INTERNATIONAL
JOURNAL
BEHAVIOR, 6(4), 501-521
OF
ORGANIZATION
THEORY
AND
WINTER 2003
DESIGNING THE LEARNING ORGANIZATION AS AN
INFORMATION-PROCESSING SYSTEM: SOME DESIGN
PRINCIPLES FROM THE SYSTEMS PARADIGM AND
CYBERNETICS
Yasin Sankar*
ABSTRACT. The major principles for designing the learning organization as an
information processing system are derived from systems paradigm, information
theory, and cybernetics. The need for these principles is demonstrated by the
information pathologies in the classical and contingency design of the
organization and information imperatives for designing the organization for the
information age. An information processing model that extends the classical and
contingency principles for organizational design is developed to provide a new
organization model for effective learning. The effectiveness of the learning
organization can be partially attributed to the design of the organization as an
information processing system. The organization learns, adapts, and responds to
innovative change through its information subsystems.
INTRODUCTION
In the learning organization, everyone is engaged in identifying and
solving problems, enabling the organization to continuously experiment,
improve, and increase its capability. The essential value of the learning
organization is problem solving, in contrast to the traditional
organization that was designed for efficiency. In the learning
organization, employees engage in problem identification, which means
understanding customers’ needs. Employees also solve problems, which
mean putting things together in unique ways to meet customer needs.
The organization in this way adds value by defining new needs and
solving them, which is accomplished more often with ideas and
information than with physical products. When physical products are
---------------* Yasin Sankar, Ph.D., is Professor, Faculty of Management, School of Business
Administration, Dalhousie University. His research interests are in leadership,
ethics, learning organizations, cultural diversity, reengineering of education.
SANKAR
Copyright © 2003 by PrAcademics Press
502
SANKAR
produced, ideas and information still provide the competitive advantage
because products are changing to meet new and challenging needs in the
environment (Goh & Richards, 1998).
Proposition 1: The organizational chart designed by both classical and
contingency principles pictures an organization as being structured
around the authority and power variables rather than around
information processing systems, information flows, and information
links so vital to effective decision-making.
INFORMATION PATHOLOGIES AND ORGANIZATIONAL DESIGN
The Classical Design
Proposition 2: The major information pathologies of the classical
organizational design are information filtration, distortion, overload
and lags in feedback. The major cause of the organizational and
information mismatch is the lag between organizational changes and
information systems to facilitate them.
Authority and power provide the rationale for organizational design
in the classical perspective. In the hierarchical structure of the classical
organization, the major design parameter is the formal authority at the
strategic apex and a chain of command. The classical approach has
always provided for routine exchange of information across the chain of
command. The coordination of workflow is done by rules, regulations,
procedures, and operations manual that regulate the information flow.
The coordination of information flow across the organizational units is
problematic. The vertical flow of information down the organizational
chain of command is imperative. The centralization of authority for
decision-making at the strategic apex legitimizes these communications
channels. Information filtration in the classical design is a major
pathology of the system. Information overload in the organizational
hierarchy is a common feature of the system. Information distortion
because of bureaucratic codes, symbols, operations manuals and
specialized information taxonomies is another pathology of the classical
design. Feedback on change initiative at lower levels of the hierarchy is
quite limited and with extensive lags. Categorization as a decisionmaking technique is common because of the ritual of categorizing of
problems into files, codes, and information taxonomies. The major
bureau pathology of the classical design is in the information-processing
field. Classical organizational structure can be visualized as a
DESIGNING THE LEARNING ORGANIZATION AS AN INFORMATIONPROCESSING SYSTEM
503
formalization of communications channels. The objective of
formalization is control over organizational activities by making the
patterns of information flow predictable and uniform through prescribed
communications channels. Two classical principles, scalar chain and
centralization, are the means by which control, predictability, and
consistency are maintained (Sankar, 2001).
The Contingency Design
Proposition 3: The major information pathologies in
organizational design are the lack of variety
communications networks, and configurations
databases. The organizational structure does not
information processing systems of the organization.
the contingency
of information,
of distributed
conform to the
In the contingency design there are some information pathologies but
they are not as problematic and dysfunctional as in the classical design.
