An introduction to the what, where, who, and what-for of Analytics

muscleblouseΤεχνίτη Νοημοσύνη και Ρομποτική

19 Οκτ 2013 (πριν από 4 χρόνια και 25 μέρες)

65 εμφανίσεις

An introduction to the what, where, who, and what
-
for of
Analytics

Contents

(pg 1 of 7)


What is “Analytics”



Where is CCCOnline in terms of
Learning Analytics?


What is the Desire2Learn Analytics
product? What can it actually do?


What have other
institutions done?
Where are other institutions going?

What are “Learning Analytics”

to us?


Analytics is processing data in some
fashion that will help us do our jobs as
administrators or instructors.


It is similar to and includes earlier
fields/fads, such as “educational data
mining”, but implies visualization of data
so as to be made more useful to faculty
and staff.


What is CCCOnline up to


Desire2Learn progress tracking


Faculty in
-
attendance alerts


Student no
-
show reports


Desire2Learn Analytics


Behavior analysis


D2L Progress Tool


Not graphical, all tables


D2L Analytics


Faculty Portal


What are my students doing at a glance?


Tool use


Grade patterns


Quiz Consistency Analysis

“Does my quiz measure just one thing?”

D2L Analytics Proper


D2L Analytics


data domains


Sessions


“When have they been in their
course?”


Tool use


“When did they go into the
discussions?”


Content access


“What have they read?”


Difficulties with content


Grades


Various
gradebook

designs


Quiz question grades


What are other institutions
doing?


What is out there that we want to achieve as
well?


Who is doing what?


Visualizing data


Standard reports
-

What happened?


Ad hoc reports
-

How many how often and were


Query/Drill down
-
Where exactly is the problem?



Alerts
-

What actions are needed?


Statistical
Analytiss

-

Why is this happening?


Forecasting/
Extrapoluation

-
What if these trends
continue?


Predictive Modeling
-

What will happen next?


Optimization
-

What’s the best that can happen?


Katholieke

Universiteit

Leuven

“Monitor Widget”


Visually compare your time in class or resources
accessed with your peers.


“Am I doing what I should be in order to be
successful?”



SNAPP

Universities of Queensland and Wollongong, Australia

University of British Columbia, Canada


University of Belgrade

“LOCO
-
Analyst”


Local
-
Analyst

Content Access & Analysis


Loco
-
Analyst

Social Network Analysis


Minnesota State College and Universities

“Accountability dashboard”


Predictive modeling

Signals


http://www.itap.purdue.edu/tlt/signals/sig
nals_final/index.htm


Signals illustrated


Signals Faculty Dashboard


Student success at a glance


Prepare and dispatch custom
intervention E
-
mails

American Public University
System


For profit university serving over 80k online
students.


Collects almost a hundred metrics based on
student demographics, prior grades, and
current course data.


Metrics are fed into a Neural Network that
compares the metrics to grades in previous
semesters, ranking the students from 1
-
80k in
their chances of success.


The user can drill down to find out exactly
what makes the network “think” a student will
fail.

Recommendation Engine


Fruanhofer

Insituttion

for Applied information
Technology at FIT


Domain Ontology


+ Usage patterns of prior users


+ Identifying feature of “this” user


a
search term, academic status, etc


= Recommended resources



Another example of a
recommendation engine…


Semantic Analysis

Open University, UK


Look into the content of posts to determine
what style of communication it is.


Challenges
eg

But if, have to respond, my view


Critiques
eg

However, I’m not sure, maybe


Discussion of resources
eg

Have you read, more
links


Evaluations
eg

Good example, good point


Explanations
eg

Means that, our goals


Explicit reasoning
eg

Next step, relates to, that’s why


Justifications
eg

I mean, we learned, we observed


Others’ perspectives
eg

Agree, here is another, take
your point

Ultimate Goal


Modeling/Predicting success


Staging the most effective interventions


Improving instructor abilities


Improving students’ self awareness


Customized learning


Learning Styles


Cognitive Load



The hierarchy of student success through Action
Analytics


Raising Awareness (Analytics IQ)


Data, Information, and Analytics Tools and Applications


Embedded Analytics in student success processes


Culture of performance measurement and improvement


Optimized student success


Dangers



“Analytics for learners rather than of
learners”
-

Dragan

Gasevic
,

Athabascau

U.


Trapping students into limiting models of
“good” behavior.


Disrupting and Transformative
Innovation


Institutions resist change