Enhancing Metacognitive Skills through Cooperative Learning

1
Enhancing Metacognitive Skills through
Cooperative Learning In a Scientific
Concept-Learning task using Hypermedia
Chanchai Singhanayok McDonald Director,
Educational Technology Unit The Office of
Biomedical Research Education and Training School
of Medicine, Vanderbilt University
Anju Relan Director, Instructional Design and
Technology Unit David Geffen School of Medicine
at UCLA
2
Purpose
Examine the effect of using metacognitive
strategies to enhance learning in cooperative
learning groups pairing students with high and
low prior knowledge.
Examine the effect of pairing students in
homogeneous and heterogeneous cooperative
learning groups.
3
Hypothesis
The exemplary verbal modeling of metacognitive
skills by high-achieving students while learning
in heterogeneours pairs will improving learning
among both high and low achievers.
Changing learners patterns of decision-making
through effective cognitive monitoring would lead
to improved performance.
4
Literatures

• - Snow, 1980 Steinberg 1977
• Empirical evidence to support the notion has
  been contentious.
• Temiyakarn Hooper, 1993
• Cooperative learning is effective for
  improvement of higher levels of cognitive
• processing.
• Johnson Johnson, 19811985
• Cooperative interaction promoted higher
  achievement and the discovery of superior
• cognitive reasoning strategies.
• Cooperative learning increases students
  mastery of higher level reasoning skills.
• Spurlin, Dansereau, Larson, and Brooks, 1984
• Individuals in cooperative learning groups used
  metacognitive strategies
• more frequently than did individuals working
  alone.
• Sharan, 1980 Hooper, Temiyakarn, Williams,
  1993
• The positive effects of cooperative learning
  have been found in higher-level concept
• learning.

5
Literatures

• Pressley, Borkowski, OSullivan, 1984
  Pressley, Ross, Levin Ghatala, 1984
• Pressley, Levin, Ghatala, 1984 Casey
  Hannafin, 1986 Garner Reis, 1981
• Glenberg Epstein, 1985 Tobias, 1987
• Students expend effort on strategies that do
  not work especially when given control of
• their own learning as is frequently the case
  with hypermedia studies.
• Merrill, 1987
• If students are not exercising decisions under
  their control wisely, the built-in monitor
• should intervene and lead the student through
  instruction.
• Reigluth, 1979
• A CAI system should provide a large degree of
  control over macro and micro
• instructional component, but it should also
  provide learners with the knowledge and
• information necessary for them to make good
  learner-control decisions.
• Flavell, 1979 Brown Smiley, 1978
• In heterogeneous cooperative learning group,
  when low ability students interact with
• high ability students in a problem-solving
  domain, a great deal of learning occurs.

6
Method
Subjects
Subjects
Materials
Design
Experimental procedure
7
183 sixth-graders of middle-class background
Subjects

Level of articulation
Level of grouping
8
Subjects
9
Method
Subjects
Materials
Materials
Design
Experimental procedure
10
Materials
Post-test
Training
Computer -presented content
Pretest
11
Materials
Training
Cooperative training
Enhance intra-group interaction and cooperation
Computer training
Provide familiarity during the computer-based
lesson
12
Materials
Pre-test
Require generation of basic science concepts
25 questions
Pre-test scores were used for assigning
students to level of prior knowledge (high or low)
K-R 20 reliability 0.85
13
Materials
Computer-presented content
Designed and developed by the experimenter using
biological concepts (Relan, 1991
TemiyakarnHooper, 1993)
Tutorial
Conceptual, Learner-controlled lesson
Hypermedia format with 2 different versions
Verbal articulation required version
Topic Ecology
No verbal articulation required version
14
Materials
15
Materials
Post-test
Used for both immediate and delayed post-tests
25 multiple choices
10 items on verbal information 15 items on
generalization
Experiment
Post-test
Post-test
When
1 week
16
Method
Subjects
Materials
Design
Design
Experimental procedure
17
Design
Design
Randomized Block Design
Prior knowledge
Blocking factor
Level of grouping
Analyzing Tool MANOVA
Cross-experimental factor
Level of articulation
Cross-experimental factor
Immediate post-tests
Dependent measure
Delayed post-tests
Dependent measure
18
Design
Experimental Design 2X2X2
19
Method
Subjects
Materials
Design
Experimental procedure
Experimental procedure
20
Experimental procedures
Experimental procedures
G1-G5
Training session
Pre-test
Treatment
Post-test
21
Results
Immediate Post-test
22
Results
Delayed Post-test
23
Results
Immediate Post-test
Delayed Post-test

• There was no sig. difference among the treatment
  groups.
• There were no sig. interaction among the three
  factors.

24

• Homogeneous group without articulation
• outperformed homogeneous with articulation.
• No difference between heterogeneous group with
• and without articulation.
• Non articulation group outperformed the
• articulation group among high prior
  knowledge.
• Low prior knowledge student performed better
  when articulation was required.

The interaction of Level of Articulation, Group
Composition, and Prior Knowledge Level.
The interaction of articulation and group
composition among low prior knowledge students.
The interaction of articulation and group
composition among high prior knowledge students.
25
Discussion
Requiring articulation
These results indicate that peer modeling did not
significantly improve the scores of low-achieving
students, and that high-achieving students did
worse in homogeneous group when required to
articulate, than in any other condition.
Heterogeneous grouping
These results suggest that requiring articulation
may be an effective strategy for learning in
heterogeneous groups, but may hinder the
performance of high-achieving students working
together in homogeneous groups.
26
THANK YOU !