The chain of command, the hierarchical structure of authority is still the
predominant mode. To cope with environmental and technological
change a high degree of differentiation among organizational units is
necessary. To coordinate the workflow among these differentiated units,
interdependence must be managed through the information flow. In
managing interdependence (pooled, sequential, reciprocal and team)
certain integrative mechanisms must be designed. Unlike the classical
design which relied heavily on rules, regulations, and manuals as
integrative mechanisms, the contingency design uses liaison roles, task
teams overlays and integrating units that may cut across the chain of
command (Galbraith, 1995). These integrative mechanisms are more
flexible for the processing and conversion of information into decision
outcomes. However, the decision centers are not necessarily aligned with
the communications channels, communications networks or information
centers. Power and authority are still the critical criteria that provide the
rationale for organizational design rather than information processing
and conversion. There are problems in the contingency design of
information filtration, information overload, information distortion and
lag in feedback on change initiatives. These pathologies are not as severe
as in the classical design because of more flexible coordination modes
for information flow in the vertical and horizontal dimensions of the
organization. However, the organizational design does not enhance the
flow of information or optimize the use of a variety of communications
504
SANKAR
networks and channels. The processing of information and the
conversion of information with a high degree of efficiency is still
problematic.
Information Imperatives in Organizational Design
Proposition 4: The information processing capacity of an organization
must match the information-processing requirements of the network
of tasks. A variety of integrative mechanisms must be
institutionalized within the structural configurations of the
organization to enhance the information processing capacity of the
organization.
The organizational structure of learning organizations has been
described in the literature as organic, flat, and decentralized, with a
minimum of formalized procedures in the work environment. Some
research has supported this finding: organizations with a strong learning
capability tend to have low scores on formalization in their organization
structure. These research results clearly show a negative relationship
between formalization and learning capability (Goh & Richards. 1998).
Other researchers have also found that learning organizations
generally have fewer controls on employees and have a flat organization
structure that places work teams close to the ultimate decision-makers
(Miller, 1990). The implication is that the five strategic building blocks
can only operate effectively when the organization has a flat,
nonhierarchical structure with minimal formalized controls over
employee work processes.
The limitations of both the classical and contingency design of the
organization as an information processing system emphasize the urgency
of articulating the information imperatives in organizational design. Four
criteria for communications effectiveness are speed, accuracy, variety,
and richness. An organizational design must meet these criteria. Speed is
essentially the rapid flow of information to decision centers along the
vertical and horizontal dimensions of the organization. Accuracy is the
minimization of errors in information distortion and information
filtration. The information taxonomies within the system must be
concise, lucid and logical to facilitate understanding of the information at
various levels of the organizational hierarchy. The use of codes, symbols,
and jargon must be minimized. Information filtration along the
organizational hierarchy must be minimized through a change in the
DESIGNING THE LEARNING ORGANIZATION AS AN INFORMATIONPROCESSING SYSTEM
505
organizational culture, its value system, code of ethics, and reward
systems. The other important design variable is information richness.
Richness pertains to the information-carrying capacity of data (Daft &
Lengel, 1998). Information richness is related to the medium or channel
through which it is communicated. Some media are richer because they
provide richer information to managers. Information richness is
important because it relates to the ambiguity of management problems.
Rich media provide multiple cues and feedback. Variety as a design
variable is also critical for speed and accuracy. When task variety is high,
problems are frequent and unpredictable. Uncertainty is greater so the
amount of information needed will also be greater. Managers spend more
time processing information and they need access to larger databases.
When variety is low the amount of information to be processed is low.
Some information imperatives are the following:
(1) Decision centers must be aligned with the information
subsystems.
(2) A network structure of control, authority, and information vs. a
hierarchical structure is mandatory.
(3) A variety of communications networks must be experimented
with and instituted in the design.
(4) Integrated and distributed databases in a variety of configurations
must be linked to the computer systems.
(5) Vertical and horizontal linkage mechanisms for coordinating the
information flow must be designed.
(6) The organization must be designed around the information
subsystems, the sensor subsystems, data processing subsystem,
decision subsystem, process subsystem, control subsystem and
memory subsystem rather than around authority and power.
