Week 4 - Learning Theory

Bodong Chen
Feb 9, 2016

A SIG should be focused on a theme in the field, and a WG project is about addressing a problem.

They could definitely be related but different in nature.

• As an example, a SIG could be social network analytics in classrooms, and a related WG project could be designing an social network analytics tool for a class of students who are using Knowledge Forum to study science.
• In this case, the WG project should definitely rely heavily on the SIG of social network analytics but could/should be informed by other SIGs as well.

Course timeline

• Plan early
• Book an appointment w/ Bodong one week in advance to discuss and refine your plan
• Driven by you. Assessed by peers.

• Totally fine to work solo
• Make it useful to your research
• Make it useful for others (academics, practitioners, etc.)
• Consider a LAK17 submission

• a hunch or a guess
• a well-substantiated explanation

• A theory is not a collection of facts; rather, it explains facts
  • Theory as “a basket”
• A (adequate) theory should be accurate, consistent, broadly applicable, simple, and fruitful (Thomas Kuhn, 1977)
• Explanatory coherence (Thagard, 1989)

• theory as paradigm: frames research design
• theory as a “lens”: defines what/how we look
• theory as new knowledge: originates from our work

• Micro-level theory seeks to explain behaviour at the level of the individual or family environment (e.g., psychology, frustration, aggression, etc)
• Meso-level theory seeks to explain the interactions of micro-level organisations (e.g., social institutions, communities, etc.)
• Macro-level theory seeks to explain behaviour at the level of large groups of people (e.g. ethnicity, class, gender, etc.)

Credit: Dr. Christopher E. Sunday

• theory vs. facts
• theory vs. hypothesis
• theory vs. practice

• “The direction and focus that theory can provide while working with large quantities of data is unquestionable. … Thus the relationship between theory and research can be two-way.” (Anu)
• “However, I didn't see them mention anything about the data collection process and how important this aspect of research design is to learning analytics. … Ideally, the learning theory should be weaved through the design of the technology and the data collection for analytics.” (Nic)
• I totally agree with the problem … that “if we collect tens or hundreds of variables from millions of individuals, in the absence of theory, how does a researcher decide which ones to include in an analysis?” (Fan)

• “Saying that the approach to science based on the scientific method is becoming obsolete is demeaning all that the scientific community has done for data deluge to emerge.” “Again, science has made possible to think about throwing the numbers into powerful computer systems. Science has made possible to find patterns in a different way. This is how science evolves. Earth was flat once.” (Astrid)
• “why starting with data which one then explains via theories is somehow superior to starting with theories and then using data to determine how viable they are.” (Jenifer)

• “data can say many conflicting things simultaneously without any of those things being correct or incorrect on the whole. Different patterns might emerge based on the particular analysis used and interpretations will inevitably vary based on the viewer's theoretical paradigms.” (Jenifer)
• “I do not think data and [reasoning with] data can expalin all the things” (Fan)

• “My theory - is that we'll still need theories - they just might be used differently when large amounts of data are available.” (Kristina)
• “It makes me think of how significant computers are affecting how we approach everything. If the case is that we don't know what we don't know, how do we even know to ask a question or present a hypothesis? This type of approach helps identify the questions that we don’t even know to ask - and I think will be incredibly important as we begin to uncover how valuable this type of investigation can be.” (Sarah)

• Knight, S., Buckingham Shum, S., and Littleton, K. (2014). Epistemology, assessment, pedagogy: where learning meets analytics in the middle space. Journal of Learning Analytics, 1(2):23–47.
• (optional) Stahl, G. (2013). Learning across Levels. International Journal of Computer-Supported Collaborative Learning, 8(1):1–12.

• Week 7, 3/1, Vitomir Kovanovic, PhD Candidate, University of Edinburgh, MOOCs & use of LA for “social” learning
• Week 8, 3/8, Elle Yuan Wang, PhD Candidate in Cognitive/Learning Sciences, Columbia University, EDM-ish