How to encourage learning with educational apps
As a new academic year begins, millions of pupils around the world will be going back to school without leaving their houses. With reopening still in turmoil in areas hit hard by the COVID-19 pandemic, many places of learning are turning to digital options to help keep everything from lectures and teaching time to coursework and exams going. As always happens, myriad solutions have sprung up, ready to help. But building an educational app isn’t an easy task. Virtual learning takes many forms and, as some students have already learned, not every case actually helps them progress.
Two weeks ago, e-learning portal Edgenuity came under fire when parents and kids realised a fundamental problem with handing over grading assignments to an algorithm. Pitched as a way to reduce teachers’ workloads by intelligently assigning marks, the system’s artificial intelligence was quickly beaten by entering a jumble of semi-relevant keywords in lieu of a properly written solution to short-answer questions. By dumping a word-salad of terms that tricked the AI into giving a perfect score, some students went from failing to achieving straight-A’s overnight.
While you have to admire the kids’ ingenuity (ha!), it’s clear that this strategy doesn’t set them up for success in the future. However, when the alternative is spending time crafting a carefully constructed (and correct!) answer that still fails to get full marks, there’s no incentive to do more.
Encouraging students to get the most out of education software is the real challenge. It’s one we encountered when developing PlanetTeach, a pioneering game that helps primary school-aged kids learn Spelling, Punctuation & Grammar, both at school and at home. Throughout the planning process and while testing the app, we kept asking one key question: how does this serve the learning outcomes?
The learning outcomes are the core goal of any educational service. Everything you do should focus on encouraging only behaviour and actions that are related to what pupils are meant to be learning. For PlanetTeach, this had a major impact on how we designed the gamification features and scoring system. Correctness and accuracy were our primary concern, while speed was less important - we didn’t want to push players to rush or risk exploits where the faster a user clicked on answers, the more likely they were to get a high score.
This is where Edgenuity fell down - the learning goals didn’t match their execution. On paper, scanning for target keywords makes sense but it’s not enough on its own. Instead of rewarding answers that hit their desired talking points, they’re essentially teaching kids to cast a wide net of vague terms at the expense of their grammar and prose-writing abilities. At best, they were learning how to use search terms more effectively.
The main takeaways from this are:
Keep your desired learning outcomes clear at all times: if something you’re doing isn’t serving them, it needs to be addressed.
Never stop testing: get your app in front of as many different people as possible. Teachers, students, parents - all of these users will come into contact with your app in some capacity. Any issues they encounter will be amplified when more people get their hands on it.
Don’t underestimate users, especially kids: like with video games, if there’s an exploit to be found, you can guarantee that someone will discover it given enough time.