(7) Information is the power variable in organizational design since
management is redefined as the processing of information and the
conversion of information into decision outcomes.
(8) The learning organization must be designed to provide
information for strategic change.
506
SANKAR
(9) Organizational design is the structuring of task and the structuring
of authority to achieve organizational goals. The structuring of
task and authority must be aligned through information
processing and conversion mechanisms.
(10) With change, complexity and uncertainty, the information
processing capacity of the organization must be increased through
all the information subsystems for effective strategic planning.
The Systems Paradigm
According to Luthans, despite divergent viewpoints, the systems
approach more than any other conceptual approach, has led
organizational theorists to take a unified view of the organization as a
whole made up of interrelated and interdependent parts (Luthans, 1996).
This view of organizations reflects the concept of synergism. The
synergistic effect means that the whole organization is greater than the
sum of its parts. This is an outgrowth of and is closely related to, the
gestalt school of psychological thought.
The recent view of organizations as information processing systems
facing uncertainty serves as a transition between systems theory and
cybernetics information theory. The information processing view makes
three major assumptions about organizations. First, organizations are
open systems that face external environmental uncertainty and internal,
work-related task uncertainty. (Galbraith, 1995, p. 165) defines task
uncertainty as “the differences between the amount of information
required to perform the task and the amount of information already
possessed by the organization.”. The organization must have mechanisms
and be structured in order to diagnose and cope with this environmental
and task uncertainty. In particular, the organization must be able to
gather, interpret, and process appropriate information to reduce the
uncertainty. The second assumption is as follows. Given the various
sources of uncertainty, a basic function of the organization’s structure is
to create the most appropriate configuration of work units as well as
linkages between these units to facilitate the effective gathering,
processing, and distribution of information. The final major assumption
involves the major organizational units of the systems. Because the
subunits have different degrees of differentiation, it is necessary to
examine the information subsystems, sensor, data processing, decision
subsystems, control subsystem and memory subsystem of these subunits.
Here we are concerned with the structural mechanisms that will facilitate
DESIGNING THE LEARNING ORGANIZATION AS AN INFORMATIONPROCESSING SYSTEM
507
effective coordination among differentiated
subunits.
yet interdependence
Tushman and Nadler (1978) formulate the following propositions
about an information processing theory of organizations:
1. The tasks of organizational subunits vary in their degree of
uncertainty.
2. As work-related uncertainty increases, so does the need for
increased amounts of information, and thus the need for increased
information processing capacity.
3. Different organizational structures have different capacities for
effective information processing.
4. An organization will be more effective when there is a match
between the information processing requirements facing the
organization and the information processing capacity of the
organization’s structure.
5. If organizations (or subunits) face different conditions over time,
more effective units will adapt their structures to meet the changed
information processing requirements.
INFORMATION PROCESSING PRINCIPLES FOR DESIGN
Proposition 6: The systems view of organization takes into account the
integrative nature of information flows. The structures of the
organization should be designed to facilitate information inputs to
decision centers since communications is the process by which
organizations change and learn, and adapt to a volatile environment.
The systems approach in organizational theory and design has been
used as a framework for describing the obvious aspects of an
organization such as the functional subsystems of an organization, its
input-output-process, the boundary of the system and its degree of
interdependence with its environment and among its subsystems. The
systems approach has not provided adequate explanation regarding the
functioning of the organization nor hypothesis on the nature of systems
interdependence or the impact of change and innovation on strategic
organizational variables. It has not been applied as an analytical
framework to predict the behaviour of the system. The organization as a
508
SANKAR
behavioural system exchanges resources, energy, and information with
its environment. It tends to maintain itself in a steady state or
equilibrium, adopts a feedback process to facilitate adaptation, operates
as a cycle of events that become institutionalized and functions in a state
of dynamic interaction among its subsystems (Figure 1).
Senge (1990) proposes that learning can occur at the individual, team
and systemic levels in an organization. The three areas are interrelated
and specific organizational systems can be developed to promote
learning at all levels. The author suggests that to encourage learning, an
organization should generate a holistic view of the organization, acquire
and interpret information from the environment, provide only minimum
specifications for jobs, facilitate alliance with other organizations and
implement systems that retain knowledge (Senge, 1990). The systems
paradigm provides such a holistic view to facilitate organizational
learning through the information subsystems of the organization.
