Transcript Normative influence in small groups - an experiment

Experimental research
Richard T. Watson
[email protected]
copyright © 2003
Normative influence in small
groups - an experiment
Clapper, D. L., E. McLean, and R. T.
Watson. 1998. Mediating group influence
with a group support system: an
experimental investigation. Journal of
Organizational Computing and Electronic
Commerce 8 (2):109-126.
Extended with the support of Mark Huber
(University of Georgia)
Behavioral models of decision
making
Simon’s Models of Man
Rational Person
Based on economics and management
science
Social Person
Based on social psychology and
anthropology
Rational and social models
Rational model
Social model
Influenced by information
Influenced by opinions, status, and
number of others
Self-interested maximizer of
subjective value
Influenced by the values and
norms of the surrounding culture
An isolated center of
consciousness to whom other
persons present themselves as
external objects
Immersed in a culture and closely
linked to significant others
Rational and social models
Rational model
Social model
Psychologically autonomous
Personality develops through a
process of internalization of social
norms and values. Values human
relationships
Rational, calculating, egoistic
seekers after pleasure
Emotional, altruistic, naturally
social creatures
Competitive
Cooperative
GSS design
Derived from the rational model
SAMM’s agenda
GroupSystem’s anonymous brainstorming
COPE’s cognitive mapping
Group decision making has
multiple goals
A high quality decision (rational)
Accepted by all group members (social)
“A poor plan well executed is better than
a good plan poorly executed”
Influence
• Information exchange
• Quantitative models
Informational
influence
Task
Outcome
Normative
influence
•
•
•
•
Input
Status hierarchy
Individual persuasiveness
individual dominance
Majority influence
Process
Output
Influence
Informational - rational
The power of knowledge and ideas
Normative - social
The power of the group, group norms, and
status
Dimensions of technological
intervention
Communication networks
The fixed, specific communication patterns, or
topologies, among members of a group
Communication modalities
The amalgam of methods and devices that may be
used to convey information
Communication strategies
The rules and procedures that structure a group's
activities
The role for technology in
group process intervention
Group process intervention with
information technology
Informational
influence
Task
Communication
configuration
• Networks
• Modalities
• Strategies
Input
Technology
intervention
Outcome
Normative
influence
Processes
Output
Research questions
Informational influence
Normative influence
Networks
How is information flow within the group
affected by different networks?
Do different networks impact the ability
of a member to dominate the group?
Modalities
Which modalities facilitate the most
rapid communication of information?
Do different modalities mediate the
power of the majority influence in the
group?
Strategies
How can the system determine the
optimal strategy for a given group task?
What strategies can the present system
use to minimize influence due to status
differences?
The effects of group pressure:
Asch, 1956
Series of judgments of lengths of lines
Unanimous, incorrect majority versus
subject
1/3 of estimates distorted by majority
pressure
Can information technology
impact group influence?
Rich versus lean media
Rich media emphasizes affective
component of messages
Problems with strict Asch
replication
Series of independent trials versus indepth analysis
Judgment of lines not very relevant to
managers
Would subjects’ behavior differ given
more time?
Would results be different for a more
ambiguous task?
Modifications
Group given decision-making task
Series of rounds of deliberations
Ambiguous versus clear-cut
How to do it?
Will subjects believe deception?
Will subjects move from initial position?
Experimental control
Informational and normative influence
Communication network
Communication strategy
Face-to-face manual
Whiteboard
Speech
S
C
C
C
Face-to-face GSS
Dispersed GDSS
Hypothesis H1
11
10
9
Round s to consensus
Different
communication
modalities will result in
a significantly different
number of rounds
required for the
subjects to move to the
majority position of the
confederates
Hypothesis supported
(p = 0.0046)
8
7
6
5
4
3
2
1
Control
GSS
Place
Type of s uppo rt
Place/time
With Control
Dunne tt's
0.0 5
Hypothesis H1a
Subjects in the leanest communication
modality (dispersed place/time GSS)
will take more rounds to move to the
majority position than subjects in the
richest modality (control)
Hypothesis supported (p = 0.0009)
Large effect size (1.09)
Hypothesis H2
Subjects working on
an intellective task will
take a significantly
greater number of
rounds to move to the
majority position than
subjects working on
judgment tasks
Hypothesis not
supported (p = 0.09)
10
9
8
7
6
5
4
3
2
1
0
Control
GSS
Place
Support type
Intellective
Judgment
Place/time
Hypothesis H3
11
10
9
Round s to consensus
Subjects working in control
groups on judgment tasks
will take a significantly
fewer number of rounds to
move to the majority
position than subjects in
dispersed place/time GSS
groups working on
intellective tasks
Hypothesis supported
(p
= 0.0058)
Effect size is large (1.27)
8
7
6
5
4
3
2
1
Control
Type of supp ort
Pla ce/time
Contributions of the research
Opens issue of rational and social goals of
GSS design
Examines the micro level of GSS technology
Demonstrates the relationship between
influence and communication modality
Emphasizes a social psychology perspective
Demonstrates the power of a carefully
designed, scripted, and controlled experiment