P1253296632UmIYq - PowerPoint PPT Presentation

Title: P1253296632UmIYq


1
VMT CSCL WorkshopJune 2004

2
Group Formation,Facilitation,Recruitment

This material is based upon work supported by the
National Science Foundation under Grant Nos.
0333493 and 0325447. Any opinions, findings, and
conclusions or recommendations expressed in this
material are those of the author(s) and do not
necessarily reflect the views of the National
Science Foundation.
3
Present
Future

• - Motivated
• Improving problem-solving skills
• Improving communication skills
• Increasing content knowledge
• Works for unlimited numbers of participants

4
How do people collaborate face-to-face?

• Middle school students
• - College/Grad students

Present
Future

• - Motivated
• Improving problem-solving skills
• Improving communication skills
• Increasing content knowledge
• Works for unlimited numbers of participants

5
How do people collaborate face-to-face?

• Middle school students
• - Grad students

Present
Future
How do people collaborate online?

• - Motivated
• Improving problem-solving skills
• Improving communication skills
• Increasing content knowledge
• Works for unlimited numbers of participants

• PoW-Wow! (AIM)
• - College/Grad students (Blackboard)

6
How do people collaborate face-to-face?

• Middle school students
• - Grad students

• - Parallel conversation is easy
• Erasing makes reconstruction hard
• Tone, expressions, gestures are easy
• Speaking is fast
• - ...

Present
Future
How do people collaborate online?

• - Motivated
• Improving problem-solving skills
• Improving communication skills
• Increasing content knowledge
• Works for unlimited numbers of participants

• PoW-Wow! (AIM)
• - Grad students (Blackboard)

7
How do people collaborate face-to-face?

• Middle school students
• - College Grad students

- ...

• - Parallel conversation is easy
• Erasing makes reconstruction hard
• Tone, expressions, gestures are easy
• Speaking is fast

Present
Future
How do people collaborate online?

• - Motivated
• Improving problem-solving skills
• Improving communication skills
• Increasing content knowledge
• Works for unlimited numbers of participants

• PoW-Wow! (AIM)
• - College/Grad students (Blackboard)

• - Parallel conversation is hard
• Typing is slow
• Tone, expressions, gestures are hard
• ...

8
How do people collaborate face-to-face?

• Middle school students
• - Grad students

• - Parallel conversation is easy
• Erasing makes reconstruction hard
• Tone, expressions, gestures are easy
• Speaking is fast
• - ...

Present
Neutral scaffolding Easy things should remain
easy, hard things should become less hard.
Future
How do people collaborate online?

• - Motivated
• Improving problem-solving skills
• Improving communication skills
• Increasing content knowledge
• Works for unlimited numbers of participants

• PoW-Wow! (AIM)
• - Grad students (Blackboard)

• - Parallel conversation is hard
• Typing is slow
• Tone, expressions, gestures are hard
• ...

9
Neutral scaffolding
Identify obstacles to exploration, provide tools
(e.g., shared whiteboard) to mitigate them.
10
How do people collaborate face-to-face?

• Middle school students
• - Grad students

• - Parallel conversation is easy
• Erasing makes reconstruction hard
• Tone, expressions, gestures are easy
• Speaking is fast
• - ...

Present
Neutral scaffolding Easy things should remain
easy, hard things should become less hard.
Future
How do people collaborate online?

• - Motivated
• Improving problem-solving skills
• Improving communication skills
• Increasing content knowledge
• Works for unlimited numbers of participants

• PoW-Wow! (AIM)
• - Grad students (Blackboard)

Positive scaffolding Make it easy to do the
right thing, and to learn what that is.

• - Parallel conversation is hard
• Typing is slow
• Tone, expressions, gestures are hard
• ...

11
Positive scaffolding
Make it clear where theyre supposed to go
easier to go there more difficult
to go somewhere else.
12
How do people collaborate face-to-face?

• Middle school students
• - Grad students

• - Parallel conversation is easy
• Erasing makes reconstruction hard
• Tone, expressions, gestures are easy
• Speaking is fast
• - ...

Present
Neutral scaffolding Easy things should remain
easy, hard things should become less hard. People
should be able to find collaborators.
Future
How do people collaborate online?

• - Motivated
• Improving problem-solving skills
• Improving communication skills
• Increasing content knowledge
• Works for unlimited numbers of participants

• PoW-Wow! (AIM)
• - Grad students (Blackboard)

Positive scaffolding Make it easy to do the
right thing, and to learn what that is. People
should be able to find appropriate
collaborators.

• - Parallel conversation is hard
• Typing is slow
• Tone, expressions, gestures are hard
• ...

13
Exploring How People Collaborate

• So far we did some experiments to get a feeling
  for this
• Face-to-face
• Fall 03 In a middle school classroom, we
  observed a face-to-face lesson
• Winter 04 Taxicab Geometry session in Gerrys
  class (university level)
• Online
• Winter 04 3 online sessions in Gerrys class
• Winter Spring 04 20 PoW-wow! sessions

14
What Factors Affect Collaboration?
Environmental Factors Group Attributes Individual Attributes
Math Problems - Difficulty - Resource availability - Learning goals
15
F2F Experiment I - Sharswoodgroup formation

• In the middle school classroom, we observed a
  face-to-face lesson in which students were
  grouped by their classroom teacher.
• Students were asked to self-assign typical
  cooperative learning roles (reporter,
  facilitator, recorder).
• Results They did not continue to use the roles
  as they worked together.

