Bodong Chen, Assistant Professor
University of Minnesota
February 11, 2015

• Beijing Normal / Peking U / U of Toronto alumnus
• Joined LT in January 2015
• Research areas
  • knowledge building
  • learning analytics
  • online learning (MOOCs)
  • social media
• (An R hacker + a fan of reproducible research)

Since learning is plural and ubiquitous, we need to:

• understand learning needs – societal, organizational, personal
• understand learners – background, beliefs, values, dispositions…
• understand learning when it occurs – what, when, how, who…
• impact/empower learning – “feedback loops”, engage stakeholders
• …

To bring technical, pedagogical, and social domains into dialogue with each other

• computer scientists, machine learning experts, statisticians, and mathematicians had the technical capacity to make sense of large data sets, but lacked grounding in education and learning theory and literature
• learning scientists, psychologists, and sociologists had the theoretical lens to evaluate the social power structures and “soft domains” of learning, but lacked grounding in emerging data and analytics methods

• 2011 – First academic conference, Banff, Canada
• 2011 – Inception of Society for Learning Analytics Research
• 2013 – First summer institute, Stanford University
• 2014 – First issue of Journal of Learning Analytics

Analysis of the Community of Learning Analytics (Nawaz et al., 2013)

Twitter community flow (Chen et al., 2015)

Topic modeling of LAK/EDM publications (Sharkey et al., 2014)

Twitter community topics (Chen et al., 2015)

Learning Analytics vs. analysis of learning data

Learning Analytics vs. Educational Data Mining (EDM)

Learning Analytics vs. Academic Analytics

• LA involves (not necessarily big) digital data
• LA involves (not always complex) automated analytical processes
• LA informs / empowers learning and teaching
• (good) LA has a clear sense of what 'learning' means
• (good) LA is aware of its pedagogical and epistemological assumptions

• Analytics around social interactions
• Analytics around learning content
• Analytics of learning pathways
• Analytics in different spaces (digital/F2F)
• Analytics on interaction with the learning system (university/K-12/corporate)
• Analytics on intervention and adaptation
• Evaluation of analytics

Five primary areas of analysis (Baker & Siemens, in press; Baker and Yacef, 2009):

• Prediction
• Structure discovery
• Relationship mining
• Distillation of data for human judgment
• Discovery with models

Five areas of application (Bienkowski, Feng, and Means, 2012):

• Modeling user knowledge, behavior, and experience
• Creating profiles of users
• Modeling knowledge domains
• Trend analysis
• Personalization and adaptation

• Hidden assumptions (Knight, et al., 2013)
• Pedagogical interventions (Wise, 2014)
  1. Integration
  2. Agency
  3. Reference frame
  4. Dialogue
• Transparency in learning analytics practice
• Ethical practice

Learning Analytics is For Learners & Learning!

To advance as a field, LA researchers and practitioners need to address the following:

1. development of new tools, techniques, and people;
2. data: openness, ethics, and scope;
3. target of analytics activity; and
4. connections to related fields and practitioners.

• Attend LAK conferences (online sessions)
• Apply for LASI – capacity building for the field
• Join online conversations (e.g., Twitter, Google Group)
• More importantly, consider integrating LA in your work!

Guest speakers in my LA course:

• Simon Knight, Feb 12
• George Karypis & Jeff Gran, Feb 26
• Stian Haklev, Mar 5