Table of Contents
A causal look at the occurrence of biases
in strategic change
Ozge Pala’
Jac A.M. Vennix
Methodology Section, Nijmegen Business School
Thomas van Aquinostraat 5.0.77
PO Box 9108
6500 HK Nijmegen
The Netherlands
tel +31 24 3611470
fax +31 24 3611599
O.Pala@nsm.kun.nl
J.Vennix@nsm.kun.nl
Abstract
Information is very important for organizations in deciding whether to change their strategy. Hence,
information processing is a fundamental task, which should be done effectively. However, the vast
amount of available information coupled with the limited cognitive capabilities make such activities
less effective than desired. To reduce mental effort required to collect and analyze information,
organizations employ various biases and heuristics. Researchers, both in psychology and strategic
decision-making, point out the persistence of biases. Such literature streams, however, mostly pay
attention to the occurrence of one bias at a time even though some biases are dependent on each
other and occur simultaneously. The proposition of this paper is that the use of biases and heuristics
reinforce the factors leading to their use. The importance of this proposition is shown with a system
dynamics model by demonstrating that the isolated effects of two biases generate different results than
their combined effect.
1. Introduction
There has been an ample amount of attention to psychological biases and heuristics in the
literature streams of psychology and strategic decision-making. Various experiments have
been designed to show the existence of biases in different decision-making settings.
Researchers have devised classifications of biases based on the decision-making stage in
which they occur. Empirical work has shown that biases exist and have effects in different
stages of decision-making. Yet, the focus has been on individual biases rather than the joint
effects of multiple biases occurring simultaneously. The starting point of this work is the
assertion that biases are related to one another. They are not independent processes that occur
' Contact author
in their own account, but their occurrence depends on the occurrence of other biases.
Showing these links or dependencies amongst biases is important to improve our
understanding of the decision-making process. Moreover, recognizing these dependencies
could help us better understand the causes and consequences of biases and ways of
eliminating (or at least, decreasing) their adverse effects.
The purpose of this paper is two-fold.
The first purpose is to show that looking at biases jointly generates different insights than
looking at them individually. We have chosen strategic change as the decision-making
process to focus on. The biases involved are those that occur in information processing as
organizations decide on whether to change. Through system dynamics modeling, we will
show that looking at the change process by looking at the effects of multiple biases
simultaneously generates a different insight into the consequences of change. Our aim is not,
however, to show the totality of the links amongst all the biases. The literature and the
number of biases are far too large for that. Our purpose is to demonstrate that it is important
to take the interdependencies amongst biases into consideration.
The second purpose is to look at the phenomenon of commitment to strategy in more
detail. As will be shown with the model, commitment to a given strategy has an important
effect on the change behavior of organizations. Yet, in the literature, there is controversy on
how commitment escalates. We will show that different opinions regarding commitment lead
to quite different consequences. We will also look into the similarity between commitment to
strategy and inertia concepts that come from different literature streams.
The paper is organized as follows. In section 2, we will focus on the importance of
information for strategic change and the biases in information processing. It is in this section
that we develop our proposition. In section 3, the model structure is explained. Section 4 is on
model analysis. In section 5, we will look into the concept of commitment of strategy.
Finally, section 6 we will present the limitations of the model and analysis and present the
conclusions.
2. Importance of Information for Strategic Change and Biases in Information
Processing
Information is one of the key components of strategic decision-making. Ungson et. al. (1981,
p117) define information as “stimuli (or cues) capable of altering an individual’s expectations
and evaluation in problem solving and decision making”. Organizations spend much time in
gathering, storing, analyzing, and communicating information during their decision-making
activities. Some researchers even argue that information gathering and processing are more
crucial to the success of the firm than strategic decision making itself (Pfeffer and Salancik,
1978; Starbuck and Milliken, 1988).
Organizations need two types of information: environmental (external) information (to be
able to detect trends and changes in their environment and make decisions) and internal
information (to evaluate their performance). Organizations need relevant, timely, and
accurate information about their environment. Successful organizations are those that can
devise strategies and structures to fit the environmental conditions. Since organizations
perceive their environment through environmental information, formulation of an appropriate
strategy depends on identifying the relevant environmental factors that will help detect the
important trends in the environment (Ungson et. al. 1981). Along with understanding their
environment, organizations also need to assess their current performance. Internal
information is required for the latter. Information on the performance levels, resource
availability etc. is very important for identifying how the organization is functioning given its
current strategy and determining whether the organization is in alignment with its goals. The
mixture of internal and external information would help the organization identify the gaps in
performance/fit and possible adaptation(s) it has to make.
