Navigating the Nuances: Quantifying Linguistic Diversity in Online Tutoring

Online tutoring, a practice integral to educational adaptability during the post-pandemic era, hinges on effective tutor-student interactions. This study investigates how tutors in virtual settings develop and diversify their language over time, contributing to a growing body of research on interpersonal communication within educational environments. By applying computational language measures to a dataset of online elementary literacy tutoring sessions, we quantify the evolution of linguistic diversity among tutors as they gain experience. Our research examines the development of tutors' linguistic diversity both across and within sessions, and how this diversity relates to the content of their discourse. Analysis of over 4,000 tutoring sessions reveals that tutors not only adapt their language in response to varied student needs and educational contexts but also face constraints from standardized curricula and protocols. Our findings propose a nuanced understanding of how personalized linguistic strategies correlate with effective tutoring, underscoring the potential of computational analysis to enhance tutor training and educational quality.