Information is the mechanism through which organizations learn,
change, and adapt to a dynamic turbulent environment.
The information-processing aspect of the computer has a general
analogy in the organizational field. Every organization has the following
subsystems: the major design principle may be stated as follows. A
strategic change in information input into an organizational system will
FIGURE 1
The Information-Processing Subsystems of an Organization
Information
Change &
Innovation
Input
Look for a
right
problem
Check data
with
taxonomy
Sensor
subsystem
Data
Processing
system
Control
subsystem
Make
decisions
Decision
subsystem
Memory
subsystem
Material and
energy input
Process
subsystem
DESIGNING THE LEARNING ORGANIZATION AS AN INFORMATIONPROCESSING SYSTEM
509
Recycle
Feedback Loop
produce changes in the information process system, the sensor, data
processing, decision, process and memory subsystem.
1. A sensor subsystem, which is concerned with the reception and
recognition of information.
2. A data processing subsystem, which is concerned with breaking
down this information into terms and categories which are
meaningful and relevant to the system.
3. A decision subsystem, in which decisions are made. Decisions may
involve:
a. self-regulatory or homeostatic processes
b. adaptive or learning processes
c. integrative processes
4. A processing subsystem, which ties the whole system together by a
set of feedback loops. These loops incorporate the equations of the
critical decision variable which, if not respected, lead to lack of
growth and the eventual demise of the organism.
5. A memory subsystem, which is concerned with the storage and
retrieval of information (Kelly, 1980).
CYBERNETIC PRINCIPLES FOR DESIGN
Proposition 7: An organization’s decision environment is a major factor
in the effective design of its management information system. The
pattern of structured - unstructured decisions varies with levels of the
organizational hierarchy and so must the information systems of the
organization.
Since the systems paradigm focuses on the dynamic interrelationship
and interaction of entities and subsystems, information, and
communication theory are important to the development of systems
theory. From a systems theory viewpoint, therefore, an organization can
be viewed on one level as a complete integrated decision-making system
designed to achieve some specific objective.
510
SANKAR
The links between the elements of a system are the communications
network within the systems. The state of the link reflects the amount of
the information in the system. If the actions of the systems were
completely predictable, as in the process of a large digital computer, it
would be a complex, deterministic system. Where the relationships are
not completely determined, as in the industrial corporation, the system is
called complex and probabilistic. The problem of control in complex
probabilistic systems is the major focus of cybernetics. The interaction of
the system elements on the environment or internal process and systems
must be regulated through feedback mechanisms. There can be negative
and positive feedback; the negative feedback serves to minimize the
distance from the norm, and the positive increases the deviation.
The operation of the cybernetic system is chiefly a matter of storing,
receiving, transmitting, and modifying information. There is high degree
of uncertainty inherent in this operation since the permutations of the
system’s data, or fact elements are enormous. The order of the
uncertainty in the data, according to Rogers (1987), produce information
and the system becomes more controlled and the end-state becomes more
controlled and the goal of the system more predictable (Rogers, 1987).
A variant of the systems approach is the cybernetics model.
Cybernetics as a concept in organizational theory integrates the linking
processes in organizational systems and generalizes them to a wide
variety of systems. The linking processes are communication, balance,
and decisions. It is through them that dynamic and basic interactions are
initiated and facilitated in an organization. Decisions, information
(communication), and control (balance or regulation) are indispensable
elements of complex systems. Stafford Beer notes that “decisions are the
events that goes on in the network and they are describable… in terms of
the information in the system, and the structuring of
communication”(Beer, 1959, p. 175). In this sense, then decisions and
information cannot be understood apart from the system’s
communication pattern, and this pattern in turn is a reflection of the
decisions required and the information necessary upon which to base
them. Balance, the third linking process, is introduced in the form of
control, or regulation, and this constitutes the heart of the cybernetic
processes. Regulation of the system network by the information
produced in it is the core of cybernetics. One of the requirements of a
cybernetic system is self-regulation. Self-regulation, usually identified as
DESIGNING THE LEARNING ORGANIZATION AS AN INFORMATIONPROCESSING SYSTEM
511
feedback, controls current activities by adjusting them after comparing
their outcomes or performance against some standard or objective.