Even though information processing is very crucial to the survival of organizations and its
importance is recognized, the process is not without problems. There are many factors that
seem to render the identification of relevant information ineffective. First, there is an
overwhelming number of potential information sources in the environment (Choudhury and
Sampler, 1997) each including enormous amounts of information. Ideally, human beings
should pay attention to cues proportional to the importance and utility of different cues
(Sterman, 2000). However, decision-makers have limited scanning capacity and limited
resources. The mismatch between the amount of available information and the limited
information processing capabilities leads to unfavorable circumstances such as information
overload and stress (figure 2, the factors leading to the use of biases and heuristics). To
simplify the stressful situation faced and to reduce the mental effort required, decision-
makers resort to biases and heuristics (in figure 2, solid line from the factors leading to the
use of biases and heuristics to the use of biases and heuristics). As a result, they pay attention
to information selectively. For instance, they choose to acquire and use information that
confirms previously implemented policies (Staw and Ross, 1987; Kiesler and Sproull, 1982).
Or they gather more of the same kind of information to increase confidence in their judgment
(Hogarth and Makridakis, 1981) rather than collecting information to identify what is really
happening in the environment. Such simplifying procedures (e.g., looking for confirming
evidence or using simple mental maps) could, in the short run, take away the pressures
caused by information overload, but they obstruct the identification of relevant information
by restricting the focus of the organization and lead to biases.
The extant amount of literature on information processing biases focuses on the factors
leading to the use of biases and heuristics and the consequences of individual biases. As such,
the interactions between the biases and the interdependencies amongst them have not been
looked at. We believe that the use of biases and heuristics further reinforce the factors leading
to the use of these biases and heuristics (Figure 2, dashed line). This link does not mean that
use of heuristics will affect all the factors identified in the box. We mean that the use of these
biases can further increase the stressful situation and the subsequent mental effort required.
And as a result, the use of the same or other heuristics and biases will be reinforced. Hence,
we propose that the biases and heuristics literature should look at the occurrence of biases
simultaneously rather than analyzing one bias in isolation from all the other ongoing biases.
Complexity of reality
© — Time limitations
e Uncertainty
Unfavourable circumstances | <Q
e Information overload x
© — Stress etc. N
\
Use of biases and heuristics
Cognitive capabilities in decision-making
e Lack of ability to forecast outcomes
© Cognitive limitations to amount of data that
can be processed
e Inability to think in terms of feedback
structures
e Inability to infer dynamics from structure
FACTORS LEADING TO THE USE OF BIASES AND
HEURISTICS
Figure 2: Positive feedback loop reinforcing the factors leading to the use of biases and heuristics
We, of course, need to show that such a proposition (i.e. looking at joint effects of biases) is
indeed important. If we can demonstrate that isolated effects of two separate biases generate
different results than the combined effect of these biases then we can say that our proposition
is of importance. In a sense, we are testing our proposition by using a system dynamics model
and showing that occurrence of one bias vs. multiple biases in information processing for a
strategic change process leads to different outcomes for the organization. Hence, any analysis
of strategic change that lacks the inclusion of the interdependencies amongst biases would
generate different results than one where such mechanisms are considered. This would mean
that if such interdependencies exist they should be taken into consideration.
In the following section, we will explain the structure of the model used for analysis.
3. Structure of the Model
In this paper, our concern is the joint effect of information processing biases on the decision
of whether to go on with a strategy or change it. We will now look at the structure of the
model and explain the change-generating mechanisms and the biases involved.
3.1 Base Model: Change-generating mechanisms
Organizations change their strategies if they feel pressure to change. The most important
factor leading to the accumulation of this pressure is the performance problems the
organization might be experiencing. The fit of the organization’s strategy determines the
performance of the organization’. A strategy is appropriate or fit when it is in alignment with
2 There are two dominant views on the determinants of organizational performance. The contingency theory
emphasizes the importance of organization-environment fit as a determinant of performance, with the view that
the appropriate organization form is contingent on the characteristics required by the environment (White and
Hamermesh, 1981, Drazin and van de Ven, 1985). Thus, to be effective, the organization should achieve a fit
the competitive conditions the organization is embedded in. If a change takes place in the
environment and this change would generate a misfit between the organization’s strategic
orientation and the competitive conditions, this would be a reason for strategic change
(Tushman and Romanelli, 1985). As long as the strategic orientation of the organization fits
the one required by the environment, the organization converges around its current strategic
orientation. If for some reason (e.g. change in the environment) the strategic orientation turns
out to be inappropriate then the pressure would start accumulating leading to the start of
organizational change. Once the changes started are implemented, the misfit can be
eliminated. This process that leads to the generation of strategic change is shown in figure 3.
