Hundreds of thousands of young people go to jail each year. Most face housing, employment, and interpersonal difficulties after they're released.
This team project aims to design a Human Computer Interaction solution for young formerly incarcerated people but providing them a sense of community and connections they are otherwise lacking. We did this over a quarter for our Introduction to Human Computer Interaction class.
Our solution is a mobile social network called LifeAfter that connects young ex-convicts to older mentors, influencers, educational organizations, NGO's and each other to create a supportive community for emotional and informational needs.
We focused on young formerly incarcerated people jailed between the ages of 18 and 25. This is a typical target demographic of young adults that is disproportionately affected by disruptions to their lives as they don't have the basic educational and life skills older people might have.
Using Design Thinking
In building the platform, my team and I followed the Stanford Design Thinking method. We chose this method because it was especially important to empathize with our sensitive population from every possible angle.
I interviewed ex-convicts, people from educational institutions, and NGOs along with my team. The short time frame of the project and general difficulty in outreach to our primary demographic necessitated semi-structured interviews instead of focus groups, contextual inquiries etc.
They provided us "rich and thick" data to unpack and to extract insights from. We often followed up with email questions.
Our interviewees included:
- Horza Gobochul (name changed) is an alumnus and board member of University Beyond Bars (UBB), an organization that “empowers prisoners to fulfill their potential through communities of higher learning that transcend prison walls.” He went into prison while still in our target demographic and as a board member of UBB, has insight into what young people face.
- Franny Rotolo, who works in a reentry program within the Department of Corrections that is tasked with preparing select incarcerates for reentry with the intention of reducing recidivism.
- Vosgi Yong (name changed), a staff member of the Freedom Education Project Puget Sound. FEPPS is an NGO that works with people incarcerated in the Washington Corrections Center for Women.
- Shura Bala, Lindsey Yuu, and Fu Isi (names changed), all mentors in the Formerly Incarcerated Group Healing Together. Although they are currently older than our target group, some had been in the same position of being incarcerated for the transition from youth to adulthood.
The users and the constraints
The final target demographic was young ex-convicts between the ages of 18-25. We found that the impact of a design solution would be greatest on the young adult group, because they are in a formative stage in their lives and face great challenges because of relatively little life experience.
These constraints were put on our research users because the initial demographic of recidivist criminals were spread too far and wide. Secondly, while we did want to design for someone other than ourselves, more than one vector of difference may have been difficult to bridge. For instance, the big differences between us and our target demographic are economic status, educational status, and legal status.
Adding recidivism would also have added a broader age group and individual instances of recidivism, which would have ballooned complexity and reduced the accessibility of the target demographic still further.
Our main insights from our interviewees were as follows:
- Most problems faced by young ex-convicts deal with a lack of information and legal status.
- Housing is critical because it presents ex-convicts with a whole host of legal problems. For instance, housing provides access to employment, because most jobs require some kind of address. Further, my interview with Horza Gobochul revealed that under Federally Subsidized Housing (NYCHA & Section 8) rules, people with arrest records and even their families can upon a background check be prevented from getting a place to live.
- Interpersonal Skills in the outside world are very different from those inside jail. According to my interview with Franny Rotolo, self preservation can often be a question of aggression within jail, which can be very harmful in outside society. They need to be on boarded in a way that doesn't affect their working lives.
- Other problems we spotted were trouble with employment, general life skills, and an education deficit. We chose to tackle the top two problems because we didn't see HCI based responses for them and because they stood out as problems that we could proceed to understand within a quarter.
In coming up with ideas for our solution, we brainstormed on a whiteboard. For my team, I pushed for two main rules:
a) Yes, and? I didn't want to reject any ideas, as long as they could be connected to our findings.
b) Consider "be-goals": I drew this theory from my Experience Design class and brought it into a Human Computer Interaction class. According to design theorist Marc Hassenzahl, a design should take into account a person's "be-goals," which "motivate action and provide it with meaning." Completing a marathon isn't a be-goal, but being admired for completing a marathon is.