From the point of view of cybernetics, any large-scale formal social
organization is a communications network. It is assumed that these can
display learning and innovative behaviour if they process certain
necessary facilities (structure) and certain necessary rules of operation
(content).
First, consider the structure of the system as it might be represented
in the language of cybernetics. Any social organization that is to change
through learning and innovation, that is, to be ultra-stable must contain
(1) certain very specific feedback mechanism, (2) a certain variety of
information, and (3) certain kinds of input, channel, storage, and
decision-making facilities. This can be stated in the form of an axiomatic
proposition: that complexity of purposeful behaviour is a function of the
complexity of the communication components or parts of the system.
More specifically, every open system behaving purposefully does so by
virtue of a flow of factual and operational information through receptors,
channels, selectors, feedback loops, and effectors. Every open system
whose purposeful behavior is predictive, and this is essential to ultrastability, must also have mechanisms for the selective storage and recall
of information: it must have memory. Does the social organization under
scrutiny behave purposefully? Does it solve problems? And does it
forecast future events? If the answers are in the affirmative, then one
must find in it certain kinds of communications, information, and control
mechanisms that are conducive to change and technological innovation
(Cadwallader, 1985).
A cybernetic model would focus the investigator’s attention on
information as the critical variable: (1) the quantity and variety of
information stored in the system, (2) the structure of the communications
network, (3) the pattern of the subsystem within the whole, (4) the
number, location, and function of negative feedback loops in the system
and the amount of time-lag in them, (5) the nature of the system’s
memory facility, (6) the operating rules, or program determining the
system’s structure and behaviour.
The operating rules of the system and its subsystems are always
numerous. Relevant for the present problem are (1) rules of instructions
determining range of input, (2) rules responsible for the routing of the
512
SANKAR
information through the network, (3) rules about the identification,
analysis, and classification of information, (4) priority rules for input,
analysis, storage, and output, (5) rules governing the feedback
mechanisms and, (6) instructions for storage (Cadwallader, 1985).
Chaffee (1980), in reviewing the literature on decision-making, has
developed a set of seven criteria for defining the requirements of rational
choice processes in terms of the collection and use of information:
1. Information is received prior to the decision’s being made;
2. Information is problem-centered and goal-directed;
3. Information documents the existence of and the need to solve the
problem or reach the goal;
4. Information includes consideration of more than one alternative for
reaching the goal or solving the problem;
5. Information has logical internal consistency in terms of posited
cause-effect relationships;
6. Information is oriented toward maximization, in that it demonstrates
the value of the various alternatives considered in reaching the goal;
and
7. Information identifies the value premises on which it is based.
One final condition could be added, which is that the choice is made
to accept those alternatives which, on the basis of the information
provided, seem to provide the best likelihood of achieving the goals or
solving the problem.
INFORMATION, CHANGE, AND INNOVATION (SOME PRINCIPLES)
The literature on organizational learning has been elusive in
providing practical guidelines or managerial actions that practicing
managers can implement to develop a learning organization. Some of the
questions raised by managers about the concept of learning organizations
are as follows: What is a learning organization? What are the payoffs of
becoming a learning organization? What should I do to encourage
organizational learning? How do I know if my company is a learning
organization? What are the characteristics of a learning organization and
how do I sustain one? Is there an implementation strategy? Clearly a
DESIGNING THE LEARNING ORGANIZATION AS AN INFORMATIONPROCESSING SYSTEM
513
discussion with a managerial perspective on how to build a learning
organization is lacking in the literature.
To answer the first and most frequently asked question, “What is a
learning organization?” we need a definition. The following definition
best reflects the conceptual approach of this paper (Goh & Richards.
1998).
Proposition 8: A learning organization is an organization skilled at
creating, acquiring and transferring knowledge, information, and at
modifying its behaviour to reflect new knowledge and insights
(Garvin, 1993).
