Introduction

Surveys are a common and useful tool for collecting feedback, measuring impact and guiding decisions when they are well designed. They are best suited for collecting quantitative “what” data about attitudes, behaviors and preferences that can be counted, compared and tracked across groups and over time. Surveys are less effective for understanding the “why” behind behaviors, long-term outcomes or deeper experiences. This fact sheet outlines practical strategies for writing effective survey questions and developing survey tools that produce useful and actionable data.

Survey Planning

Before writing questions, clarify the survey’s purpose and goals. A clear purpose keeps the survey focused, improves question quality and increases the relevance of the data. Planning ahead also helps later with selecting what types of questions are most appropriate. When establishing survey goals, consider:

• What information is needed?
• What decisions will the survey data inform?
• Who is the target audience?

Define 1-3 core goals and use them to decide what to include and what to leave out. Every question should serve a clear purpose tied to a survey goal. Avoid “nice to know” questions that do not support decision-making or provide meaningful insight. Extra or unfocused questions can confuse respondents, lower response rates and complicate analysis.

Digital vs. Paper Surveys

Both digital and paper surveys have advantages. The best choice depends on the audience, goals and context. Digital surveys (e.g., Qualtrics, Google Forms, SurveyMonkey) are easy to distribute widely, allow instant data collection and often include built-in analysis tools. They also support advanced features that affect survey design and functionality. However, digital surveys require internet access and may exclude people who are less comfortable with technology. Paper surveys are useful for in-person events or populations with limited internet access or digital skills. However, they require manual data entry and analysis.

The choice between paper and digital surveys should reflect what the audience is most likely to respond to, how broadly the survey must be distributed, what design features are needed and the program team’s capacity for data entry and analysis.

Survey Design

Survey design affects whether people complete a survey. Well-designed surveys are easy to complete, respect respondents’ time and clearly communicate the value of participation. Poor design can lead to low response rates, incomplete answers and inaccurate data.

Keep It Short and Simple

People are more likely to complete short, easy-to-understand surveys. Digital surveys under 10 minutes generally perform best. Long or complex surveys reduce completion rates and response quality. Minimize question types, prioritize the most important information and use plain, jargon-free language. Keep surveys accessible by considering the target audience’s reading level, language needs and accommodations. Group similar questions together and use skip logic when appropriate to avoid asking irrelevant questions. Begin with a few easy questions to help engage respondents (Galesic and Bosnjak, 2009).

Make It Easy to Complete

Most digital surveys are completed on mobile devices, so surveys should be optimized for phones and easy navigation. Include a direct survey link in the invitation whenever possible to minimize extra clicks. The easier a survey is to access, the more likely people are to complete it (Qualtrics, n.d.).

Survey Appearance

A clean, simple layout performs better than a cluttered or complex one (Dillman et al., 2014). Use readable fonts, clear headings and consistent formatting.

Distribution Planning

A strong survey also requires a strong distribution plan. Response rates are influenced by how a survey is shared, who receives it, when it is distributed and how participants are invited.

Matching the distribution method to the audience, crafting intentional invitations and using reminders and incentives thoughtfully can improve participation and data quality.

For more detailed information on distribution planning including outreach and recruitment strategies please see the Increasing Survey Responses fact sheet.

Writing Survey Questions

Writing good survey questions is both an art and a science. Effective questions are clear, unbiased and easy to answer (Bradburn et al, 2004). The following considerations can help improve question quality.

Ask One Thing at a Time

Avoid combining multiple ideas into one question. For example, “Was the workshop useful and enjoyable?” forces respondents to answer two questions at once, which is a problem if they have different answers for usefulness and enjoyment. Instead, ask separate questions about usefulness and enjoyment. Asking one thing at a time ensures responses are answering only what is being asked directly.

Make Each Question Count

Avoid asking the same question in multiple ways. Asking the same question in different ways (e.g., “To what extent did you gain knowledge?” and “How much did your understanding increase?”) can confuse respondents, create survey fatigue and weaken data quality. Instead, clarify what specific information is needed and avoid redundancy. Each question should provide unique, actionable information directly tied to the survey’s goals (Pew, 2023).

Use Follow-up Questions

Questions should collect enough detail to support meaningful action. For example, if respondents indicate that a presentation was ineffective, ask a follow-up question that probes which sections or presenters were problematic. If participants report applying what they learned, ask what they applied. Follow-up questions provide data that leads to clearer decisions and stronger programs.

