Learning Analytics and Mediation of Collaborative Learning Processes (CSCL 2015)
1. Learning Analytics of and in
Mediational Processes of
Collaborative Learning
Dan Suthers
Alyssa Friend Wise
Betrand Schneider
David Williamson Shaffer
H. Ulrich Hoppe
George Siemens
2. The collection, analysis and reporting
of data traces related to learning in
order to understand, inform and
improve the process, outcomes and/or
environments in which it occurs
Learning Analytics
3. Rapidly growing interest in learning analytics by the
CSCL community
CSCL’15: 9 papers + 3 posters +
this invited sessions
CSCL’13: No mention
in program
A rose by any
other name…
Intrigue
Puzzlement
Hesitancy
Energy
4. Generation of insight through computational
analytic methods
Informing of human action and decision
making (“closing the loop” in a tighter cycle)
Development of indicators, models and data
representations
Some Critical Characteristics of
Learning Analytics
5. One Schematic of Learning Analytics
Adapted from Tyne (2015)
Data Access,
Capture &
Management
Analysis &
Creation of
Insight
Processes that
Impact Student
Learning &
Success
6. Learning Analytics and CSCL
Fundamental shared concern with learning
processes and innovating to improve them
Analytics can help uncover mediation of social
interactions by physical, digital + conceptual artifacts
- Identify patterns in “big” or “deep” data
As well, learning analytics themselves are material
that can further mediate these interactions
- Possibilities for data-informed practices
7. Focus of the Panel
Explore potential of learning analytics for generating understanding
of and participating in mediational processes in CSCL
Session Structure
Four illustrative examples of CSCL learning analytics projects
Comments and questions to the panel by George Siemens,
Founding President of SoLAR (Society for Learning Analytics Research)
Extended discussion based on questions from the audience, live +
via twitter #cscl2015 #learninganalytics
8. Mediation of Discussion Forum Activity by
“Messages”: A Learning Analytics Approach
Supported by the Social Sciences and Humanities Research Council of Canada
CSCL 2015 ∙ Gothenburg, Sweden
Alyssa Friend Wise
Simon Fraser University Vancouver, Canada
With grateful thanks to the entire
e-listening research team:
Trisha, Hsiao, Farshid Marbouti,
Jennifer Speer, Yuting Zhao, Simone
Hausknecht & Nishan Perera
9. Origins of Online “Listening”
• From a social constructivist perspective, the goal of
online discussions is for learners to build understanding
through dialoging with other. At a basic level this involves
Externalizing one’s ideas
by contributing msgs to
an online discussion
Taking in the externalizations
of others by accessing
existing msgs
The messages are thus conceptual and
interactional resources that mediate the
process of discussing online
10. Distinct Characteristics of Listening Online
• Listeners (rather than speakers) determine timeline
by which messages and ideas are accessed
• Large decision space
– Frequency and length of log-in sessions
– Which posts attended to, in what order, for how long
– Revisit posts as many times as needed
– Reply when ready, unlimited time to prepare
11. Not Just the Messages, but their
Presentation also Mediates Interaction
11
12. Microanalytic Case Studies of Listening
Date Time Session Action Duration
(min)
Length
(words)
Message #
6/3/2011 23:46 1 Read 44.43 413 447
6/3/2011 23:52 1 Read 1.73 60 455
6/4/2011 00:08 1 Scan 0.23 117 459
6/4/2011 00:09 1 Read 12.51 413 460
6/4/2011 23:49 2 Post 3.18 120 477
Dynamic Discussion
Map: A record of the
discussion to show the
historical appearance
of the discussion forum
at any point in time
Log-file Data
of Student
Actions
So how do we study this?
13. Common Online Listening Patterns
Pattern Characteristic Behaviors
Disregardful
Minimal attention to others’ posts (few posts viewed; short
time viewing). Brief and relatively infrequent sessions of
activity in discussions.
Coverage
Views a large proportion of others’ posts, but spends little
time attending to them (often only scanning the contents).
Short but frequent sessions of activity, focusing primarily on
new posts. *May be socially-oriented or content-driven.
Focused
Views a limited number of others’ posts, but spends
substantial time attending to them. Few extended sessions of
activity in discussions.
Thorough
Views a large proportion of other’s posts; spends substantial
time attending to many of them. Long overall time spent
listening; considerable revisitiation of posts already read.
14. Developing Metrics for the Patterns
14
Dimension Metric Definition
Listening
Breadth
% of posts viewed
Number of unique posts that a student viewed divided
by the total number of posts made by others.
% of posts read
Number of unique posts that a student read divided by
the total number of posts made by others.
Listening
Depth
% of real reads
Number of times a student viewed other’s posts that
were slower than 6.5 words per second, divided by the
total number of views.
Av. length of real
reads
Total time a student spent reading posts, divided by the
number of reads.
Listening
Reflectivity
# of reviews of
own posts
Number of times a student revisited posts that he/she
had made previously in the discussion
# of reviews of
others posts
Number of times a student revisited others’ posts that
he/she had viewed previously in the discussion
16. Connections with Speaking
• Greater revisitation of others’ posts is
associated with richer responsiveness
• Greater listening depth (% of real reads) is
associated with richer argumentation
• Initially no relationship found between
listening breadth and quality of speaking
18. Designing Learning Analytics to Mediate
Learner’s Interactions w/ Messages
Extracted Analytics
Metric Your Data
(Week X)
Class
Average
(Week X)
% of posts read 72% 87%
% of real reads
41% 66%
Av. length of
real reads
2.37m 4.12m
#of reviews of
own posts
22 13
#of reviews of
others’ posts
8 112
19. Summary
The E-Listening Project as a brief illustration of how:
1. Analytics helped uncover mediation of
discussion forum activity by “messages” as
conceptual / interactional resources
2. This information could then be used to create
material that could further mediate these
interactions
20. Alyssa Friend Wise
Simon Fraser University
afw3@sfu.ca
@alywise
www.sfu.ca/~afw3/research/e-listening
Karrie Karahalio’s keynote yesterday showing the variety of ways in which she visualizes learners collaborative processes and showing them back to them seemed very much to be learning analytics
How analytics helped uncover mediation of discussion forum activity by “messages” as conceptual / interactional resources
This information could then be used to create new material that could further mediate these interactions
- But while one half of the process has been extensively studied, the other is often simply assumed to generally occur