2

The current product I'm testing requires users to connect their financial information (such as their bank, credit cards, etc.); after this info is connected, the user is shown their balances, credit lines, etc. What should I keep in mind, given that not all users will be comfortable inputting/having this info displayed in front of a researcher/recorded?

Could I provide users with "fake" log-in credentials they can use? The pitfall with that is I'm also testing how many accounts the user would connect until they feel satisfied. I don't want to pre-determine how many accounts a user is likely to input; rather I'd like to get a sense of how many they feel is enough for our database to get an accurate understanding of their financial status.

I think this also speaks largely to the question of how to conduct usability testing with sensitive/private info.

4

We work in mobile and have developed apps for the NHS and other healthcare clients. Clearly the patient's details and medical information is of a private and sensitive nature.

The way we have tackled this in the past is that our developers have wrote routines to anonymize the data. For some clients it is simply enough to replace names, addresses and dates of birth with stub data, for others it involves time consuming detail replacement, as 'Lorum ipsum' is rarely an acceptable substitute.

We have had one or two clients accept masking, but this is more rare than the norm.

Sadly, it's going to cost effort. There are some frameworks available at different cost levels, if you search for 'anonymize data framework' or similar.

One thing it's really important to make clear with your test subjects is how you will protect their information, not just from a storage perspective, but how you are going to present it when carrying out UX testing. If you can give them confidence in your procedures, they'll be more willing to take part.

  • Yes, so helpful. Thanks for that. I remember when I was conducting my thesis at a healthcare facility, I went through numerous round of IRB, until I realized that what I was missing was simply letting my participants know that they'll be disclosing private/confidential info, but my job was to let them know that all of that info would be anonymized and destroyed and not part of the overall study. Overall, very helpful response. Thanks! – Chris May 1 '15 at 16:17

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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