Use Yes/No Questions Sparingly

Yes-or-no questions work best for screening or eligibility questions (e.g., “Have you attended this program before?”). They are less useful for data collection because they lack detail and nuance. For example, asking “Did the program provide useful information?” with only yes or no options leaves no room for understanding what specific parts were valuable, what could be improved, or what questions or needs participants still have. In most cases, scaled or open-ended questions provide more actionable information.

Use Neutral Language

Question wording influences responses. Avoid leading (positively or negatively) or emotionally loaded language.

• Biased: “How great was the event?”
• Neutral: “How would you rate the event?”
• Emotionally loaded: “Do you support wasting taxpayer money on public parks?”
• Neutral: “To what extent do you support funding maintenance and improvements for public parks?”

Avoid Absolutes and Double Negatives

Using absolute words like “always” and “every” in questions leaves little room for nuance and can produce misleading data. A question like, “Do you always make purchases online?” risks excluding or confusing respondents who shop online occasionally, producing misleading results. Instead, ask about frequency and include a time frame when possible: “How often have you made an online purchase in the last 90 days?” Similarly, avoid confusing double negatives, such as: “Do you favor or oppose not passing the policy?”

Use Precise Language

Avoid vague terms like “regularly,” “often” or “good,” which can mean different things to different people. Replace them with specific behaviors or time frames to help ensure you and respondents have the same understanding of what is being asked. Instead of: “Do you regularly use the information you learned?” Ask: “In the past three months, have you applied information from the workshop?” Specificity also applies to common questions. Instead of asking “What is your income?” make the question more intentional by including a time frame, like monthly or yearly income.

Similarly, common words like “favorite,” “best” or “enjoyable” are open to interpretation. When asking for feedback about a program, you can’t know if a section of the program was their “favorite” because it was fun, useful, short or had the best speaker. Be specific about what you really want to know in relation to the survey’s goals. Are you actually asking about the most informative, impactful or inspirational parts? Or do you need to know what speakers should be invited back for the next program? Ask about that instead. Finally, avoid broad questions like: “What did you think of the program?” Instead, ask about specific aspects such as clarity, usefulness, or relevance.

Clarify Terms and Avoid Jargon

Surveys should speak the language of their respondents, not of industry insiders or researchers. Use language respondents will understand. Avoid jargon, acronyms and technical terms whenever possible. If they must be used, define them clearly and spell out acronyms at least once. Ask yourself whether all respondents would understand the terms regardless of their background or education level. Testing the survey with a sample of the target audience can help identify confusing language before launch.

Speak Like a Human

Survey questions do not need to sound overly formal. Conversational wording is often easier to understand and more engaging. For example, a formally written question might ask, “To what extent did facilitators adequately disseminate relevant knowledge?” A better-written question would ask in a conversational tone, “How well did the presenters share useful information?”

Use Balanced and Intuitive Scales

Rating scale or Likert scale questions should include balanced response options with equal positive and negative choices. Poorly designed scales can unintentionally bias responses. Similarly, a balanced scale shows clear differences between the answer options. For example, the following question does not have a balanced scale. It includes several scale points that express the same or a similar degree of satisfaction (fairly, somewhat, mostly) and includes only one negative option and four positive or slightly positive options.

For example, this scale is unbalanced:

• Very satisfied
• Fairly satisfied
• Somewhat satisfied
• Mostly satisfied
• Satisfied
• Not satisfied

Use Inclusive Answer Options and Avoid Excessive Forced Responses

Answer choices should be both all-inclusive and mutually exclusive. For example, “How old are you?” with ranges 0–20, 20–30, 30–40, 40–50 leaves out anyone over 50 and overlaps at ages 20, 30 and 40. Write numeric ranges to ensure there are no gaps or overlaps.

There will be situations where the answer options do not fit all respondents. In these cases, it is best to include answer options to capture these unique responses, such as opt-out and fill-in-the-blank answer options (e.g., “Not applicable,” “Other”). With questions asking about intent to adopt practices or behaviors, it’s useful to include an answer option for people who have already adopted that practice.

Additionally, it can be important to give respondents an option to not answer certain questions with an answer option like “Prefer not to answer.” Avoid the temptation to include one catch-all opt-out answer containing multiple possibilities. For example, “I don’t know/Not applicable” covers two different responses with different meanings; combining them fogs your data. Providing opt-out answer options respects respondent boundaries and prevents “bad data” from forced choices. Similarly, most survey questions should be optional rather than mandatory. Required questions can drive dropouts or random, low-quality answers when respondents don’t know or want to answer.

