1. Ask About the Right Things

Focus on Essentials

Keep surveys concise by asking only essential questions. Avoid including information you can obtain from other sources, such as demographic data already available.

Avoid Impossible Questions

Surveys are best for capturing quantitative attitudinal data. Avoid qualitative or speculative questions that respondents cannot accurately answer, such as:

Poor Example: What would make a button stand out for you on a website?
Better Approach: Use usability or A/B testing to address these types of questions.

2. Use Neutral, Natural, and Clear Language

Eliminate Bias

Avoid leading questions. For example:

Biased: We are committed to achieving a 5-star rating. How would you rate your satisfaction?

Neutral: How would you rate your experience?

Simplify Language

Avoid jargon and overly formal wording. Pilot your questions with the target audience to ensure clarity.

3. Don’t Ask Respondents to Predict Behavior

Avoid Predictions

Respondents are poor predictors of future behavior. Replace predictive questions with those based on recent behaviors:

Poor Example: How likely are you to use this product?

Better Example: Approximately how many times did you use this product in the past 7 days?

4. Focus on Closed-Ended Questions

Surveys are inherently quantitative. Use closed-ended questions like multiple-choice or rating scales, supplemented by a few open-ended questions for qualitative insights. Include a final question such as:

Optional Open-Ended Question: Is there anything else you’d like to share?

5. Avoid Double-Barreled Questions

Split double-barreled questions into separate parts. For example:

Poor Example: How easy and intuitive was this website to use?

Better Example:
How easy was this website to use?
How intuitive was the process?

6. Use Balanced Scales

Ensure rating scales have an equal number of positive and negative options.

Unbalanced:
Excellent, Very Good, Good, Poor, Very Poor

Balanced:
Excellent, Good, Neutral, Fair, Poor

7. Provide All-Inclusive and Mutually Exclusive Options

Design answer choices to cover all possibilities without overlap.

Poor Example:
How old are you?
0-20, 20-30, 30-40, 40-50

Better Example:
How old are you?
0–19, 20–29, 30–39, 40–49, 50+

8. Provide an Opt-Out Option

Always include opt-out answers like Not applicable or Prefer not to answer. Avoid combining multiple opt-outs, e.g., I don’t know/not applicable, as it creates ambiguity.

9. Allow Most Questions to Be Optional

Avoid forcing responses, which may lead to:

• Bad Data: Respondents guessing answers.
• Drop-Offs: Abandoning the survey entirely.

10. Respect Your Respondents

Use Neutral Phrasing

For sensitive topics, avoid language that assumes negativity. For example:

Insensitive: Do you suffer from hypertension?

Neutral: Do you have hypertension?

Be Mindful of Privacy

For sensitive topics, ask only if necessary. If needed, provide context explaining why the data is being collected.

By following these guidelines, you can design surveys that yield accurate, reliable data while respecting respondents’ time and comfort.