Fit wear off
time pulse amount
Change _ Fit decrease due to-
environmental change
Required
performance Performance
“a gap
Average
performane
23
4) formance
Acceptable
] pearl correction
Implementation performance gap . }-
time = ial Delay
Relative performance gap
performance gap
Perceived pressure \
to change
+
Exerted pressure if
Presute tole ——c)__to change
| change | Pressure:
— dase Table for pressure by
Soe chang performance gap
~ -\ Pressure build
up time
Figure 3: Change-generating loop
Fit of strategy — signifies the appropriateness, effectiveness, alignment or fit of an
organization’s strategy with respect to the external environment. This is a state variable,
because it is the accumulation of the effects of past strategic changes and the changes in the
environment. It ranges between 0 and 100. Zero means that the strategy of the organization is
totally in misfit with its environment whereas 100 signifies perfect fit. Fit can change in two
ways: change in strategy (implementation of the ongoing changes) and environmental change
(e.g. a new law, a better technology, a new product). The fit of the strategy determines the
performance. Hence, the performance gap actually refers to the amount of misfit of strategy.
It is assumed that the organization strives for perfect performance (of 100).
Pressure to change — signifies the amount of pressure the organization experiences regarding
whether to change or not. Pressure to change is a state variable since it signifies the
cumulative pressure that builds up over time. The main source of pressure is the average
performance gap (Tushman and Romanelli, 1985). As the average performance gap increases
so does the pressure to change.
between the environmental requirements and organizational strategy and structure. Congruency theory, on the
other hand, stresses the importance of internal consistency between elements of structure (Child, 1972). How
well the organization functions internally would affect how well it performs. In this paper, we have decided to
limit ourselves to the contingency theory to keep the model simple. Hence, performance is determined by the fit
of strategy.
Change in process — signifies the amount of ongoing changes in the organization. It is
measured in terms of the amount of fit gap to be closed as a result of the on-going changes.
Change in process increases as new changes are started due to the accumulated pressure to
change and decreases as the changes are completed.
3.2 Biases and heuristics
We will now move our attention to the biases that are represented in this model. We do not
claim to include all the biases into our model. The model should be seen as a first attempt in
identifying the linkages between information processing biases. The totality of the
relationships and loops that are in the model are shown in figure 4. For the full model
specification is included in the appendix.
Below, we will look at external and internal information processing separately. However,
we first need to explain the commitment to strategy concept since it has an effect on both the
external and internal information processing.
Fit wear off
time pulse amount
Expected
_> performance
Y
(—# perceived «—
performance
Normal ~
confidenc
table information
overload
4 Intensity of search for
__ started
confirming evidence hoectaa
S/o eR ertgenane
Infrmaton | table
Tendency to relate
sto strategy
Table for
‘Acceptable tendency
Implementation performance a 7
* time ie oP Delay
Relative performance gap
performance gap
a | comin |
f Asaittios «= : ita pri /
| istration ion /
‘= [Commitmen| } Table for pressure by Vi
Information to strategy performance gap
‘Commitment
Comanitmerbg
Shc spb? ee ieee
7 \ oes vo
| build-up time — ” Pressure build
oe i ee uptime
procesingenpacity XY) ae
Effect of Me
Table for eff
information processing »
‘capacity
Figure 4: Biases in change process — The whole model
Commitment to strategy — signifies the extent of support for the current strategy. There are
two main streams of literature that explain how commitment to a certain action increases. The
trial and error learning literature assumes that behavior associated with success is repeated
whereas behavior associated with failure is not (Lant and Mezias, 1992; Levitt and March
1988; Lant and Hurley, 1999). The escalation of commitment literature, on the other hand,
states that even unsatisfactory performance leads to persistence and further commitment to a
course of action* (e.g. Staw, 1981, Staw and Ross 1987). The latter literature stream states
that decision makers want to justify their prior action by committing more resources as a
means of showing that they were not mistaken by their initial decision. Moreover, people
seem to be attached to sunk costs (money, time, and other resources) that have already been
devoted to a (losing) course of action. Even though such persistence may be beneficial in
> Even though escalation of commitment literature primarily focuses on responses to negative feedback, Staw’s
(1976) original ideas predicted that escalation would occur both in positive and negative feedback situations.
stable environments, in the face of environmental change, it might prove to be harmful since
it would disable the organization from changing.
