Write questions for user research

Contributors

@andreea-macoveiciuc-content-expert


Business Benefits

Get more accurate, unbiased responses.


Add an opening clause to normalize behavior and make respondents feel more comfortable.

For example, “The average person spends more than two hours per day on social media. How much time, on average, do you spend?” or “The Mayo Clinic reports that red wine may be heart healthy. How often do you drink red wine?”

Add phrases that normalize negative responses to counter social desirability bias.

For example, “Do you happen to jog, or not?”

Use phrases that suggest the knowledge in knowledge-based questions is not expected.

For example, “Do you happen to know that…” or “Can you recall offhand the…”

Use opinion questions instead of knowledge questions for more candid responses.

For example, “In your opinion, what are…”

Include Sleeper Answers or an “I don’t know” option in closed-ended responses to prevent inaccurate answers and manage social desirability bias.

This can prevent inaccurate answers when respondents feel compelled to select one.

Include responses with an artificially high range for closed-ended questions that suggest norms.

For example, for the question, “How many hours, on average, do you watch television each day?” you could provide the options 0–2 hours, 2–4 hours, 5–7 hours, more than 7 hours to put respondents at ease about average versus extreme responses.

Order questions from most general to most specific so that the answer to the specific question doesn’t influence the response to the more general one.

Opening with a general question like “How would you describe your marketing team’s overall knowledge” allows you to follow with a more specific question such as, “How would you rate (1–10) your marketing team’s knowledge of lifestyle stages?”

Ask someone outside your organization to review the questions and help identify questions that may be confusing or suggestive.

Test open-ended survey questions with a pilot group.

Use open-ended response from the pilot group to determine value ranges for a closed-ended survey.