The average researcher works with an incredible variety of data, including material on all manner of sensitive topics that demand careful research design and collection.
The goal is to conduct research in a way that’s ethical, makes participants feel comfortable, and is fully compliant from a legal point of view. In this blog, we describe nine best-practice techniques to do exactly that, empowering researchers to handle even the most sensitive topics with confidence.
1. Know your regulators.
Before we go any further, it’s crucial to understand what’s meant by “sensitive” in the first place.
At GWI, we treat the GDPR as our gold standard for global research – with good reason.
The GDPR was one of the first consumer data protection laws, and sets very high standards in terms of the rights and safeguards it guarantees individuals. Many frameworks that came after the GDPR borrow directly from it, while others don’t go as far to protect the consumer.
That’s why being compliant with the GDPR means being compliant pretty much everywhere – which is the best thing we can do to protect our respondents.
Following the GDPR framework also makes good business sense when looking at the future of data protection.
While it’s still possible to operate in countries without data protection laws, it’s only a matter of time before this changes. Data regulation is rightfully becoming more protective of respondents, so early compliance is essential for survival.
2. Define “sensitive topics”.
With the help of the GDPR, understanding what qualifies as “sensitive” is much easier. In fact the GDPR lists “special categories of data” for exactly this purpose. This includes information about a person’s political leanings, health conditions, sexual orientation, race/ethnicity, and religious beliefs, among other things.
But even these categories aren’t exhaustive. Deciding whether something is a sensitive topic or not requires a bit of judgment, so researchers need to ask themselves two key questions to decide:
- How comfortable would people reasonably feel volunteering this data about themselves?
- Could revealing the data collected about a participant ever adversely affect them?
If they’ve any doubt about a participant’s comfort or safety, the responsible thing to do is consider it a sensitive topic.
It’s also important to take a broad view when thinking about “adverse effects” and how to protect respondents.
Looking beyond obvious risks like criminal liability or identity theft; it’s important to consider things like loss of reputation or potential discrimnation at work, school, and so on.
3. Ask why you need to collect this data.
With even the most careful approach, asking people about sensitive topics can cause discomfort. So before embarking on your research, it’s good to make sure you really need to collect this type of data.
Researchers often have hoarding tendencies, mainly because it makes sense to collect more data than you may need so you don’t get caught out later.
But with sensitive topics, we have to resist that urge.
If we’re asking these questions to ensure representation in our sample, or because the study’s objective is to understand a topic that’s sensitive by nature, then it’s all good.
But if a question is a “nice to have” rather than a “need to have”, we should stop and leave it unasked.
4. Get informed consent.
Anyone trained in the social sciences will tell you that, when researching people, getting informed consent is a critical first step.
Informed consent means more than just making sure a participant has said yes to a study. It means making sure they have the right information to make a clear decision.
This includes letting people know how the data collected on them will be used, whether it’ll be shared with other parties, and whether their confidentiality and anonymity will be protected.
It also means making sure only those fully able to give consent are asked to do so.
For survey-based market research, kids and teens under 16 can’t give consent, and the same thing goes for respondents who are given the terms of a study in a language they don’t understand.
Researching sensitive topics demands an extra layer of consent to make sure participants are fully aware of what they’re being asked and why it’s being asked in the first place.
Respondents taking a survey about their sexual health, for example, might be greeted with a screening question informing them about the nature of the study and giving them the chance to opt-out. And even if they opt-in, all questions should have a “prefer not to say” option.
5. Protect your data.
Sensitive data, if linked back to a respondent, has more potential to harm them than non-sensitive data. Information about a participants’ drug use, for example, could get them fired at work or even lead to legal problems. The same could never be said about data relating to their music or sports preferences.
This is a key reason why researching sensitive topics demands stringent data protection.
In market research, these efforts are often two-fold. One part of the process relies on researchers and panel providers working together to create processes that guarantee full respondent anonymity.
The other part relies on having the right tech infrastructure. Using secure communications channels to transfer data, secure cloud storage to house it, and using best practice in terms of encryption or password-protection all help protect against data leaks. Having a good back-up plan in case the worst does happen is also essential.
6. Maintain neutrality.
When researching sensitive topics, being respectful of the respondent’s experience is key.
A big part of consent is being able to withdraw it at any time.
Respondents who suddenly find themselves uncomfortable answering questions about sensitive topics may well decide to withdraw their consent – and experienced market researchers know to expect a higher drop-out rate when probing into these topics.
Creating a safe space for people to answer honestly, however, can reduce this effect.
This means taking great care with language, and designing questions that are as neutral and judgment-free as possible. This not only makes respondents more comfortable, it can also help with data quality by reducing social desirability bias, where people give responses that are more socially acceptable or appropriate versus what they actually think or feel.
An example might help illustrate this point. If we wanted to understand alcohol behavior, asking respondents “how heavy is your drinking?” has a different tone than “how much alcohol do you drink?” While neither question is totally free of social desirability bias, the second option is undoubtedly more neutral.
7. Treat vulnerable populations with care.
With the right approach, researching sensitive topics among the general population can be done securely and respectfully. When researching more vulnerable groups, however, an added level of care is required.
Some respondents can’t effectively give their consent to participate in research studies.
As we mentioned earlier, kids and teens under the age of 16 are part of this group. While studying kids is possible with the consent of their parents, researchers need to avoid sensitive topics for obvious reasons – they’re often not age-appropriate or relevant, and have the risk of causing distress.
Vulnerable populations also include those who are technically able to give consent, but are particularly susceptible to coercion or distress. This might be people currently or formerly in prison, victims of traumatic events, people with cognitive disabilities, and older respondents (65+).
If in doubt, look at the comprehensive guidance provided by Institutional Review Boards for advice on identifying and researching vulnerable groups ethically.
8. Understand the role of inclusivity.
There are huge overlaps between identity characteristics and sensitive data – think race, sexual orientation, and disability, among many others. And, as we progress toward becoming a more inclusive society that recognizes and respects diversity, our research practices have to keep up.
In the UK, the Market Research Society looks at how we collect sensitive data and how to set a more inclusive standard. Still, this is a tricky area with plenty of challenges.
For example, how do we ask about gender identity in countries where cultural definitions of this concept are very strict? How do we embrace a more social model of disability in research that includes neurodiversity? What does it mean now for research to be truly “nationally representative”?
Working through questions like these will ultimately make all data – not just the sensitive variety – more inclusive, something that’s in all our interests.
9. Go forth and research with care.
It’s impossible to completely avoid sensitive topics as a researcher. The data we get from these topics is hugely beneficial, helping us tackle social inequalities, design better products, and ultimately understand the full human experience of our audiences.
The key is doing so with care, keeping in mind not only compliance with regulators and data security, but also putting respondents and their safety first.