In a learning organization:
1. The rate of innovation is a function of the rules organizing the
problem-solving trails (output) the system; [This sentence does not
make sense as written]
2. The capacity for innovation cannot exceed the capacity of the
information processing systems of the system;
3. The rate of innovation is a function of the quantity and variety of
information;
4. A facility, mechanism, or rule for changing organizing patterns of
a high probability must be present to facilitate flexibility of
processing;
5. The rate of change for the system will increase with an increase in
the rate of change of the environment (input). That is, the changes
in the variety of the inputs must force changes in the variety of the
outputs or the system will fail to achieve ultra-stability.
6. As cybernetics recognizes the need to study interactions of the
subsystems in an organization, it focuses on information as the key
to analyzing and understanding organizations as learning systems.
Consequently, communication theory and information theory were
central in the development of cybernetic systems and offer the following
principles:
1. Communication is the basic process facilitating the
interdependence of the parts of the total system; it is the
mechanism of coordination.
514
SANKAR
2. Information processing is the main function performed by all
organizations;
organizational
systems
were
essentially
communication systems.
3. Management in cybernetic language is the processing of
information and the conversion of information into decision
outcomes
4. The greater the task uncertainty, the greater the need for
information processing.
5. The larger the number of elements in the decision-making process
such as the number of departments, the greater the informationprocessing requirements of the organization.
6. The greater the degree of interdependence among the elements of
decision-making, the greater the information processing.
7. To increase the capacity of the organization to process information,
more flexible integrative mechanisms are needed.
8. Task variety, change, and complexity will produce different
degrees of uncertainty in the information domain and decisionmaking.
9. The structural configuration of an organization must be designed
with information as the design parameter.
10. Different structural configurations are needed for different
information domains within a complex organization.
The currency of organizational theory is information. As information
is infinite in quantity, it is expensive to collect, must be treated
selectively, can only be packed up with a taxonomy (classification) based
on yesterday’s experience (and thus is to some extent irrelevant), causes
surprise (which is how we measure it), and is fraught with uncertainty.
AN INFORMATION PROCESSING MODEL
The systems and cybernetic concepts of organizations focus on the
dynamic interaction and intercommunication among components of the
system. Systems theory subordinates the separate units or departments of
an organization to decision-making and communications networks. The
cybernetic view of organizations is concerned with decisions, control,
self-regulation, and feedback operating in dynamic equilibrium. Both
DESIGNING THE LEARNING ORGANIZATION AS AN INFORMATIONPROCESSING SYSTEM
515
take into account the integrative nature of information flows. This
concept is demonstrated in Figure 1 where each organizational entity is
seen as an information system with the components of input, processor
and output. Each is connected to the others through information and
communication channels and each organizational entity becomes a
decision point. Thus the communications process is dynamic, because as
the objectives of the organization change, the content of communication
must also change in order to alter or reinforce the actions of various
segments of the organization. In the language of the systems model and
cybernetics, communication is a linking activity of organizations.
At a practical level, designing the organization as an information
processing system means that the following information subsystems, the
sensor subsystem, data processing system, decision subsystem, process
subsystem, control subsystem, and memory subsystem may be integrated
into each structural configuration within the organization and along
levels in the organizational hierarchy.
Proposition 9: The organization as a learning system must be designed
around the information domain and those parameters of information
such as complexity, interdependence, variety, uncertainty, and
information subsystems and databases.
The application of
management information systems (MIS), executive information
system (EIS) and decision support systems (DSS) to the right task
and types of managerial decisions is essential to the effectiveness of
learning organizations.
An important aspect of organizational structure is the way in which
the parts of an organization communicate and coordinate with one
another and with other organizations. Vertical linkages coordinate
activities between the top and bottom of the organization, while
horizontal linkages are used to coordinate activities across departments.
Advances in information technology can reduce the need for middle
managers and administrative support staff, resulting in leaner
organizations with fewer hierarchical levels. In some organizations, such
as Microsoft and Andersen Consulting, front-line employees
communicate directly with top managers through e-mail. Information
technology can also provide stronger linkages across departments and
plays a significant role in the shift to horizontal forms of organizing.
Coordination no longer depends on physical proximity; teams of workers
from various functions can communicate and collaborate electronically.
