Data Analytics in Web-based Education in the Higher-education Classroom

Abstract

Attention span of students in a classroom is very short. To overcome this, different active learning methodologies have been used in the past. Active learning keeps the students busy and engaged throughout the lecture. It breaks the lecture into certain time intervals by intermixing breaks, demonstrations and questions after each interval. For using active learning, clickers and laptops are commonly used in higher education classroom. Most experiments in higher education classroom studying different characteristics of students like learning performance and attention, use clickers and laptop. But, most of these experiments are in a controlled setting, not scalable and compromise the privacy of students. We overcome these problems in an active learning setup in the higher education classroom where we use a web-mediated teaching tool called ASQ. ASQ is a web application that helps to give presentation in a classroom where the presenter has control over the flow of the presentation. ASQ also allows the presenter to interleave the presentation with questions, videos and other interactive JavaScript components. Anyone can anonymously join a presentation in ASQ using a web browser. ASQ tracks the activity of every student interaction by generating event logs each second. In the previous work using ASQ, it has been shown that these logs could be used to infer the attention level of students in the classroom. The goal of this thesis is to gather insights about the fine-grained study behaviour of students in a higher education classroom by analyzing these event logs.

We investigate (i) the effect of lecture elements (like the difficulty, relative positioning and spacing of questions; and duration of discussion in the slides) on study behaviour (like attention level, performance and reaction time while answering questions) of students; (ii) the relationship that might exist between attention percentage of students and their participation in the in-class questions; (iii) if students are taking external help when answering questions during the lecture and the relationship that might exist between their tendency to take external help with the difficulty of questions. We conduct our study in a classroom of around 300 students, for 15 lectures in the Web and Database Technology course at TU Delft taught by 2 instructors. We find significant effect of (i) spacing of questions on reaction time and instructor on performance; (ii) length of discussion time associated with a slide on the attention level of students which agrees with past studies; (iii) relative positioning of questions on the performance of students. However, we do not find significant effect of difficulty of questions on performance and reaction time of students while answering these questions. We also find significant effect that students with more attention percentage participate more in the in-class questions. Finally, we find that students take external help while answering questions but the tendency to take external help does not depend on the difficulty of questions.