16
F2F Experiment II TaxiCab Geometry Sessions

• Student surveys were used to form groups
• Decisions were mainly based on availability of
  students for synchronous communication.
• Students were not highly motivated
• Depending on their math knowledge and problem
  solving skills, each group followed very
  different approaches.

17
F2F Collaboration in General

• Easy to coordinate discussion, taking turns
  etc.
• Effective communication channels, can benefit
  from gestures, body lang. etc.
• - Resource limitation, discussed ideas are
  volatile, could be forgotten etc.
• - Access to information resources is limited

18
Our Experience with Online Experiments In Class

• Groups were initially grouped randomly for the
  first online session.
• For the second two virtual problems groups were
  the same as in taxicab sessions. (longevity)
• Groups used Blackboard software (shared
  whiteboard, synch chat, equation editor)
• Groups did better as they became familiar with
  the software and their group members
• Facilitation instructions
• Inscribed-Circle Tangent-Square
  Ducks-in-a-row

19
Our Experience with Online Experiments PoW-Wow!

• Collected information
• Age
• Grade
• Gender
• Math class
• Knowledge about math
• Knowledge about technology
• How often they join a Pow-wow

• Group size
• Original goal of 3-4 students
• Later goal of 3-5
• With small numbers of registrants, groups were
  usually small

20
PoW-Wow! - Facilitation

• In initial Pow-wows, instructions were given
  within the chat room. Later general instructions
  were printed only on the problem page.
• Goal basic, clear information
• Script Facilitator Information

21
PoW-Wow! - Recruitment

• We have promoted the Pow-wows on the Math Forum
  site
• Problem of the Week index page
• Algebra and Geometry problem pages
• Instant Messages from MFpowwow to see if
  registrants are serious and if past participants
  are coming back
• We contacted some active PoW submitters

22
PoW-Wow! - Statistics

• 20 Pow-wows
• 37 different participants
• 7 repeating participants (2, 3, 3, 4, 4, 6, 11)
• 64 never-shows (some registered more than once
  some participants were no-shows at times)
• VMT facilitators joined 4 chats to fill space

23
PoW-Wow! - Getting Return Visitors

• Problems
• Time zones, home schedules, other homework
• Kids who register dont show up
• A few rude participants
• Software?
• Purpose not clear?
• Whats in it for the participant?
• Socializing, chatting
• Math help

24
Online Collaboration in General

• - Coordination is difficult (taking turns,
  following ideas etc.)
• - Limited communication channels
• Hard to share your work (e.g. diagram)
• Typing issue
• Resources, easy keep a log of the discussion
  for future reference
• Math resources Can benefit from online
  searches to retrieve necessary math info

25
What Factors Affect Collaboration?
Environmental Factors Group Attributes Individual Attributes
Math Problems - Difficulty - Resource availability - Learning goals Activity structure (roles) Software Interface - Affordances - Usability of the system Facilitation Group Size Member composition Longevity Cohesion Math skills - Math background - Problem solving skills Social skills - Communication style - Linguistic skills Technical skills - Familiarity with tech Availability Learning style Personality Motivation/Confidence Familiarity with tools/collaboration.
26
Some Desirable Outcomes of Small Group
Collaboration in Our Context

• Mastering problem solving (logical reasoning)
  skills
• Mastering communication skills, becoming better
  collaborators
• Developing mathematical thinking logically
  linking topics learned in class to solve a
  problem
• Asking insightful questions
• Raising different viewpoints, perspectives
• Paying attention to other members thoughts and
  needs providing feedback, help
• Critiquing other members ideas in a constructive
  way
• Learning from mistakes, making corrections/refinem
  ents with the help of other members
• Intertwining of perspectives through a process of
  negotiation collectively constructing a solution
  to the problem

27
To get to the desirable situation

• We need to find out how we should set the
  parameters effecting the collaboration.
• I.e. we need to answer questions like
• How we should form our groups?
• What sort of software tools do we need to provide
  scaffolding?
• What sort of facilitation/help features can we
  build into the system?
• What sort of problems should we use?

28
Approximating Parameters

• Answering previous questions requires us to
  approximate the values of key parameters
  (especially those regarding skills of group
  members)
• Possible data sources
• Self-assessment data obtained from surveys
• Observed performance in previous sessions (if
  such historical info is available)

29
Automation Issues

• We expect to serve 100s of users around the globe
• Due to the projected scale of the system we need
• a reasonable group formation scheme
• an automated facilitator agent (handle group
  formation, providing feedback etc.)
• a scheme for forming/updating user profiles
• updating user profiles according to observed
  performance

30
A simple framework for collaboration...
Present Future ?
Sharing drawings, resources is awkward. Sharing drawings, resources is fast and easy. Remove this obstacle.
Typing is slow, error prone. Other input modes are supported. Remove this obstacle.
Students dont seem to understand problem reduction. Tracking and coordinating sub-problems is facilitated by the environment. Provide this structure.
Hypotheses already rejected can resurface. Tracking confirmed or rejected hypotheses is facilitated by the environment. Provide this structure.
Limited perspectives lead to circling, stuckness. Multiple perspectives increase alternatives. Form groups more effectively
Mismatch in knowledge levels leads to excessive catch-up discussions. Compatible knowledge levels support facile exploration of problem space. Form groups more effectively
In wiki...