columbia tc
DESCRIPTION
Presented to Teachers College, Columbia UTRANSCRIPT
The Structure and Logic of the Learning Analytics
FieldGeorge Siemens, PhD
January 9, 2013Teachers College
Columbia UniversityNew York
- Publicly funded research university- 38,000 students- One of four research universities in Alberta- Only US accredited Canadian university (MSCHE)- Bachelor, masters, doctoral degrees- Fully online
My Work
1. Social network analysis and concept/knowledge development (NSERC/D2L, with Gasevic, Dawson, Haythornthwaite)
2. Knowledge mapping and competencies
3. Structure/Logic of LA: epistemology/assumptions/trends/deployment/impact
4. Developing post-baccalaureate (Masters, 2014, fingers crossed) on data analytics
Previous EdLab sessions
BrusilovskiBakerWoolfStamper (future)
Structure of Learning Analytics Field
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.
Scope of focus is important factor
EDM: specific variables and factors
LA: systemic and in-context factors
(but this distinction is not hard; overlap and blurring is occurring)
9
Domains of LA activity & impact
Learning & knowledge growthNetwork analysis – social and knowledgeContent analysisPersonalization and adaptationPrediction & InterventionSystemic impact
Inter-disciplinary emphasis
Bringing technical, pedagogical, and social domains into dialogue with each other.
Analytics around social interactions
Analytics around learning content
Analytics in different spaces (digital/F2F)
Analytics on interaction with the learning system (university/k-12)
Analytics on intervention and adaptation
Assessment of analytics
Siemens, Long, 2011. EDUCAUSE Review
What is happening globally in LA
Africa: end userAustralia: end user, gov’t, networkChina: end userEurope/UK: end user, gov’t, networkHong Kong: end user, gov’tIndia: end user, networkLatin America: end userUSA/Canada: end user, gov’t, network
http://lakconference.org
literature and research in three primary domains: networks and social media, learning analytics and datamining, and the future of learning and learning institutions.
Open Learning Analytics
Trends (?) in literature
Xavier Ochoa
Logic of LA field
Logic of analytics
Sensemaking and wayfinding
Comp683, Stat110 (Blitzstein)
What is research/science?
Essentially, discovery (identification) of connections
Validity of connection interrogation techniques
(Guba & Lincoln, 1994, 2005)
The research model in learning analytics– Holistic, not reductionist– Focused on systemic level– Impact and in-context evaluation– Social, technical, pedagogical
An EDM paper (best paper 2012 conf)
Terms: Student model, CTA, cognitive tutor, instructional techniques
Goal: automated technique for the discovery of better student models using input from previously generated models.
Data: DataShop, 300 datasets, 70 million student actions
Methodology: Knowledge components mapped to instructional tasks, LFA algorithm to find better models
A LAK paper (2012 conf)
Terms: learning sciences, engagement, design, learning ‘power’, competencies
Goals: “a learning analytics infrastructure for gathering data at scale, managing stakeholder permissions, the range of analytics that it supports from real time summaries to exploratory research, and a particular visual analytic which has been shown to have demonstrable impact on learners.
Data: survey data, Learning Warehouse, >40,000. Not auto-tracked, instead: learner self-disclosing
Methodology: ELLI visualization and validation using Learning Warehouse (ELLI intends to “provide educators with a practical tool to enable rapid assessment and intervention of a complex quality, to stimulate change in learners”)
A LAK paper (2012 conf)
Terms: networks, community structure, SNA, distributed learning
Goals: Resolving: “multiple means of participation, each with their own Interactional and social affordances…Events in these media may be logged in different formats and databases, disassociating actions that for participants were part of a single unified activity.”
Methodology: “developed an abstract transcript representation called the Entity-Event-Contingency (EEC) graph that provides a unified analytic artifact [33]; and an analytic hierarchy derived from the EEC that supports multiple levels of analysis”
Data: Participant interaction in TappedIn, 150k members
Challenges
Scope of data captureIdentifying critical elements that matterGenerating multi-dimensional models
“Learning and knowledge creation is often distributed across multiple media and sites in networked environments. Traces of such activity may be fragmented across multiple logs and may not match analytic needs. As a result, the coherence of distributed interaction and emergent phenomena are analytically cloaked”
Suthers, Rosen, 2011
http://mashe.hawksey.info/2012/11/cfhe12-analysis-summary-of-twitter-activity/
Systemic: Impact and deployment
gsiemens @gmailTwitterSkypeFBWherever
www.elearnspace.org
www.connectivism.ca
www.learninganalytics.net