1

We are creating a survey that allows companies to see a detailed report about their organisational pain points. The survey is comprised mostly of questions based on Likert Scale. However, we also ask for free-text responses that users (employees) fill out so that managers and company executives can get direct feedback from their employees.

We are essentially faced with privacy concerns as showing the full responses to managers (even though we don't show names) may still contain contextual clues that identify users. We want to show full responses, along with sentiment analysis and word frequency etc.

I'm looking for some advice on whether it is viable to allow managers to see the full responses if we add a disclaimer to the free-text response questions, explaining who the responses will be visible to. I've made a mockup below to show how this would look:

enter image description here

Any feedback and help would be highly appreciated.

1 Answer 1

1

This is an excellent question, and I don't think there's a correct, bulletproof answer, only answers that provide fewer damage scenarios.

The thing is that no matter what you do, managers will identify the employee, or worse, they'll think they've identified the employee and the answers will belong to someone else.

So no matter what you do, there's a risk.

You may be asking yourself, what to do then?

Well, as I said, there are no right answers, but here are some tips:

Think of the company's values: it's tempting to assume that managers will identify employees and perhaps retaliate. But. is that really the case? And if it's, are there mechanisms in place to prevent it? Many departments at HR have some policies in place to prevent workplace abuse.

Research underlying issues: it's a good idea to find out if the potential problems are just superficial issues, or if there are significant issues that require more stringent privacy measures. You can conduct 2 or 3 surveys to do this. You could even ask someone in a management position. Then it's possible to create a list of problems that you can measure quantitatively instead of asking open-ended questions.

Be realistic in your expectations: You're worried about privacy, and that's fine. But it'll be really strange for you to find important issues. People will be afraid to talk and may lose their jobs or be punished somehow. To be honest, there's a good chance they'll complain about little things, and no manager is going to care much about answers like "I want more light" or "I want a better chair." For really important problems, there are usually other mechanisms.

Consider quantitative research: while it's tempting to rely on qualitative methods, if privacy concerns are so important that they affect your own research (not the responses, but the research itself), try a quantitative analysis, such as a Likert scale or Semantic Differential with a wide range of questions. Also, quantitative research is much easier to analyze.

Investigate the possibility of working on ethics in the workplace: If you think the issues are important, you may be able to suggest working on a workplace education program and developing different values that cover the issues that exist. Again, this is usually the responsibility of HR departments, but if there's no HR department, this may be a good idea.

Don't show the answers to managers: just show the answers to high-level stakeholders OR have someone study the answers and then filter out sensitive questions.

Take bias into account: Remember that these types of responses will be very biased. Sometimes because employees are afraid (as mentioned above). Sometimes because employees think they're anonymous, and they might also give unfair answers.

As you can see, there's a lot to consider, but I hope this can help you with your project.

3
  • 1
    Thank you so much for this detailed answer. You've given me a lot to think about - But I must also comment on a few items you've mentioned here: We've run a few test surveys and come up with some very interesting and useful feedback on various issues - so the value is there. The survey itself is mostly quantitative with Likert scales like you've mentioned, the free-text is an added response option that people can add further responses. My big question: Am i in the clear with GDPR privacy issues if I add a disclaimer above the responses? Whether we show to managers or high-level people. Commented Jul 19, 2022 at 17:35
  • 1
    You're correct on what you say about the GDPR and your process is sound. My answer is more oriented towards the fact that there isn't a bulletproof options in scenarios like this, hence providing some alternatives for you to think about
    – Devin
    Commented Jul 19, 2022 at 17:39
  • You're right - there isn't a perfect solution to this and your suggestions to mitigate the damage are what I'll discuss with my team. Thank you again! Commented Jul 19, 2022 at 17:45

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.