Educational Analytics and Predictive Modeling for Students
1. It is time for
Educational Analytics.
DR. SREERAMA KV MURTHY
Talk given at eIndia Education Summit, 24th July 2013, Hyderabad International Convention Centre.
2. 1. Quality institutions
of the future must be
data-driven.
Good institutions must invest in
comprehensive data collection,
warehousing, reporting & analytics
infrastructures.
Near real time data needs to be made
available, to Heads, Faculty, Placement
team as well as Students.
Case Study (2012)Concept
Implementation Steps
Informed decision making & better
accountability at all levels, irrespective of size
of institution
Clear definition & tracking of performance
metrics
Expected Benefits
3. 2. Predictive
Analytics can help
students choose
the right courses /
programs.
Deploy smart counseling at upper primary
& secondary school levels
What programs suit your interests?
What programs are you likely to be
successful at?
Case Study (2012)Concept
Implementation Steps
Reduce dropout rates in subsequent years
Prevent issues of weak learning
Expected Benefits
4. 3. Learning impact
can & should be
measured, especially
for our teachers.
Integrate continuous measurement during
& after teacher training. Quantify learning
impact on the teacher.
At all levels: higher education, vocational
skills, K-12
Case Study (2012)Concept
Implementation Steps
Quality trickle-down
Cost-to-benefit measurement of funds
deployed for teacher training.
Expected Benefits
5. 4. Digital Learning allows
unprecedented
measurement, and hence
personalization.
Implement e-learning wherever feasible.
Deploy platforms that support extensive
data collection & reporting.
Aim for individualized, personalized
training.
Keep the learners continually in-the-know
on where they stand.
Case Study (2013)
Concept
Implementation Steps
Better learner engagement.
Accelerated learning.
Expected Benefits