Johnson STEM, an after-school STEM program in Atlanta, had all of their data recorded on paper. Johnson STEM doesn’t have the data they need to make informed decisions about their operations. All residents, staff, coaches, and visitors have to sign in at the security booth before they are allowed access to the Center. Their existing pen and paper sign-in has resulted in backed-up lines and stacks of pages of data that have never been analyzed.
Here's what we knew from talking to Johnson STEM:
10,000
estimated participating students per year
?
average hours a resident spends at JSTEM
12-18
age range of students
0
concrete data to back any claims up
?%
of JSTEM residents that go to Title 1 schools
As you can see, we had no solid data to know what problem area to first tackle. We decided that collecting data to know what the problem was the first step.
After interviewing several JSTEM employees, we identified multiple pain points, including a lack of standard communication format, difficulty tracking data, confusion over lab scheduling, and an inefficient system to track attendance.
When surveyed, 80% of JSTEM residents ranked having a new sign-in system as being of high importance
After sketching out ideas for a sign in system that required a physical device like a card scanner or thumb print reader, we created paper-prototypes to send to JSAC and told them to record themselves trying to figure out how to use it. In response, they said they couldn’t figure out how to record themselves. Quickly, we realized that the lower tech we could design, the better it would be adopted at JSAC.
One limitation we saw quickly was that our ideas of either bringing an item to identify and check-in a student (ie a wrist band, or ID card, or fingerprint) could either be lost, not secure, or costly to implement. After rounds of brainstorming, we figured that an installed tablet at the security booth that we required just a username to check-in and check out of would be the most low-tech and foolproof method we could implement.
An account was necessary for our system to have so that a username that would sign in would also sign in with all the correct metrics associated with the user. Instead of just signing in “John Smith” on a piece of paper, our users would sign in with an account “johnsmith3” that has all of John Smith’s data associated with it.
Hypothesis: Having residents, staff, and coaches create digital accounts for signing in at the booth, wait times could be reduced. In addition, Johnson STEM would have concrete data and stats on the demographics of the school.
Our account creation went through four rounds of revisions. The second round of revisions had the most drastic changes, with adjusting how many categories of attendee status (from two categories to five) and order of information presented.
After our testing with the general public of BYU, we found oversights we could correct like our code didn’t recognize names with accents or design choices with text boxes and button sizes.
What went well?
Testing at BYU revealed a few oversights with our forms that we had.
We developed a great working relationship with the shareholders of JSTEM through our multiple rounds of testing.
What did not go well?
I had the opportunity to travel to Johnson STEM from Utah myself. I wanted to hand deliver a tablet to them to install and see how the physical location would be for our design. After interviewing the security guards, I realized what oversights we had just from being so physically removed from the location we were working for. “The tablet doesn’t really hook up to the wifi,” one guard said.
Another one seem to be reluctant to put the pen and paper away and bring out the tablet, citing it was hard to see the screen of the tablet in the sun.
Once those fixes were implemented, we got real data. The estimated data of 10,000 students coming annually went to… 1,115 students. Students who had real metrics and forms and demographics associated with them that JSTEM can use to get future grants.
1,115
actual students
1
grants secured using real data