516
SANKAR
New technology enables the electronic communication of richer, more
complex information and removes the barriers of time and distance that
have traditionally defined organization structure. A special kind of team,
the virtual team, uses computer technology to tie together geographically
distant members working toward a common goal (Daft, 2001).
At the strategic level of the organization where there is high
uncertainty in decision outcomes, objectives, and the means-ends chain
more information processing is critical for strategic planning. The sensor
subsystem must be activated through a variety of information systems,
EIS, SIS (strategic information systems), DSS, and DDB (distributed
databases). At the tactical and operational levels of the organizational
hierarchy, planning goals and action goals respectively mean that the
other information subsystems which use other varieties of information
systems and data-bases may be more effective for semi-structured and
structured decision-making.
At each level, the information mechanisms such as channels,
networks, feedback loops and distributed data process (DDP)
configurations will vary with the elements of the information domain. At
the strategic level the mechanisms will require some degree of flexibility
with information channels, networks, and configurations linked to
decision centers. At each level of the organization, the strategic, tactical,
and operational, the information needs will vary because of the elements
of the task and the information domain. Linking information channels
and networks to decision centers is more effective than the traditional
linkage to power centers. Information emerges as the critical contingency
variable in organizational design rather than power and authority. The
structural configurations will also vary with the information domain.
Organizational decisions vary in complexity and can be categorized
as programmed or nonprogrammed (Simon, 1960). Programmed
decisions are repetitive and well defined, and procedures exist for
resolving the problem. They are well structured because criteria of
performance are normally clear, good information is available about
Organizational decisions vary in complexity and can be categorized
as programmed or nonprogrammed (Simon, 1960). Programmed
decisions are repetitive and well defined, and procedures exist for
resolving the problem. They are well structured because criteria of
performance are normally clear, good information is available about
current performance, alternatives are easily specified, and there is
DESIGNING THE LEARNING ORGANIZATION AS AN INFORMATIONPROCESSING SYSTEM
517
relative certainty that the chosen alternative will be successful.
Examples of programmed decisions include decision rules, such as when
FIGURE 2
Decision-Making and Organization Design
to replace an office copy machine, when to reimburse managers for
travel expenses, or whether an applicant has sufficient qualifications for
an assembly-line job. Many companies adopt rules based on experience
with programmed decisions.
518
SANKAR
Nonprogrammed decisions are novel and poorly defined, and no
procedure exists for solving the problem. They are used when an
organization has not seen a problem before and may not know how to
respond. Clear-cut decision criteria do not exist. Alternatives are fuzzy.
There is uncertainty about whether a proposed solution will solve the
problem.
Typically, few alternatives can be developed for a
nonprogrammed decision, so a single solution is custom-tailored to the
problem (Daft, 2001).
The decision about how to deal with charges of faulty Pentium chips
was a nonprogrammed decision. Intel had never faced this kind of
problem and had no rules for dealing with it. Many nonprogrammed
decisions involve strategic planning, because uncertainty is great and
decisions are complex (Pocanowsky, 1995).
Particularly complex nonprogrammed decisions have been referred
to as “wicked” decisions, because simply defining the problem can turn
into a major task. Wicked problems are associated with manager
conflicts over objectives and alternatives, rapidly changing
circumstances, and unclear linkages among decision elements. Managers
dealing with a wicked decision may hit on a solution that merely proves
they failed to correctly define the problem to begin with. Today
managers and organizations are dealing with a higher percentage of
nonprogrammed decisions because of the rapidly changing business
environment and globalization.
This information contingency model is designed (1) to increase the
information processing capacity of the organization (2) to provide
flexibility in information systems for coordinating the information flow
and workflow (3) to maximize the use of a variety of data bases for
information processing (4) to align the conjunction of informationcommunications channel with decision centers (5) to facilitate the use of
a variety of communications networks (6) to locus and integrating
information subsystems within a modular units of a complex learning
organization (7) to speed up the decision making process (8) to apply
cybernetic principles of feedback, control, and self regulation to design
information systems (9) to focus on the information domain rather than
power and authority as the design parameter (10) to assist in the design
of adaptive learning organizations in the context of the information
domain.
