I was really excited this week to be presenting at the #PacRim2019 conference on UX research involving people with disabilities, one of my favorite things to talk about. I will be giving the same talk with with my co-presenter David Fazio at #CSUN2019 next week, and on my own at Shape, an internal VMware design conference in April.
If you know a fair amount about user research, but haven’t done research involving participants with disabilities, there are a few significant differences that the moderator or interviewer needs to be aware of.
Give Your Personas Disabilities
The purpose of personas is to create realistic representations of your most important user segments for reference. You may have Jason, who is a mid-level IT staff member who does SaaS administration, or Sandy in accounting who does PCI compliance and is working on her MBA. Your personas may also contain real-life details such as how many years of experience they have, what languages they speak, and personality traits that impact their software use (impatience, for example). Your personas should:
- Match the major user groups for your software/websites
- Identify significant needs and expectations of these major user groups
- Describe real people with backgrounds, goals, and values
Most UX research personas *don’t* have disabilities, which is astonishing given that 18 % of the population does. Making Jason an individual who is on the Autism Spectrum, or giving Sandy a temporary broken dominant wrist from a spring skiing accident forces the researcher to look at that individual’s software interaction through the lens of someone with a disability. That, in turn, may help your UX group identify improvements that can be made in your product or services — Jason might want a single location he can turn off all motion which he finds distracting, Sandy may get frustrated and tired having to hit the tab key 27 times to get to the footer where a “skip to footer” bypass block would be of significant benefit to her.
Finally, use these personas with disabilities when recruiting for your user interviews. You can generalize from what is known about some disabilities, but without talking to people who live with these disabilities day in, day out, you may miss an important need that will be expressed in an interview which will drastically improve their experience.
There are several differences in recruiting people with disabilities for user research.
Focus Groups vs. User Interviews
People with disabilities, especially screen reader users, prefer individual, one-on-one user interviews over focus groups. Imagine six people all using screen readers in a single room. It is complete chaos because if the participant is not wearing headphones, you have lots of voices talking over each other continually. When your participants are wearing headphones, they can’t hear the moderator. Also it can be awkward to discuss one’s disabilities (especially the health aspects) in a group setting. Hence, individual interviews are much easier to accomplish and the benefits far outweigh the increased costs and time involved.
Many people with disabilities don’t drive. Relying on public transportation or Paratransit can introduce timing issues for arrival and departure. Giving the participants a ride share credit covering their round trip to where you are conducting the interviews is money well invested.
When screening potential candidates to see if they fit your personas that you’ve extended to include disabilities, you will have to inquire about:
- The nature of the candidate’s disability
- The length they have experienced the disability
- What assistive technology they use
- Their proficiency in their preferred assistive technology
It is important to try to get a cross-section of people with both acquired and congenital disabilities. Two people with identical disabilities may have very different attitudes towards that disability depending on whether they perceive it as having an ability which was recently lost versus a long-standing disability.
Conducting the Users with Disabilities Interviews
Once you have recruited the candidates (and make sure you try to get an extra “floater” in case somebody cancels at the last minute) there are also several differences about conducting the actual user interview when your users have disabilities
Bring their own hardware
Users either disabilities, especially those who have been disabled for a long time, prefer to use their own hardware and AT. Usually they have highly personalized setups, and requiring them to use other hardware would actually present a barrier to those participants and invalidate your results. This is another reason why individual interviews work better than focus groups, because it is likely that many of your participants will be using different hardware setups.
You need an AT specialist at the interview
Because it’s pretty much a given that your users will bring their own hardware, there’s always the possibility that there will be configuration issues between their assistive technology and whatever it is that you’re trying to test. Having an assistive technology specialist on standby is always a very very good idea.
Interviews will take longer
You need to allow for 15 minutes to set up whatever it is you’re testing on your participants hardware. Also, people who are using assistive technology take longer to process data. What you would normally accomplish in an hour with a user without a disability will likely take an hour and a half for user with a disability. to be fair, you will want to compensate the participant for the full 90 minutes.
You Need to Have a Working Prototype
You need to be further down your design and prototyping pipeline when working with people with disabilities, especially vision loss.
When assistive technology is not involved, you can show a person two designs and ask them which they like better, and why. As soon as assistive technology becomes the intermediary between the person and the design, the design cannot be a 2-D printed sheet of paper, it must be functioning and accessible enough to expose the elements you are testing through the screen reader
How to Convince Management that UX Research with PwD is Important?
UX research is time consuming and expensive, and research with people with disabilities even more so. However, there are several reasons that management may listen to that will hopefully persuade them to agree with the cost.
Finding issues earlier makes them cheaper to fix
There is a commonly cited “ten fold” escalating model for fixing products. What costs
$1 to fix in design costs
$10 if not fixed until development
$100 if not fixed until QA
And $1000 if the fix goes into production.
Doing UX research in design when you can, and when you have a working prototype if you need one, reduces the chance of needing those very expensive production fixes.
Save your customers money
If you are selling a product or service that will be used in a business setting, identifying repetitive tasks will save your customers money which is a competitive advantage. These might not even be noticeable for people without disabilities but they’re much more noticeable when you’re testing with people with disabilities. Saving two to three seconds on a task that’s executed 60 times a day multiplied by a couple hundred users starts to turn into real money fairly quickly.
Less friction in purchasing means higher customer conversion
If you have an end user that is a consumer, making the product work for people with disabilities typically involves removing friction points. Eliminating friction points benefits everybody, not just people with disabilities, they are a “Curb cut”
The “Halo Effect”
Companies can improve how others perceive them when they do positive things that become widely known. This is caused the halo effect. This week, for example, Southwest Airlines achieved a halo effect by delivering a forgotten bridesmaids dress to Costa Rica in time for a wedding. Even people who have a generally negative opinion of airlines smiled when they heard about the story. Doing something positive for people with disabilities can give a company a halo effect.
What is the value of the lawsuit not filed? I call it “accessibility zen” it because it reminds me of the zen koan “What is the sound of one hand clapping?”.
For starters, there are enormous legal fees, both yours, and possibly the plaintiffs as well if they are the winning party. Then there is the cost of the settlement itself, reputation damage if there are reports of the lawsuit in industry publications, and opportunity costs if engineers are pulled from implementing business features to undertake emergency accessibility fixes.
All of this may be avoidable if user research is performed including people with disabilities and incorporating their commentary.
Expand your potential customer base
Certain sales such as those to entities receiving money from the federal government including the military, state governments, local governments, colleges, universities, and hospitals, require section 508 compliance. This means that the product being sold must be usable by people with disabilities. Doing UX testing with people with disabilities helps the product achieve 508 compliance as well as allowing their product to be used by anybody who is using assistive technology.
UX research is a really important part of the design process. You can’t know with 100% certainty that you’re giving your users what you want unless you ask them. Not including users with disabilities in this research is not only discriminatory and exclusionary, it’s short-sighted. Why would you ignore the needs of 18% of your potential audience? Understanding the specific needs of users with disabilities with respect to their participation in focus groups will go along way towards making sure that their input is valid and understood.