Dealing with Sensitive Survey Topics

By Frank Markowitz, Academia, SSI

Academic and public opinion researchers often want to ask people about sensitive topics – even topics that involve illegal activity such as drug use or certain driving habits. There are very good reasons for doing this type of research, which helps to inform the public and policy makers.

As a partner with public opinion researchers for 40 years, SSI has provided sample and data collection for thousands of academic and public opinion surveys. In doing so, we walk a fine line between our belief in the value of public opinion research and our duty of care to survey respondents, based on our understanding of what they expect when agreeing to take part in phone research or join an online panel.

That “duty of care” leads us to not allow questions to be asked of our online panelists or our phone respondents which would require them to report any illegal activity. Examples of these topics might be questions about drug use, teenage alcohol consumption, or sensitive topics such as immigration status.

While our policy on “illegal activity” questions is black and white, there are other questions which fall under the “sensitive” category where we must make judgments on a case-by-case basis. Examples include testing insulting or insensitive messaging, sensitive topics such as politics, religion, sexuality or race, or inclusion of graphic visuals. In some cases, we can work out a wording change, or add special introductory wording which makes the question acceptable, but there are some questionnaires, or specific questions that SSI must decline.

If the sensitive question is asked in a phone survey context it may be possible for a skilled interviewer to use tone or context wording to reassure a respondent and avoid offence, while online, where we only have words on a screen to reassure (words which many respondents won’t read anyway) we may be unable to reassure the respondent adequately and therefore cannot allow the question.

In all cases, we’re guided by keeping in mind that our panelists are people. When coming to a decision in difficult grey areas it can help to think about whether we would be comfortable with our 18-year-old son or daughter, or father or grandmother answering the question, or being shown these words or images.

Here are some best practices for dealing with sensitive topics:

  1. Be aware of cultural norms: politics, religion and other topics are much more sensitive in some regions.
  2. Do not assume that your view of topic sensitivity applies to everyone—ask friends and colleagues.
  3. Use clear introductory wording and an opportunity to decline the survey and still receive a reward. Be very specific: are there graphic images or videos? Will slang or explicit words be used?
  4. Explain who is sponsoring the study, its purpose and how the data will be used.
  5. Give people the option to take the survey later (e.g. when alone rather than in public).

Here is an example of Sensitivity Screener wording:

This survey will ask questions and contain content about [INSERT TOPIC], which may be sensitive to some people. Your participation in this research study is voluntary. You may choose not to participate. If you decide to participate, you may withdraw at any time.

Would you like to continue taking this survey?

Yes, I would like to continue

No, I would not like to continue [Terminate]