2. What do we mean by ‘analytics’?
How can we classify different kinds of analytics?
What are the challenges we face in the LMS?
What is happening in the Moodle project (and
beyond) in relation to analytics?
This session
3. “…exploring the unique types of data that come from
educational settings, and using those methods to better
understand students.”
A broad definition
Source: Wikipedia entry on Educational Data Mining
6. Type Level/object Who benefits?
Learning
Analytics
Course level Learners, Academics
Faculty level Learners, Academics
Academic
Analytics
Institutional Administrators, Funders,
Marketing
Regional Funders, Administrators
National and
international
National governments,
Educational authorities
Multiple perspectives
Source: http://www.educause.edu/ero/article/penetrating-fog-analytics-learning-and-education
Penetrating the Fog: Analytics in Learning and Education – Phillip Long & George Siemens
7. Type Level/object Who benefits?
Learning
Analytics
Course level Learners, Academics
Faculty level Learners, Academics
Academic
Analytics
Institutional Administrators, Funders,
Marketing
Regional Funders, Administrators
National and
international
National governments,
Educational authorities
Multiple perspectives
Target Consumer
Source: http://www.educause.edu/ero/article/penetrating-fog-analytics-learning-and-education
Penetrating the Fog: Analytics in Learning and Education – Phillip Long & George Siemens
11. One extreme: LMS only data analysis
Simple to implement
Losing ‘market share’ of data over time leading to lower
reliability of analysis
Other extreme: Learning ecosystem
Complex – needs open standards model and external
repository
The holy grail of educational analytics?
Analytics Scope
12.
13. One extreme: Manual analysis
Run reports, analyse data, take actions
Other extreme: Automated analysis
Automated analysis against known metrics
Alerts & notifications to consumers
Analytics automation
20. Only one part of the story
Lecture capture? Library systems?
Virtual classroom? Social learning?
Logs in Moodle are good, but not
comprehensive
‘LMS only’ analysis problems
21. Data gets big(gish)
‘Typical’ Moodle uni generates
between 50,000,000 and
100,000,000 log records in a year
LMS only analysis problems
23. Solution part 1: abstraction
Log
store
Adapted from http://docs.moodle.org/dev/Logging_2
Log data
24. What they are saying:
SOLAR: “Development of a common language for
data exchange*” – first item on their roadmap
IMS: “The … ‘holy grail’ of data interoperability
is an agreed upon “learning/progress map” that
all tools and assessments could populate.**”
Solution part 2 - standards
* source: http://solaresearch.org/OpenLearningAnalytics.pdf
** source: http://www.imsglobal.org/blog/?p=258
25. Increase amount of data gathered in Moodle’s logs to
support more in-depth analytics.
Solution part 3 – more data!