Gamifying the classroom is a bad idea

I was part of a generation of children who learned about dysentery through a video game. Every week, I sat in front of a bay of Apple II computers in the school library with my fourth grade class and played The Oregon Trail for an hour. First developed by an eighth grade teacher in Minneapolis, the now-legendary game depicted the difficulties of settling the American West in the 18th century. It introduced children across the country to topics like cholera, typhoid, squirrel hunting and river fording by including them in its basic mathematical system. But did anyone really learn anything from playing the game?

I definitely remember the infamous “You have died of dysentery” screen. But I never really learned what the disease was, why it was fatal and how long it took to play out; rather, I learned the word as a unit within the game’s competitive system. Dysentery was a “game over” screen, and not much else. 

In the years since I was in fourth grade, there has been a concerted push among educators to promote video games in schools. As more criticism has been directed at standardized testing of students, video games have emerged as a blessed compromise, combining a digital system of performance tracking with an excitable drive to “modernize” education and make learning more creative. An influential 2006 report from the Federation of American Scientists argued that video games could benefit the American education system, and a follow-up survey by the Gates Foundation–backed SRI International found that students taught with video games performed 12 percent better in cognitive tasks than students taught without games.

And yet the use of games in the classroom actually sustains some of the worst elements of the failing American education system: They use binary pass/fail states as the ultimate measure of success, reduce the complexity of thought to a simple log of input commands, depict dramatically oversimplified mathematical models of their subjects and often come with predetermined goals that students have little say in choosing. The games thus impose a predigested value system rather than challenging kids to identify and execute their own goals.

Video game companies have, naturally, been more than happy to play along with the interest of education reformers. Last fall, Electronic Arts announced it would release an educational version of SimCity, a game in which the player manages the growth of a city like an urban planner. In 2012, an educational version of the global phenomenon Minecraft was made available for schools, with a Wiki of different lesson plans and activities building on the game’s basic concept of gathering stone and wood to make cabins, castles or whole cities. And in 2012, Valve, maker of the legendary Half-Life series, repurposed its Portal 2 game into an educational platform called Teach With Portals, which uses the game’s puzzle room structure to teach kids about geometry, basic physics and spatial reasoning.

Education reformers, in turn, are happy to gamify the classroom. Educational games such as SimCity are efficient, deferring the data collection and evaluation to the computer while building games that compress pedagogy, practice and test into one cohesive experience of play. But their structures as digital systems also make them ideal companions for standardized testing: They measure a wide range of standardized performance metrics — from how many mouse clicks were used during a session to how the student distributed her money — and then judge these metrics against other students’ performances. Testing a student’s ability against a standardized baseline is certainly easier when it comes in the form of an entertaining game and not a fusty set of multiple-choice questions. Yet the outcome is strikingly similar, with the end product ultimately being a series of data points about what the player chose to do in what instance, just like a test record of what answers a student chose to enter after each question prompt.

What are the effects on students? In an interview with games website Eurogamer, Joel Levin, a New York City teacher who was instrumental in showing how the game Minecraft could be used in schools, described his students' consuming interest in the game overriding other areas of study, turning their entire curriculum into a de facto Minecraft seminar. “It was like nothing I’d ever seen before as a teacher,” he said. “These kids couldn’t wait to get to school to get a taste of Minecraft. They were writing stories about it in English class, they were drawing pictures of it in art class.”

Games such as Minecraft and SimCity channel this excitability back into systems of standardized testing, with assignments designed to closely match the Common Core, a controversial attempt to create a national set of standards for what students should be able to do at each grade level. These include benchmarks such as expecting eighth graders to be able to use the Pythagorean theorem or 10th graders to be able to properly cite a source in a written paper. Because computers can log data and give instant performance feedback, they make standardized forms of student evaluation more tenable, and when student performance is measured with the highs of solving game problems, results are bound to improve.

Presently, students spend anywhere from 60 to 110 hours a year preparing for the various state and federal standardized tests, which have become an immovable part of American pedagogy. Spending on standardized testing has risen at a staggering pace in the United States over the past half century, from $7 million in 1955 to an estimated $1.7 billion in 2012. Over the same period, the performance of American students declined dramatically, with drops in all categories of the Organization for Economic Cooperation and Development’s most recent test of global education systems.

Yet there is the threat with these declines in U.S. test scores to overemphasize testing to compensate. As has been noted again and again, increasing student scores rarely correlates to intelligence. In 2012, researchers from Western University published results from a survey of 100,000 people arguing that standardized intelligence tests were poor measures of cognitive ability. The correlation between noneducational factors such as poverty with high test scores has also long been known, with one Washington State University survey finding that 80 percent of the variance in college admissions test scores could be accounted for by parental income.

There are other dangers in overestimating what video games teach students, and in using them to prop up a failing approach to education through quantifiable student evaluations. Video games are essentially mathematical simulations that nevertheless cover up their algorithms and equations. The most popular ones give their players an intuitional sense of a set of mathematic relationships — how long you have to hold down the run button to change the arc of Mario’s jump, or how many coal mines you can build in a town before your citizens start to turn on you. They simplify the world in deceptively unreal ways; to use them as educational tools is to teach students to engage with the world as a series of artificial systems.

None of this is to say video games should be banished from education. But as they become increasingly popular among educators, it’s important to think of them as abstract systems and not real-world problem simulators. In the same way that it would be absurd to teach children about French history by having them all try and paint Jean-Louis David’s “Napoleon Crossing the Alps,” using a SimCity to teach children about the socio-political realities of pollution confuses a depiction of a system for the larger system that houses it.

In “Education for a New World,” the Italian physician and educator Maria Montessori described education as “a natural process spontaneously carried out by the human individual … acquired not by listening to words but by experiences upon the environment.” As artificial systems, video games are ultimately just another way of making kids listen to an authority, though they are more tolerable because they come with colorful visuals and seem to respond directly to inputs. Video games are beautiful, provocative and transfixing, but they are also isolating, artificially simplistic and too easily drawn into the matrix of standardized education.

Video games have a place in schools as examples of what can be done through computer code and how computers can be played with, but their capacity to teach players about things that are not games is as limited as that of any other art form. Poetry can describe math in a beautifully simple way, but you wouldn’t teach calculus with a poem. Likewise, a video game can only ever euphemize its subject, be it urban planning or physics. The pleasure of seeing the world’s most complicated phenomena reduced to a simple series of interactive parts is what makes games so entertaining in the first place.

Games lie to us, making the unmanageable easy to control. In the same way that standardized tests reduce the vastness of the human intellect to narrow sets of data, games reduce the world to a series of inputs and outputs. These are wonderful diversions, but we risk oversimplifying the world by training our children in euphemisms, just so the system can keep better track of their progress.