IET Learning Analytics Workshop (15-05-14)

Today I attended and presented at an OU internal Learning Analytics workshop organised by my institute – the Institute of Educational Technology, at the Open University, UK.  This blog post is my notes from that event.

Eileen Scanlon

3rd in this workshop series – IET researchers joined by those from KMi and visitors from University of Amsterdam

Starfish – Networked Knowledge Sharing
Sander Latour and Natasa Brouwer (University of Amsterdam)

  • Starfish – finding a better way to disseminate best practice
  • (Video)
  • Community driven network
  • TPACK (Model of teaching good practice and technology) based labels.  See:
  • Entities linked into a network – aid to exploration
  • c.f. Google+ Communities
  • First working system in place – building network with other faculties/institutions
  • Potential for EU project
  • Research topics –
    • Dealing with difference in vocabulary
    • Effective search and exploration
    • Evaluating quality of information
    • SNA / Expert finding
    • Effect on Teachers of TPACK beliefs (Mishra and Koehler 2006) – e.g. technology used effects how we teach

Analytics insights into Epistemic Commitments in collaborative information seeking
Simon Knight, KMi

  • Link to epistemic games group in Maddison
  • Evolution in forms of assessment – moving away from pure summative towards performance assessment
  • Epistemic beliefs – a lens on learners views (Broome, 2009)
  • Removing the thought element – not decontextualised beliefs but situated and contextualised
  • Moving away from psychometrics
  • In “Search”
    • selecting sources
    • collating information
    • etc.
  • Epistemic commitments
    • The connections people make
    • Certainty characterised as …
  • Surface answers or deeper reasoning – use of search results
    • E.g. question on Marie Currie
  • Epistemic Frames for Epistemic Commitments
    • Views on how we see the world
    • described as skills, knowledge, values, identities, and epistemological rules
    • Discourse orientated
    • E.g. “we should try looking on Wikipedia”
      • select token, make connections
    • Epistemic Frames allows classification of activities in this process
  • Epistemic Network Analysis
    • Edge – indicates communication between nodes – edge gets thicker in proportion to level of discourse between the nodes
    • Move from log data – to keywords and concepts
    • Some maths happens – stanzas – don’t worry about how many times word occurs but how sourced
    • Principle Component Analysis (PCA) across stanzas
    • As analysis builds up some nodes become more central
    • Used a pair and two trios of 11-year-olds
    • Hypothesis – collaboration might be linked to number of sources researched
    • Some questions in exercise open – some closed – students asked to justify their answers and cite sources
    • Differences in groups
      • G1 – Successful “it has got all the important information” – i.e. less sense making more whether source had answers to questions
      • G2 – Also stressful but used different strategies and discourse was thus distinct from G1 (talked a lot about authority)
      • G3 – Poor results so discourse related to this
  • Claim that ENA offers a representational tools and can be used for hypothesis testing


Learning Analytics approaches to target support for disabled students in particular and to identify accessibility deficits in teaching and learning presented on the VLE
Martyn Cooper, IET

My presentation so not noted but slides on SlideShare at:

Learning Analytics for Academic Writing
Duygu Simsek, KMi

  • Machine code to identify good attributes in academic writing
  • How use this to support students and academics
  • Academic Writing –
    • Critical for students
    • Need to communicate validity of claims of automatic system
    • Meta-discourse analysis
    • Students find it challenging to learn academic writing but also find it difficult to understand meta-discourse cues
  • XIP – Xerox Incremental Parser uses NLP
    • Pulls out key features in academic writing
    • XML format output but not suitable for the learners
  • What are key features in student writing:
    • relevance
    • demonstration of knowledge
    • linguistic quality
    • argumentation
    • etc.
  • Argumentation:
    • mapping between good academic writing and XIP rhetorical functions
  • XIP needs a learning analytics approach to be useful in this context
  • Research Questions:
    • To what degree can XIP be used to identify good academic writing practice?
    • In what way shouldXIP output be communicated to students and educators?
      • Used a dashboard in a pilot study
    • To what extent to students value this approach?

Six different learning analytics metrics, but which one(s) best predict performance
Bart Rienties, IET

  •  Simon Buckingham-Shum: “We should move towards depositional learning analytics”
  • Learner Data vs Learning Data
  • E.g. from footballer tracker data
  • Study at Maastricht of students on compulsory maths/stats course presented on Blackboard
    • deep learner vs surface learner
    • motivation
    • diagnostic pre-test
    • demographic data
    • Blackboard data
    • Results on quizzes
  • Research Question 1 – to what extent predict performance?
    • Level of clicking in VLE poor predictor
    • More sophisticated tools shown to be better predictors
    • Combined metrics better predictors
    • Best predictions of assessment is assessment itself – so predictions get better after initial assessment on course
  • Research Question 2 – When should “coaches” intervene?
    • After first test good prediction but too late to intervene?
  • Research Question 3 – Dispositions or Learning Analytics?
    • Dispositions combined with early assessment provides good early warnings
    • Dispositions can be changed – feedback


— Close, drinks (Dutch treat) and onward discussion —



Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s