language, it asks you follow-up questions on screen, and when you answer by choosing, a draft of the questions is created. You can also attach materials such as notes from last year's review meeting. Staff can review and edit the draft, so people decide the final questions.
- How many options should we provide?
- Provide a number that lets people choose without gaps or overlaps. Adding "Other" or "Don't know" for people to whom none apply avoids forcing them to choose.
Try Baisoku Survey
Baisoku Survey starts with the Free plan (¥0, 10 per month). Write the questions, collect responses, and organize the free-text answers on one screen.
Open the Baisoku Survey site (opens in a new tab)