Based on those constraints, we came up with the following design solutions:
- Design a mobile-only social media network as our minimum viable product. Former incarcerates need to have cell phone access. Their lack of housing makes access to a desktop terminal hard. Most jobs and housing applications have an online component, and finding community in a disparate group can be difficult. They are the cheapest, most ubiquitous, and most effective Human Computer Interface yet.
- Provide ways for former incarcerates to find short and long term housing. One of their biggest persistent problems is finding housing. Finding housing is intimately linked to finding employment, getting respect in society, and having a base to operate from.
- Provide ways for incarcerates to build their interpersonal skills. We found that one of the bigger unmet factors preventing our target group from achieving their ends was a lack of interpersonal skills. They had to be built "the hard way," with no support or mentor system provided that was effective and ever present.
- Build mechanisms (like checklists) to provide clear guidance and specific tasks for achieving their ends. In a highly regulated environment like prison, most incarcerates have every decision made for them. Thus, when they are released from jail, formerly incarcerated people have difficulty making everyday decisions.
In coming up with prototypes, we focused on developing the housing and relationships tasks, because we didn't find enough user generated mobile solutions that rolled in many of the other tasks. The paper prototypes were built to be high fidelity enough to undergo basic usability testing. The videos below describe the initial interactions.
Usability Testing: running it by the audience
For usability testing, I created two tasks for subject for housing and relationships, set their beginning and ending conditions, and used Think-aloud protocol. The latter is a technique where the researcher requests the test subject to "think aloud" as they work their way through the product. This helps designers better understand what the subjects are going through as they complete their tasks.
In choosing our subjects, we had some difficult obstacles. firstly, our target group of young ex-convicts proved extremely hard to connect with. Left with no options, we turned to classes of users who shared some of the attributes of our core demographic.
a) We user tested two students in the 18 - 25 year range. They represented the tech savviness typical of young people today.
b) We user tested a military veteran, who had led a very regimented life analogous to those in prison.
In creating the tasks, I was cognizant of how "real" the situation had to feel and hence carefully laid out tasks that felt true to life. The tasks were as follows:
1) Relationships task
Beginning status: You have just been released from jail and have your sister’s borrowed cellphone with LifeAfter downloaded. Your first priority is to connect with an old flame. Use the app to get to a support network. As you go through the system, please talk as you make your choices.
Success criteria: Making it to the U-District Same-Sex Relationships page.
2) Housing task
Beginning status: You have been out of prison for a week and have downloaded LifeAfter. You've also signed up for a housing checklist and want to make the next move: meeting with a roommate. However, you are apprehensive about what that might entail and have some questions specific to your situation. How would you tackle that on the app?
Success criteria: Going to the checklist page, finding the specific task, and clicking on "Need Help?" to go to the relevant forums page.
Usability Testing: the changes we made
While we didn't spot any major issues with the design as a whole through our usability tests, we still had some notable changes, especially as regards information clarity. Ranked by severity of the issue, we found a number problems and acted to solve them as follows:
- No forum page: The earlier prototype did not have a forum page because the forum was not explicitly needed for any of our tasks. However, that proved to be a simple impediment to users who needed to go somewhere when then clicked on the large tab labeled "forum" on top.
Solution: Added a forum "home" page just like the checklist and group tabs.
- Limited access to groups for non-members: The first prototype had limited access to groups for non-members. For instance, if a user wanted to check the U-District Same-Sex Relationships page, he wouldn't immediately have access to any of the content. This would prevent people from making informed decisions, limiting user control and freedom, one of Nielsen's heuristics.
Solution: Provide the top three comments from the last 24 hours in each group visible even to non-members.
- No guest access: More than one subject said that they would be interested to see the interface with a guest login to allow people who don't have immediate access to email to access the product. This would be useful, as housing is an immediate item that needs to be accessed right after a person is released. What would be more contentious is deciding to what extent the community is open to them.
Solution: Provide a "guest" login with a limited suite of essential features.
The Final Product
Shown below are the final screens submitted. The carousel changes once every 5 seconds, and shows some of the completed screens for our prototype. The team created most of the screens on Proto.io (available in this LINK). I handled the login flows, the group screens, and the final color schemes.