Commentary: Find out how artificial intelligence could help shape new, game-changing forms of student assessment.
One silver lining of COVID-19 is that it exposed failed legacy systems and processes, forcing organizations to digitally transform. One area that has benefited is education. There is a quiet revolution underway that is challenging assumptions we have held for more than a century. The success of so many EdTech companies over the past year shows us new ways to better address old challenges. Artificial intelligence, for example, is sometimes cast as a villain in education, displacing teachers. However, what seems to be happening instead is that AI is being used to free up teacher time to improve how they engage students.
Perhaps in no area of education is this better illustrated than testing.
The AI will test you now
I’ve written before about how Riiid Labs applies AI to testing. Riiid recently raised $175 million from Softbank and is making waves. I connected with the Silicon Valley-based CEO of Riiid Labs, David Yi, to learn more about how AI can improve testing. Like getting rid of standardized tests, for starters.
“Nobody likes standardized tests, even though most of us agree that we need some sort of objective measure of education,” Yi said. “But in the century since the practice was widely introduced to American schools, the standardized testing system has grown bloated and corrupt. People with money have long figured out how to game it.”
One of the outcomes of the coronavirus pandemic has been a pause in testing at many schools, and a number of prestigious American universities dropped standardized tests altogether, such as the ACT, GMAT and SAT. Yi argued this is a good thing. He said AI can do much better than standardized tests.
“Within ten minutes of interaction with an AI system, we can predict with over 90% accuracy a student’s standardized test score and know what she’s weak at and what she’s strong at,” Yi said. “We can predict what questions she’ll get wrong before she even tries to answer them. We can even predict when she is going to get tired and disengage.”
One application of AI could be to “pre-fail” students, similar to “pre-crime” in “Minority Report.” But this isn’t how Yi sees things going.
How AI can help students succeed
What people overlook in AI too often is that it can also personalize education at a granular, individual student level in a way that standardized textbooks, tests and teaching cannot. AI can optimize a learning plan for each student. That is a game changer.
“Suddenly, the assessment itself becomes formative,” Yi said. “It’s not just a conclusive, one-test value judgment of the student. It becomes a formative learning process, and we can micro assess her, with low stakes, multiple times throughout a school semester or school year. That’s an exciting proposition, but we thought it would take a decade before people were ready to accept this kind of methodology. COVID-19 has accelerated everything. Because people haven’t been able to gather at a physical location and everything’s asynchronized, we’re finding more people eager to adopt this mode of assessment.”
SEE: Artificial intelligence ethics policy (TechRepublic Premium)
If you have a system that a student’s interacting with on a daily or weekly basis, and that system can predict students’ scores at any point in time and can recommend the learning path they should follow for optimal results, it obviates the need for standardized testing, Yi argued. A standardized test is a snapshot that people cram to get ready for. They take the picture, and then there’s no follow-up. But if you have a system that’s continually evaluating the student, you don’t need that snapshot. You know at any point in time where they are and what the likely outcome is.
What changes is the purpose or the intent behind the assessment. Instead of, “Hey, you learned it” or “You didn’t learn it; you’re a failure” or “You’re great,” AI helps teachers assess the student to help her learn better. In this way, it’s not a value judgment about her in the way test scores can be. When you’re micro assessing, you’re saying, “Here’s where you are. Here’s what you’re weak at. And here’s what you need to improve.” This approach has the potential to completely change how we approach education.
“If we, as a community, society, and nation agree that the learning process is valuable and that it says something about the student’s final outcome, we should be able to replace that final summative test score with this formative assessment,” Yi said. AI enables that.
Disclosure: I work for AWS, but the views expressed herein are mine.