DESIGNING THE LEARNING ORGANIZATION AS AN INFORMATIONPROCESSING SYSTEM
519
The information-processing model shifts the emphasis from the
contingency design parameters. The optimal degree of differentiation and
integration is determined not only by the degree of environmental change
but also by the information domain, variety, complexity, change and
uncertainty of information. Environmental determinism is still critical
but so is the type of information domain that must be designed to cope
with environmental change and globalization. With reference to the
technological imperative, the organizational structure must cope with
technical complexity in the Woodward’s conventional paradigm and with
types of interdependence (pooled, sequential, and reciprocal a la
Thompson’s menu. Now the structure must also cope with computer
technology, integrated vs. distributed data-bases, information taxonomies
based on degree of complexity and variety of information inputs,
information processing capacity of information subsystems, variety of
DDP configurations, interdependence among the information
subsystems, feedback loops within structural configurations, elements of
task uncertainty and information processing needs and types of
information systems. Principles from cybernetics, the systems paradigm,
and information theory alluded to earlier, are more relevant and
informative as design principles for the learning organization than the
conventional paradigms of design.
CONCLUSION
The importance of this information processing view of
organizational design is that it determines the pattern of problems and
information needs within the organization. The information support
systems and organizational structure should provide information to
managers based upon patterns of decisions to be made. When the
organization is designed to provide correct amount and type of
information to managers, decision processes work well. When
information systems are poorly designed, problem solving and decision
processes will be ineffective and managers may not be able to solve the
problem. The processing of information and the conversion of
information into decision outcomes can be effectively facilitated if the
structural configuration matches the elements of the information domain
or the information needs of the task, its degree of complexity, variety,
and change. Information becomes the contingency variable rather than
authority and power for the design of modular units within complex
520
SANKAR
learning organizations, virtual teams and network systems. Shared
information is one of the building blocks of the learning organization. In
the move toward information, knowledge and idea based organizations,
information sharing reaches extraordinary levels, hence our major focus
on the design of the organization as an information processing system.
REFERENCES
Beer, S. (1959). Cybernetics and management. New York: Willey.
Cadwallader, J. (1985). The cybernetics analysis of change in complex
organizations. In Litterer, J. (Ed.). Organizational Analysis (pp. 132146). St. Paul, MN: West Publishing.
Chaffe, E. (1980). Decision models in university budgeting. Unpublished
Ph.D. Dissertation Stanford University, Palo Alto, California.
Daft, R. (2001). Organizational theory and design. St. Paul, MN: West
Publishing.
Daft, R. L. & Lengel, R. H. (1998). Organizational information
requirements, media richness and structural design. Management
Science, 32, 554-71.
Galbraith, J. (1995). Designing complex organizations. Reading, MA:
Addison-Wesley.
Garvin, D. (1993, July-August). Building a learning organization.
Harvard Business Review, 2, 78-91.
Goh, S., & Richards, G. (1998). Bench-making the learning capability of
organization. European Management Journal, 2 (2), 575-583.
Kelly, J. (1980). Organizational behavior: Its data, first principles and
applications. Homewood, IL: Richrad Irwin.
Luthans, F. (1996). Organizational behaviour. New York: McGraw Hill.
Miller, D. (1990). Organizational configurations: Cohesion, change and
prediction. Human Relations, 43, 781-789.
Pocanowsky, M. (1995). Team tools for wicked problems.
Organizational Dynamics, 23 (3), pp. 167- 175
Rogers, R. (1987). Organizational theory. Boston, MA: Allyn & Bacon.
DESIGNING THE LEARNING ORGANIZATION AS AN INFORMATIONPROCESSING SYSTEM
521
Sankar, Y. (2001). A value-based model of organizational design
(Working Paper No. 6). Dalhousie University. Canada.
Senge, P. (1990). The fifth discipline: The art and practice of learning
organizations. New York: Doubleday.
Simon, H. (1960). The new science of management decision. Englewood
Cliffs, NJ: Prentice Hall.
Tushman, M., & Nadler, D. (1978, July). Information processing as an
integrating concept in organizational design. Academy of
Management Review, 20 (2), 158-176).