More than half of all learning apps on Apple and Android target kids ages zero to five. But new research from the University of California Irvine School of Education has found that most are developmentally inappropriate and fail to teach the age group effectively. The 2018 study Are educational preschool apps designed to teach? An analysis of the app market, led by associate professor Stephanie Reich and Harvard post-doctoral student Melissa Callaghan, finds that none of the 171 apps tested provided all of the prompts, modelling, feedback and rewards that facilitate learning in preschoolers.
“There are decades and decades of research looking at how children learn best, but developers are really good at getting kids’ attention in short periods of time—so [the apps] are not really structured to let them practice or persist through challenges,” says Reich.
The study tested the top 10 premium and free math and literacy apps from Google Play, Amazon Appstore and Apple’s App Store over three months. Reich and her team started by creating a basic framework to capture and analyze the various learning features used in the games. Then Callaghan and three undergraduate researchers played each game several times in as many ways as possible: correctly, incorrectly, quickly, slowly, and sometimes the app was simply opened and left untouched.
What they found was that very few of the math apps taught broader lessons like even and odd numbers or how to count currency. And oddly, barely any literacy apps teach vowels and consonants, comprehension or sentence completion. The study recommends more apps offer open-ended learning that challenges players, rather than closed-ended questions that just quiz kids on their existing knowledge.
Some apps that stand out in a positive way are PBS KIDS’ The Cat in the Hat Builds That and Vectorkpark’s Metamorphabet. The former incorporates an estimation game into the narrative starring the mischievous Cat and his human counterparts, Nick and Sally. And Metamorphabet is an interactive alphabet game that animates and transforms letters that kids tap on into various things that start with that letter.
The study also found that 18.7% of the apps don’t provide instructions on how to accomplish the goal of the game, and less than 5% use modelling to demonstrate how to get to correct answers, even though research shows preschoolers need clearly defined goals and models to learn.
“It’s much younger kids using these, so they need a guide on what to do; and then as kids get older, the [offerings] can get more interactive,” says Reich.
Repetition is also key, but only 15.2% of the apps tested repeat instructions, and even fewer rephrase instructions for clarity after a pause in active play. Along with in-play guidance, preschoolers need feedback that explains why something was correct or not. Most apps give feedback for correct answers, but few provide it for incorrect answers to explain why. Most games just respond with something like, “No, try again.” But Reich explains that a more effective clue response, like “You’re looking for a triangle, which has three sides,” would be much better.
Beyond needing to provide explanatory feedback, Reich says scaffolding (a tactic that requires children to gain a foundation of basic concepts before they’re allowed to jump to more complex learning) should also be present for this age group. Shockingly, only 11.7% of the apps included in the study let users select their own difficulty level, which creates the potential for learners to place themselves in a level that’s too difficult for what they actually understand, and also to progress without showing improvement.
Half of the apps tested give rewards for correct answers, which tends to cause preschoolers to simply rush through the lesson to get to the prize. “Preschoolers learn through what we call baked-in rewards,” says Christine Elgersma, an expert app reviewer at Common Sense Media. “This means a seamless existence between the game and reward. Learning the narrative becomes the reward, so it doesn’t have this dynamic where the learning is the boring part and the reward is the fun part.”
The Homer: Kids Learn to Read app earned five stars from Elgersma for its research-based development and efficacy research showing that kids actually do learn from it. It features verbal instructions and visual prompts so kids can get through the content, even without parents present.
Elgersma sees a multi-disciplinary approach to educational preschool apps as the way of the future. “I would like to see experts on staff who are actually really deeply involved in the development and can add research-based input,” she says.
The first step towards improvement is to have educational experts and developers push for a clearer distinction about which apps qualify as educational, and which would be better off being branded as just for fun. Next, developers should add clear prompts and defined goals, and they must start including better modelling that demonstrates how to achieve the goal, along with more visual and audio feedback for both correct and incorrect responses. Repeating and rephrasing the instructions throughout the game will also benefit learners.
“The more scrutiny that apps and digital media get over time, the more it will impact business models and practices,” says Elgersma.