Innovations in Researching Regulation of Learning in Solo and Collaborative Tasks

session on SRL last year

Ludo Van Meeuwen, The Open University of the Netherlands; Saskia Brand-Gruwel, The Open University of the Netherlands; Paul A. Kirschner, Open University of the Netherlands; Jeano de Bock, Air-Traffic Control, the Netherlands; Jeroen J.G. Van Merrienboer, Maastricht University

complex and dynamic

• mental picture
• chunking
• decision of order
• r/t, labeling

tree phases (Schunk & Zimmerman, 1998)

Forethought phase

• goal setting
• strategic planning
• self-efficacy
• intrinsic interest

Performance phase

• self-observation
  • monitor task performance
• self control
  • adjust task strategies

Self-reflection phase

• self-judgmented

Self-efficacy, sense of agency to motivate, and self-directed learning skills

Question: Relation SRL-skills, SDL-skills and SE-beliefs.

Goal: measurement of SRL minimally influencing learning process

• use of cued retrospective reporting

Procedure

• learning task given
• eye-movements recorded
• saw eye-movements back (slowed down)
  • opportunity to say what he thought at the time

RQs:

• what skills can be identified using cued retrospective report?
• how are skills/believes connected to complex task performance?

Method

• training task
• training assessment 1 (2 weeks)
• training assessment 2 (6 weeks)

Self-efficacy: how well do you think you will do?

Analysis

• coding
  • forethought phase
    • orientation and planning
  • performance control phase
    • monitoring and adjusting
  • self-reflection phase
    • evaluation in action, on action, on learning

Findings

• some correlations between coding scheme and dependent variable

Roger Azevedo, McGill University; Reza Feyzi Behnagh, McGill University; Jason Matthew Harley, McGill University; François Bouchet, McGill University

Srl, Co-RL, and SSRL (Hadwin et al 2011) - handbook of SRL, read this!

in the literature “extended” SRL models (Dismoore et al 2008, Saab, Hadwin et al 2011).

Use of artificial pedagogical agents

MetaTutor

• detect model and trace scaffold cognitive and metacognitive self-regulatory processes
• avatars
• four pedagogical agents, one human
  • for different kind of processes (students learn to associate them)
• different layouts - affords different self-regulatory processes
• agents pop up in the box
• self-regulation palet
• indicate that you want to summarize, or click on summary
• overall learning goal, sub-goals, pages

Assumptions (lots)

• cognitive-affective … CAMM

Goal

• having agents as self-regulating systems that learn from their interactions with student (and results)

Method

• screen capture
• eye tracking
• video coding for emotion
• concurrent think aloud
• quizzes
• user-generated summaries
• note-taking behavior

(hard to temporally align these - across time and across multiple data channels)

Concepts and theories

• what are defining criteria that differentiate different regulated learning categories
• how do current frameworks account for temporal nature of regulatory processes
• can machine learning and educational data mining be used to understand SRL
• are inferences valid
• how do different instrictuional conditions, environments and contexts impact learners' ability to regulate aspects of their learning
• how do students “graduate” from pedagogical agents, scaffolding

Work

• examine dialogue moves between learners and agents
• design and test various versions of MetaTutor to test assumptions

Sanna Jarvela, University of Oulu; Hanna Jarvenoja, University of Oulu; Jonna Malmberg, University of Oulu

Hadin, Jarvela and Miller 2010, Winne, Hadwin and Perry 2010

Lack of theoretical and empirical congruence around social aspects of SRL

Target regulatory processes in challenging situations and how individuals and groups collectively respond to and resolve challenges

achieving success depends upon

• co-constructing of shared task representation, goals and strategies
• regulating learning through shared metacognitive monitoring and control of motivation, cognition and behaviour

Co-regulation in collaboration

• interactions between peers that mediate regulatory work
  • planning
  • goal setting
  • motivation

Shared regulation in collaboration

• shared aweareness of goals, tasks etc

Aim

• investigate socially shared regulation
  • individual and shared statements of challenges experiences
  • how do students collectively regulate contextual challenges
  • collaborative learning results compared to emerging shared regulation

Design

• 3 weeks
  • face to face
  • individual
  • collaborative
• nStudy (Winne & Hadwin) to support collaborative planning
• notes
  • solo notes (statements of challenges and regulation strategies)
  • collab notes
  • collaborative task performance
  • log file traces of contributions chat discussions

Results

• three groups
  • strong shared regulation
  • progressive shared regulation
  • weak shared regulation

Conclusions

• three kinds of profiles
• need to specify regulatory processes engaged to support learning and knowledge co-construction
• need for research designs for investigating solo and shared strategic regulation of learning and triangulation of different types of data

Allyson F. Hadwin, University of Victoria; Mariel F.W. Miller, University of Victoria

Coordination

• moving toward shared goal, joint task
• leveraging individual's unique and distributed knowledge and expertise
• achieving something beyond what any individual could do by themselves
• does not happen automatically

Success

• not only individually self-regulate
• co-regulate
• socially shared regulation of learning

Research

• infancy
• SSRL complex process not often achieved
• methodological challenge
  • capturing individual and group level knowledge
  • exploring alternatives to self-report measures
    • biased
    • prone to miscalibration

Study

• 1st yr undergrad course
• two major collaborative assignments (same groups, 3-4 students)
• data from second assignments
• choose high-performer and low-performer group
  • based on showing difference between individual final grades, and group grade
• each student an expert on one topic, so they had to be part of the assignment

Coding

• individual planning tool
• shared planning tool
• type of discussion
  • convergence
  • divergence
  • emergence
    • arose during discussion, not previously in individual response
• consensus index

Findings

• high performers had higher consensus in shared plans
  • could also be because they were already well-aligned with the instructor's perceptions of the task
• little evidence of active and purposeful construction of shared metacognitive planning knowledge in discussion
• unequal participation in both groups
• rarely the whole group at a time
• often one or two members were responsible for most of the ideas
• mostly accepted suggestions at face value

aims of symposium

• examine regulating
  • one's own learning
  • with others
  • together

Winne, Zhou, Egan (2011) in Assessment of higher-order thinking skills

How does a learning tactic play out (COPES)

• conditions
  • internal
    • knowledge
    • learning skills
    • efficacy
    • interest
  • external
    • time
    • task complexity
    • access to feedback
• operations
  • primitive operations
  • searching
  • monitoring
  • assembling
  • rehearsing
  • translating
    • if-then tactic
    • pattern of tactics: if-then-else-then
  • subject matter productions (rules, eg. resolving vectors)
• products
  • information created when operations are applied
    • knowledge
      • declarative
      • procedural
    • motivational interpretations
    • affect
• evaluations
  • fit of products to standards
  • optimal ideal or satisficing qualities of
    • product
    • engagements: tactics and strategies
    • improved or degraded predictability about whether/how conditions modulate engagements and products

(updating this list to reflect shared/social)

Facets of contexts

• focal tasks
  • main work
• supporting tasks
  • some contribute
  • some involve knowledge/skills outside
  • some are extraneous
• access to info
• task offload
  • opportunity of offloading work to contex and group into the environment (annotate and index content, rely on peer)
  • divide and conquer - how to put it back together again?
• density
  • ration of demand on working memory, group processes, relative to resources that support engaging in a task

Questions

• metacognition (megacognition)
  • can a person “keep up”?
  • can the group “divide and conquer”?
  • suggestion: look for COPES in help-seeking and help-giving
• grain size
  • how to traverse grain sizes and units of observation
• how to examine adaptation - dynamic social processes across time & task?
  • graph theory? dynamic social processes = event contingencies (n-partite graphs)