Rebecca Ferguson’s Overview of Learning Analytics (LA) H817 Activity 4.5
Thursday, 9 June 2016, 07:22
Visible to anyone in the world
Edited by Steve Bamlett, Sunday, 12 June 2016, 09:50
NOTES for later:
The key point in the introduction is that, although a lot of
effort can go into differentiating categories of things that are and are not
LA, the point is that these differentiated activities each appeal to the
interests of different kinds of stakeholder in the enterprise, each with a
distinct approach:
·business intelligence,
·web analytics,
·educational data mining (EDM), &
·recommender systems.
The point is to bring these distinct approaches, if not the original
stakeholders themselves together in some participatory manner. The paper
focuses on academic analytics and EDM as approaches requiring a rapprochement.
Driving Factors
1.Big Data: particularly turning the raw data
(over-abundant, lacking [evident] connectivity externally and internally) into
a visible usable form (into VALUE) by: a.Extracting from it, b.Aggregating parts of it, c.Reporting, d.Visualization(s) of it.
2.Optimising it to facilitate more effective online
learning.
3.Using it politically – measure, demonstrate
& improve performance. My feelings here (see Activity 4.2) are that there is a
danger in confuting ‘intelligence’ with political interests. Knowing about inequality
does not change it and may just validate it.
4.The role of stakeholder interest groups: a.Government b.Education providers c.Teachers / learners.
EDM – techniques in
computing:
·Decision tree construction
·Rule induction
·Artificial neural networks
·Instance-based learning
·Bayesian learning
·Logistical programming, &
·Statistical algorithms. (LOOK ALL THESE UP,
Steve)
Good quotations here on use for teachers & learners (Zaïane 2001) on EDM for PEAR.
Learning-focused perspectives – pedagogy.
Social Network
Analysis (SNA) – ‘considers knowledge to be constructed through social negotiation.’
Rebecca Ferguson’s Overview of Learning Analytics (LA) H817 Activity 4.5
NOTES for later:
The key point in the introduction is that, although a lot of effort can go into differentiating categories of things that are and are not LA, the point is that these differentiated activities each appeal to the interests of different kinds of stakeholder in the enterprise, each with a distinct approach:
The point is to bring these distinct approaches, if not the original stakeholders themselves together in some participatory manner. The paper focuses on academic analytics and EDM as approaches requiring a rapprochement.
Driving Factors
1. Big Data: particularly turning the raw data (over-abundant, lacking [evident] connectivity externally and internally) into a visible usable form (into VALUE) by:
a. Extracting from it,
b. Aggregating parts of it,
c. Reporting,
d. Visualization(s) of it.
2. Optimising it to facilitate more effective online learning.
3. Using it politically – measure, demonstrate & improve performance. My feelings here (see Activity 4.2) are that there is a danger in confuting ‘intelligence’ with political interests. Knowing about inequality does not change it and may just validate it.
4. The role of stakeholder interest groups:
a. Government
b. Education providers
c. Teachers / learners.
EDM – techniques in computing:
Good quotations here on use for teachers & learners (Zaïane 2001) on EDM for PEAR.
Learning-focused perspectives – pedagogy.
Social Network Analysis (SNA) – ‘considers knowledge to be constructed through social negotiation.’
Dawson (2008) – read.
Ethical issue - should students be told their activity is being tracked’
Uses in Learning are:
Challenges