This weekend, I received an email from my son’s kindergarten teacher. It described the activities planned to commemorate the children’s 100th day of school. I smiled as I read the first paragraph outlining plans for the 100th Day Museum: a collection of art projects the children are to make and display, with each piece featuring 100 “somethings”—cheerios, pennies, etc.
My smiled faded, however, as I read the second paragraph. The children were also encouraged to dress up as 100-year-olds. For this, kids are supposed to wear mismatched, outmoded clothes, bring canes and walkers, and “pretend” to be feeble, infirm, and out of touch.
This reduces the oldest members of our society to creatures of ridicule, promoting and encouraging beliefs that people above a certain age no longer have value or relevance. That just because they are above a certain numerical age, they no longer have a sense of style, a connection with what’s happening in the world today, or something to contribute to society. It teaches our children that it’s okay to make fun of people who aren’t young.
In response, I wrote a letter to the school administrators asking them to please reconsider this aspect of celebrating 100th Day. I also reflected on how frequently I encounter evidence of bias against the elderly in the user experience field.
“So Easy Even Your Grandma Can Do It”
We’ve all heard it. We’ve all seen it. Some of us have even said it. “Let’s make this so easy even a grandma could do it!” This is supposed to be a positive statement that expresses a common UX goal – ultimate ease of use. It’s anything but positive, however. It stems from the assumption that just because someone is older (and, let’s face it, female*), they are automatically less capable of engaging with technology.
That’s why it’s offensive.
My son is incredibly fortunate to have two great-grandparents involved in his life. While neither is 100 years old yet, they are both in their nineties. We connect regularly through FaceTime, text, email, and Facebook. For years, they've met regularly with a group of like-minded senior citizens to share tech tips and discuss which mobile apps are most effective. They buy things online, wear Fitbits, and create holiday wish lists on Giftster. In fact, my grandmother (my son’s great-grandmother) had a first-generation iPad before I did. My grandfather has always been an early adopter. In the 1980s, I remember him lugging home his Kaypro personal computer and writing programs to perform statistical analyses for his university research.
They were grandparents then and they’re great-grandparents now. Just because they’ve aged does not mean that their interests and abilities, particularly with regards to technology, automatically disappear.
The “Older User” Persona
Another place where the “grandma can’t do tech” theme rears its ugly head is in personas. Personas are fictional constructs used to represent types of users who engage with a product or service. They are often employed by design and development teams to understand more about who they’re building for and to empathize with real people’s goals and challenges. Robust personas are created from extensive research of actual users, doing actual tasks, in actual situations. They’re amalgamations of many different types of people who have similar goals, roles, and/or behaviors.
Robust personas are rare.
More often, personas are created without any research at all. Instead, they’re pieced together from a combination of assumptions, anecdotes, and stereotypes. This is where casual ageism (and other biases) can creep in.
A common “persona” I’ve encountered throughout my career is the “Older User” persona, often typified by an unwillingness or inability to engage with technology. The problem with many of these personas is that they rely on the sole factor of age to convey something meaningful about the user.
Here’s an example. Let’s say you’re designing an app to help people keep track of medications. One way to approach personas is the following:
What information in these personas is truly actionable? Well… hardly anything at all. The second bullet in each gets close, but ultimately falls short. For the “Older User,” we know he “often forgets to take meds,” but we don’t know why. Perhaps he has age-related cognitive impairments (which is what we’re supposed to infer from the inclusion of age). But it could also be that he’s traveling frequently in retirement and the switch in time zones causes forgetfulness.
That’s the problem with including demographic information and other vague statements in personas – each nugget of information lacks the specificity to inform design decisions. In this example, the reader must infer what else might be true of the user based on stereotyped assumptions about age.
Testing Personas for Bias
To identify and eradicate bias in personas, the first step is to ensure that they’re created from research and not from assumptions. The second step is to ask yourself if demographic information is necessary at all. If you removed the connection with age (or gender, race, socio-economic status, education, sexual orientation, religious affiliation, etc.), does the persona stand on its own? If you’re relying on demographic information to trigger certain assumptions in the reader about the user, then it’s better to spell those out clearly for the design team and not leave them up to interpretation.
Applying this test to the personas above, here’s a revision that provides teams with specific information about user behaviors and motivations:
These personas** are actionable because they include specific content that ultimately impacts design choices. Knowing that Arthur needs help setting up and installing apps, for example, the design team might choose to create a “caregiver” role in the app. This functionality could allow the caregiver to set medication reminders for Arthur and check to see if he's taking meds correctly.
Notice that in both personas, inclusion of cognitive limitations, discomfort with technology, and visual impairments are all listed. It’s not that we should shy away from including these details if they’re identified by research and relevant to the application we’re designing for; it’s that we shouldn’t rely on age-based assumptions to convey this crucial content.
One last note about bias in personas: If you include age and other demographic factors in your personas, please know that this post is not meant as an attack against you, or as evidence that you willfully introduce bias into your work. (Plus, there are some instances when age-related details are important to include, such when only certain users can access age-restricted content, for example.) I believe with my whole heart that UX researchers and designers do not intend any negative intent with the artifacts they create in the pursuit of better design. I truly believe this. But I also know that not having negative intent does not prevent negative outcomes.
That’s why it’s important to have an open mind and to revisit artifacts that inappropriately and misleadingly reference age (and perhaps other bias-laden content) to ensure that we’re not perpetuating tired tropes.
Footnotes
* Google results for search string, with quotations:
“so easy even your grandma can do it” = 47,000
Google results for search string, with quotations:
“so easy even your grandpa can do it” = 3
** I created the persona examples in this article completely from my imagination, so none are true personas! They’re for illustrative purposes only and are not related to any work I have conducted.
Kommentare