Exploring Characteristics of Fine-Grained Behaviors of Learning Mathematics in Tablet-Based E-Learning Activities

Attributes of teaching and learning contexts provide rich information about how students participate in learning activities. By tracking and analyzing snapshots of these attributes captured continuously throughout the duration of the learning activities, teachers can identify individual students who...

Full description

Saved in:
Bibliographic Details
Published inInternational Association for Development of the Information Society
Main Authors Yeung, Cheuk Yu, Shum, Kam Hong, Hui, Lucas Chi Kwong, Chu, Samuel Kai Wah, Chan, Tsing Yun, Kuo, Yung Nin, Ng, Yee Ling
Format Report
LanguageEnglish
Published International Association for the Development of the Information Society 01.07.2017
Subjects
Classification
Computer Oriented Programs
Data Analysis
Data Collection
Educational Technology
Elementary School Students
Handheld Devices
Mathematics Achievement
Mathematics Instruction
Problem Solving
Student Attitudes
Student Behavior
Student Surveys
Teaching Methods
Technology Uses in Education
Telecommunications
Online AccessGet full text

Cover

More Information
Summary:Attributes of teaching and learning contexts provide rich information about how students participate in learning activities. By tracking and analyzing snapshots of these attributes captured continuously throughout the duration of the learning activities, teachers can identify individual students who need special attention and apply different pedagogical actions to them. This paper describes the results of the work-in-progress study in exploring characteristics of fine-grained behaviors of learning mathematics in tablet-based e-learning activities. An experimental platform called SkyApp is built. Through SkyApp, teachers can create e-learning activities and track learning records of students after the delivery of the activities. SkyApp supports capturing, storing and analyzing of fine-grained behaviors of students. Pilot tests have been done in two primary schools for eight months. The review of the tests demonstrates the potential in performing learning analytics. By applying clustering algorithms on multiple learning metrics of marks, time and number of attempts for students in solving mathematics questions, classification of students by learning characteristics of performance and engagement can be formulated. [For the complete proceedings, see ED579335.]