Generalizable prediction of academic performance from short texts on social media

TL;DR

This study demonstrates that short social media texts can be used to predict students' academic performance across different platforms, leveraging large-scale language models for generalizable and interpretable predictions.

Contribution

It introduces a model that predicts academic performance from short texts and shows its applicability across different social media platforms, enhancing generalizability.

Findings

Model can predict academic performance from VK posts.

Model can reproduce school and university rankings.

Model generalizes across Twitter and VK data.

Abstract

It has already been established that digital traces can be used to predict various human attributes. In most cases, however, predictive models rely on features that are specific to a particular source of digital trace data. In contrast, short texts written by users $-$ tweets, posts, or comments $-$ are ubiquitous across multiple platforms. In this paper, we explore the predictive power of short texts with respect to the academic performance of their authors. We use data from a representative panel of Russian students that includes information about their educational outcomes and activity on a popular networking site, VK. We build a model to predict academic performance from users' posts on VK and then apply it to a different context. In particular, we show that the model could reproduce rankings of schools and universities from the posts of their students on social media. We also find that the same model could predict academic performance from tweets as well as from VK posts. The generalizability of a model trained on a relatively small data set could be explained by the use of continuous word representations trained on a much larger corpus of social media posts. This also allows for greater interpretability of model predictions.