Most questions should remain optional. Required questions can frustrate respondents and increase dropouts. However, some questions may need to be required when using skip logic, collecting essential demographic data, validating age or other screening questions, or measuring key outcomes. However, requiring too many answers can frustrate respondents and reduce completion rates, so it’s best to limit forced questions to those that are essential.

Avoid Asking Impossible Questions

Some survey questions fail because respondents cannot realistically answer them. For example, asking “What would make you likely to subscribe to our newsletter” requires speculation, not lived experience. Similarly, while rank-choice questions are popular, the results can be misleading if respondents lack experience with all the items being ranked. For example, asking respondents to rank the most useful presentations at a workshop can be effective, but if participants did not attend all the sessions, the ones they had little or no experience with will naturally be ranked lower, not because they were less useful, but because respondents didn’t experience them.

Qualitative Questions

Surveys primarily collect quantitative data through closed-ended questions (e.g., multiple-choice or rating-scale questions). Open-ended questions (e.g., short-answer questions) can also gather qualitative information about participants’ experiences and perspectives. Quantitative data is good at answering the “what” and qualitative data is useful for answering the “why” questions about a project (Creswell, 2018).

Strong surveys rely mostly on closed-ended questions with a few carefully placed open-ended questions for additional context. Data collection benefits from one final open-ended question to allow respondents to share anything that wasn’t directly asked about, such as: “Is there anything else you would like to share?”

If a project’s information needs are mainly focused on understanding why participants think, feel or act a certain way (such as their decision-making processes or long-term experiences) or a survey draft becomes mostly open-ended questions, this is often a sign that another method, such as interviews or focus groups, may be more appropriate.

Sensitive Topics

Before asking sensitive or demographic questions, determine whether the information is truly necessary. If sensitive questions are needed, use respectful, up-to-date terminology and explain why the information is being collected when appropriate.

Some demographic questions require special care, such as race and gender questions. Allow respondents to skip these questions if they want. Inclusive language not only improves data quality but also shows respect for respondents (American Psychological Association, 2023). Also consider what the specific audience for a survey will feel comfortable answering. Cattle producers may feel comfortable providing data on the size of their operations, but farmers may not want to share their acreage.

Similarly, assess questions for negativity or biased wording. For example, “Do you suffer from hypertension?” may read as judgmental and negative, while “Do you have hypertension?” is more neutral. Similarly, questions like “What challenges have you observed from the implementation of X policy?” lead with the implied understanding that the policy change has produced challenges. Instead, ask the more neutral, “What changes have you observed from the implementation...” or pair it with a question asking about the positive effects of the policy.

Before Distribution

Before distributing a survey, pilot test it with a small group. Ask whether the questions were clear, the survey was easy to complete, and the logic worked properly. Test the survey on phones, tablets and computers to identify technical or usability issues. As of 2025, 16% of U.S. adults are “smartphone-only” internet users who do not subscribe to home internet, so ensure surveys are mobile-friendly to allow more respondents to complete them on their phones (Pew Research Center, 2025).

Conclusion

Surveys are widely used in Extension and research, but their value depends on thoughtful design. Effective surveys ask clear and purposeful questions and use appropriate formats to generate actionable data. Careful planning, strong wording, and selecting the right question types can make the difference between confusing and insightful data. Applying the practices outlined in this fact sheet can create surveys that are not only easier for participants to complete but also more effective in producing the insights needed to guide meaningful action.

References

American Psychological Association. (2023). Inclusive language guide (2nd ed.).

Bradburn, N. M., Sudman, S., & Wansink, B. (2004). Asking questions: The definitive guide to questionnaire design—for market research, political polls, and social and health questionnaires (Rev. ed.). Jossey-Bass.

Creswell, J. W., & Creswell, J. D. (2018). Research design: Qualitative, quantitative, and mixed methods approaches (5th ed.). Sage.

Dillman, D. A., Smyth, J. D., & Christian, L. M. (2014). Internet, phone, mail, and mixed-mode surveys: The tailored design method (4th ed.). Wiley.

Galesic, M., & Bosnjak, M. (2009). Effects of questionnaire length on participation and indicators of response quality in a web survey. Public Opinion Quarterly, 73(2), 349–360.

Pew Research Center. (2025, November 20). Mobile fact sheet.

Pew Research Center. (2023). Writing survey questions.

Qualtrics. (n.d.) Survey design best practices