A Discussion of Learning Analytics on Big Data 基于大数据的学习分析研究
1. A Discussion of Learning Analytics on Big Data
基于大数据的学习分析研究
THEOL
Dr. Cheng, Jiangang 程建钢
Dr. Mo, Yun 莫 昀
Tsinghua University, Beijing
2. About Us - Educational Technology Institute (THETI)
Background
• We have been focusing on exploring the potential of information and
communication technologies (ICT) in learning, teaching and administration in
higher education for 14 years.
• We offer master’s and doctoral (PhD and EdD) degree for graduate students.
• We developed a series of software products for e-learning and e-management
which are widely used in China.
3. Research Team
Faculty, research fellow,
and graduate assistants (30)
Development and
programming (30)
Support and service (40)
400-500 School clients
5million Users
Background
4. Statistics for TsingHua Education OnLine (THEOL):
Partners: more than 400 universities and community colleges;
5,000,000 users (student/instructor/administrator)
500,000 online courses have been established based on THEOL platform;
More than 1,000,000 students and instructors visit THEOL every day;
More than 10,000,000 logs are saved everyday.
Big data spans four dimensions: Volume, Velocity, Variety, and Veracity (IBM)
Data in THEOL
8. Big Data Analysis based on THEOL
Current research data based on
Student-, teacher- and institutional level information from MIS;
Teaching and learning activities from e-learning platform;
Rating instruction/evaluation information from e-assessment system;
More than 10 years longitudinal data.
9. Big Data Analysis based on THEOL
Example:
Teachers’
preparation for
class
Teachers’
instruction
Students’
self reading
Students
learning
Student-teacher
interaction
# of Log in times;
# of posting notice;
# of posting documents.
# of Log in times;
# of reading notice;
# of reading documents.
# of adding FAQs;
# of quiz;
# of assignments.
# of taking survey;
# of taking notes;
# of submitting assignment.
# of students
asking questions;
# of discussions;
10. Type of analytics Level or object of analysis We are doing now
Learning
analytics
Personal level: analytics on personal
performance in relation to learning
goals, learning resources, and study habits of
other classmates.
Course-level: social networks, conceptual
development, discourse analysis, “intelligent
curriculum”
Designing a new Learning
analytics system
Departmental: predictive modeling,
patterns of success/failure
Designing E-portfolios
Academic
Analytics
Institutional: learner profiles, performance
of academics, knowledge flow, resource
allocation Multiple comparisons based
on THEOL series softwareRegional (state/provincial): comparisons
between systems, Quality and standards
National & International
A New Design of Learning Analytics System
Learning and Academic Analytics (George Siemens and Phil Long)
11. Learning Analytics System
Data Collection
From THEOL series
software
Learning Analytics Output
Students’activities
Instructional activities
Course materials
……
Tables
Graphs
……
Database
Analytical
Tools
Framework