Learning Analytics: Today, Tomorrow, and When We Get Flying Cars
Megan Bowe@meganbowe
Learning Analytics, the definition(s)
1. Learning analytics is the measurement, collection, analysis and reporting of data about learners and their contexts, for purposes of understanding and optimizing learning and the environments in which it occurs.
2. Learning Analytics use techniques from information science, sociology, psychology, statistics, machine learning, and data mining used to analyze data collected during education administration and services, teaching and learning to create applications that directly influence educational practice.
1. https://en.wikipedia.org/wiki/Learning_analytics 2. https://sites.google.com/a/umail.iu.edu/iuncc/whatis
Educational Data Mining
Developing methods and using techniques from statistics, machine learning, and data mining to analyze data collected during teaching and learning (e.g., from a courseware platform) to test learning theories and inform educational practice
https://sites.google.com/a/umail.iu.edu/iuncc/whatis
So. Many. Words.
• Data mining is digging through data sets to clean up and find patterns
• Analytics take those patterns into reality to find meaning and give insights
Great! How do we do this?
My students use X and I want to
know Y
We’ve been using this LMS and SIS for
decades, there must be something to learn in
there The most business critical issue is student
retention, tell me who is at risk of dropping
outDefine a goal, then…
Uh, thanks Internet. This is totally clear now…
Better.by Timothy Hartfield http://timothyharfield.com/blog/2014/09/11/learning-analytics-what-is-it-why-do-it-and-how/
Define your data sources
Activity 1 Activity 3 Activity 5
Learning experience designActivity 2 Activity 4 Activity 6
Get all of the data together
Data Warehouse? Data Mart?
Database? Learning Record Store?
Dig around, mine if you must.
• What activity creates what data?
• Is there an ideal path through the activities?
• What data or sequences are missing or unexpected?
• What clean-up work needs to be done?
Design ways to show others your insights! How do people use pdfs?
Design ways to show others your insights! How do people use pdfs?
Today, most of us have an LMS (like it or not)
• Grades
• Scores
• Duration
• Logins
• Page/Course access
Tomorrow, we get more holistic views
LMS Flashcard app Youtube
Class Attendance
Report!
eBook Degree Progress
Standards and standard units of measurement
The Experience API and matching equivalent data
Past Present Future
InformationWhat happened?
(Reporting)
What is happening now?
(Alerts)
What will happen?
(Extrapolation)
Insight
How and why did it happen?
(Modeling and experiment
design)
What’s the next best action?
(Recommendations)
What could happen?
(Prediction, optimization, simulation)
Davenport et al “Analytics at Work”
…and the flying cars.
(it gets creepy)
–Tony Shan
“All in all, Fast Data, Actionable Data, Relevant Data, and Smart Data (FARS) are well poised
today to replace Big Data for the new paradigm.”
“Big Data is Really Dead”
• Fast Data - processing a large amount of data in real time to find immediate insights and patterns
• Actionable Data - a combination of predictive analytics and hypotheses to make recommendations and use feedback to make decisions
• Relevant Data - relationships in data help to identify patterns that seem unrelated on the surface
• Smart Data - using meaning and algorithms to make predictions and support decision making