Duolingo enters the testing business

Duolingo has announced they’re developing a $20 app-based English-proficiency exam to challenge big test makers such as TOEFL, according to the Wall Street Journal. The test will cost $20, and will use your camera or tablet’s front-facing camera to record sound and video during the exam to prevent cheating. Apparently, they want their test to be accepted at companies and universities as proof of language proficiency.

All the articles I’ve read about this focus on the innovative measures to prevent cheating. That’s fine. But it kind of seems to me that neither Duolingo chief Luis von Ahn, nor the journalists who’ve written about the announcement, are intimately familiar with the TOEFL, or language assessment. If they were, I think the more pressing question that would come to mind would be how a mobile-device based, crowd-sourced, grammar translation-driven language proficiency test is ever going to come close to being able to credibly test the range of skills that something like the TOEFL iBT can.

Is it going to test the productive skills? Production of anything more than sentences isn’t really in Duolingo’s repertoire, and even within the language learning app, there are big issues with divergence of answers. Will reading happen beyond the sentence level, and will listen comprehension be anything beyond a nonsensical computer-generated phrase? And communicative competence? That term doesn’t even exist in the world of Duolingo. A certificate of proficiency stating that “User X is fully able to match words to pictures” just won’t pack a lot of punch out there in the real world.

Perhaps they have something completely mind-blowing ready to unveil that’s going to be a sea change from the exercises and games that make up Duolingo’s language learning experience (which I have a lot of fun with in my studies of Portuguese, for what it’s worth). If so, then, bring it on! A free/cheap, accessible, widely-accepted and trusted international English proficiency test would be game-changing.

Don’t get me wrong; I’m no fan of the big business of the TOEFL/IELTS testing industrial complex and the way it has pulled the focus of much EAP teaching, for example, away from real language development toward acing the standardized exam. But the research and development that have gone into TOEFL and IELTS over the years is phenomenal, and as evil and profit-driven as they may be, they test reading, writing, speaking and listening in a generally valid and reliable manner. (Not without major limitations and heaps of criticism, of course! But nonetheless…)

As far as Duolingo producing a test on par with those the big boys that will be accepted by universities? I’ll believe it when I see it.