Does the interface of an OLE influence aspects of the contributions made to it

1
Does the interface of an OLE influence aspects of
the contributions made to it?

• Dr. Chris Hughes1, Dr. Lindsay Hewson2,
• Ms. Sophie di Corpo1
• 1 School of Public Health and Community Medicine
• 2 School of Medical Sciences
• University of New South Wales

1
2
2
3
Background

• Research on facilitation online has been
  conducted on single OLEs.
• Yet there are OLEs with radically different
  interface designs.
• Might not the interface designs have an impact on
  the processes they support?
• Focus is on the Communication Tools, not the
  whole omnibus OLE.

3
4
Research questions

• What indicators of deep learning processes can be
  identified in the transcripts of online classes?
• Do features of the interface impact on the
  character of the contributions made to online
  classes, and more specifically
• Do features of the interface impact on indicators
  of deep learning in the online environment?

4
5
System One WebCT (CE)

• Features a quite standard newsgroup type
  interface for communications.
• Few structured supports for teaching strategies
• Standard newsgroup interface, with contributors
  to a thread able to reply to individuals out of
  chronological sequence.
• Relatively uninformative email notification

5
6
(No Transcript)
7
System Two WebTeach

• Features a strictly chronological listing of
  contributions.
• Provides explicit structural support for a range
  of strategies, including discussion,
  brainstorming, case studies, questioning,
  debates, commitment activities, quizzes,
  task-setting
• Contributions are to the activity ( thread), not
  in reply to another posting.
• Email notification includes the activity title, a
  sample of the contributed text, and the URL.

7
8
(No Transcript)
9
Issues encountered

• Ethical clearance! Opt-in and opt-out.
• Nomenclature (threads vs. activities)
• Coding scheme(s).
• Coding incommensurate systems (eg. anonymity,
  private activities, strategies).
• WebCT does not easily support structured
  processes.
• Quest for reliability

9
10
Indicators of deep learning

• Learner activity and engagement
• Access to the knowledge base of the discipline
• Structured group interaction
• Some level of learner control
• Support for phatic or social processes, and
• Opportunities for reflection.
  (Ramsden, 1992)

10
11
Deep learning and e-learning

• Learner activity and engagement
• intensity of contributions, contribution rate per
  student.
• identification of contributions that build on the
  work of others.
• Structured group interaction
• identification of clear tasks and processes,
  focused on intentional pedagogic action by
  teachers.
• Support for phatic or social processes
• identification of affirmative/process management
  activity
• naming rates?
• Opportunities for reflection
• reading contributions, contributing, observe the
  interactions of others, option to be anonymous.
• (The degree of Learner control and Access to the
  knowledge base, are clearly strategy, discipline
  level dependent)

11
12
Deep e-learning indicators?

• Contribution rates (perhaps)
• Naming rates (perhaps)
• Intensity of contributions (contributions per
  day) (perhaps)
• Level of interactive contributions (perhaps)
• Percentage of student contributions in the total
  (perhaps)
• Level of Affirmative/process contributions
  (perhaps)

12
13
Our coding scheme

• Allocated percentages to up to three
  communicative intentions per contribution,
  threshold of 20
• Individual
• Collaborative
• Affirmative/process
• Plus objective data on contribution and naming
  rates, identity of contributors etc.
• (Salmon, 1999 Corich, 2004)

13
14
Initially the coding was complex

• Individual initiate new topic articulate or
  explain own views give examples or
  illustrations reflect or re-evaluate personal
  views.
• Interactive expand on others ideas summarise
  prior contributions propose actions share
  resources.
• Affirmative/process affirming or phatic
  meta-comment or group process.

14
15
Data Set

• All assessable post-graduate subjects, fully
  online
• Clear pedagogic tasks.
• Single coder (some reliability, less valid).
• Discussion mode only in WebTeach.
• 5 WebCT classes 2 WebTeach classes.
• 19 WCT threads 28 WT activities.
• 161 students 54 students
• 418 posts 762 contributions.
• 776 thread-days 357 activity-days.

15
16
Most indicators comparable
17
Three indicators stand out

• Naming rate
• WebCT 44.7
• WebTeach 24.9
• Intensity of contributions (posts/topic day)
• WebCT 0.54
• WebTeach 2.13
• Contribution numbers and rates (posts/student)
• WebCT 418 posts 2.6 posts/student
• WebTeach 762 posts 14.1 posts/student
• Could these differences be attributed to the
  interfaces?

17
18
(No Transcript)
19
(No Transcript)
20
Naming rate

• Higher in WebCT, perhaps because
• In WebCT you respond to anothers contribution.
  So you are addressing the original author
  directly, as in an email.
• In WebTeach you respond more to the discussion as
  a whole.

20
21
Intensity and contribution rate

• Higher in WebTeach, perhaps because
• More informative email notification
• Simpler presentation of contributions
• More activities, fewer classes, stricter
  deadlines on activities, shorter timeframe for
  activities
• Only the first two factors are clearly related to
  the WebTeach interface, although the third set
  may be related to it too.

21
22
We also noted

• Tendency in both systems for early contributions
  to be coded as individual, later ones as
  collaborative.
• Clear, but relatively unchallenging tasks, and
  none that explicitly called for collaboration.

22
23
Impact on indicators

• WebCT interface may have an impact on naming
  rates
• WebTeach interface may have an impact on
  intensity and contribution rates
• So these indicators may be interface dependent,
  as may others

24
References

• Corich, S., Kinshuk, Hunt, LM, (2004) Assessing
  Discussion Forum Participation In Search of
  Quality, International Journal of Instructional
  Technology and Distance Learning Vol 1 (12).
• Hewitt, J. (2003) How habitual online practices
  affect the development of asynchronous discussion
  threads. Journal of Educational Computing
  Research, 28(1), 31-45.
• Ramsden, P. (1992). Learning to teach in higher
  education. (London, Routledge).
• Salmon, G. (1999) Reclaiming the Territory for
  the Natives, Online Learning Exploiting
  technology for training, London, Nov. 23 24,
  1999.

24
25
Any questions?
25
26
Thank you!

• Correspondence
• Dr. Chris Hughes
• School of Public Health and Community Medicine
• University of New South Wales
• Sydney 2052
• Australia
• c.hughes_at_unsw.edu.au

26