Abstract
Learning a foreign language is important, it improves connection between groups of people, gives a greater understanding into another culture, and improves access to economic and social opportunities, especially through immigration. However language learning is an inherently difficult process, so many, especially those who cannot afford a language course or tutor, turn to apps to help guide them. However the most popular foreign language learning apps are deceptive, and fail to effectively teach their users. This creates a linguistic gap between those who can afford tutors and courses, and those who either rely on these apps or who try to learn on their own. This means that the poorer learners, the ones whose lives often could be vastly improved by learning a foreign language, don’t have access to an effective tool to learn. In my STS research paper, I explore why these language learning applications are ineffective, how they deceive their users, and what alternatives there are to popular apps like Duolingo. Then in my technical paper I explore how an effective app might be constructed to guide users along an effective path without limiting their progress.
For my STS paper I analyzed the issue that popular language learning apps have hundreds of millions of users despite the fact that these apps tend to be ineffective at teaching a language. This is an important problem, as these apps are seen as the standard way to self learn a language by many, and so tends to cause users to waste time, effort, and potentially money using these apps with little reward. This is especially important for those who need to learn a language quickly, for example immigrants or those who need it for a job. My research question is this: How do these apps continue to successfully lure users into ineffective learning with a lack of transparency in methods. In my STS paper I first analyzed academic research done on effectiveness and gamification in language learning apps, afterwards moving to an autoethnographic analysis of Duolingo, the most popular language learning application. Through this analysis I show that popular apps like Duolingo are not only ineffective, but can be detrimental to learning and users’ mental health. Then I conclude by analyzing different emerging apps and showing how these apps provide a transparent and effective approach. Though these apps give some hope, they are ultimately overshadowed in revenue and customers by Duolingo and other popular apps, which leads to a new question: what can be done to overturn the dominance that Duolingo and other popular apps have on the language learning application industry?
My technical paper is centered around the same problem as my STS paper, but instead of analyzing why the problem persists, I instead design an app that could solve the issue. I do this first by analyzing language learning frameworks and features in current apps that fail/benefit users. In my analysis I discovered that while apps like Duolingo and Babbel fail to use effective methods, their course structure can relieve stress and confusion for beginners and those who are less knowledgeable on creating a language learning routine. Similarly apps like LingQ use effective methods and give users a lot of freedom on what to learn, but can be confusing for beginners and those who want a more structured approach. So my design focuses on maximizing target language content like LingQ, while giving the user a course structured like a tree, where it is more rigid at the beginning, but allows users to branch out into what they want to learn about as they become skilled.
A very clear next step would be to prototype the design from my technical, and test if the design truly does improve learning. I think it is difficult to consider contributions from my technical without implementing it first. But overall, showing why these apps are ineffective, and giving alternative approaches for how to improve provides transparency to those who may not know whether popular apps are effective or not.
I would like to acknowledge the help of Professor Caitlin Wylie with my STS paper, especially on figuring out the overall problem. I would also like to thank professor Rosanne Vrugtman for helping with my technical, and my fellow students who helped me fix my paper during peer reviews.