In this paper, we represent this concept as commitment to strategy. Commitment is a
stock variable since it builds up as a result of previous devotions of the decision-maker to the
particular strategy. We assume that commitment increases continuously over time (with a
time delay of approximately 1 year). This assumption reflects the idea that as the organization
continues to use the same strategy, more time and resources are devoted to that strategy and
hence, the commitment will increase. At the same time, commitment can decrease as well.
Commitment of an organization is to a given strategy. When a withdrawal from a strategy
takes place (ie., if the strategy is changed partially or abandoned completely) then the
commitment should decrease accordingly. Hence, in the model, commitment decreases
proportional to the amount of changes made to the strategy’. If the changes made are large so
will be the decrease in commitment.
It should be pointed out that we are taking the escalation of commitment literature (rather
than the trial and error learning literature) as a basis for our commitment construct. Hence,
possible effects of performance on escalation are not represented in the structure.
Performance effects on (de-)escalation are a matter of debate in the commitment literature.
We will come back to this issue later in section 5.
Commitment to strategy has effects on both external and internal information processing.
We will first look at its influence on external processing and then, internal processing.
Biases in external information processing — Confirmation bias and (over)confidence —
External information processing refers to the search for environmental information to
determine whether the organizational strategy is fit to the environment. It has been shown
previously that individuals and organizations look for confirming information. This is also
known as the confirmation trap, “individuals tend to seek confirmatory information for what
they think is true and neglect the search for disconfirmatory evidence” (Bazerman, 1998).
What is important for our purpose, however, is why people search for confirming evidence.
Caldwell and O’Reilly (1982) state “aside from justifying one’s actions through continued
resource allocation, trapped decision makers may also buttress their choices by actively
seeking out supportive information to rationalize their position and by constructing
arguments that justify their choice and undermine alternative approaches that might have
been selected”. Kiesler and Sproull (1982) state that organizations value their commitment to
their policies and procedures more than data. They have shown that managers accept
environmental information that confirms their beliefs to be diagnostic and they do not pay
attention to disconfirming information: “Motivationally driven social cognition theories
would suggest that even had the [disconfirming] information been available, executives
would have either attacked its validity or ignored it because of their commitment to their own
program” (p558). They, moreover, state that when managers have heavily invested in any
situation, they are likely to discount information about the environmental changes detrimental
to that situation. Hence, if decision-makers believe in a certain course of action then the
information processing will be directed at the confirming evidence disregarding the
disconfirming information (Staw and Ross 1987; Caldwell and O’Reilly 1982; Conlon and
Parks, 1987).
Along with the effect of commitment, another information biasing takes place due to
information overload, which is caused by the lack of balance between available information
and information processing capacity (Hogarth and Makridakis, 1981; O’Reilly, 1980). This
* Amount of change in fit is compared to the existing fit. The ratio of the two gives a proxy of the extent of
change that has been made to the strategy.
overload can be overwhelming, especially if the organization perceives itself to be under
threat. Smart and Vertinsky (1977, see pages 640-641) have proposed that under stress, the
cognitive abilities’ and information processing abilities are reduced, leading to information
overload which decreases the quality of information. The threat-rigidity theory (Staw et. al.
1981) also points to the linkages between threat/stress and information processing
capabilities: ““a threat may result in restriction of information processing, such as narrowing
in the field of attention, a simplification in information codes, or a reduction in the number of
channels used” (p502). When the threat is felt, the information search decreases due to the
overloading of communication channels. Organizations rely upon prior expectations and
hence, restrict their information processing to information that supports the chosen policy.
The search results in information that is similar to that of the past (Starbuck et al., 1978).
In the model, both relationships are represented (see figure 5). Increase in commitment to
strategy leads to an increase in intensity of search for confirming evidence. On the other
hand, the threat the organization feels, as represented by the perceived pressure to change,
leads to a reduction in the information processing capacity and hence, increase in information
overload, which, in turn, constitutes the second factor increasing the intensity of search for
confirming evidence. The more confirming evidence the organization has, the more confident
it becomes with the chosen strategy (Einhorn and Hogarth, 1978; Hogarth and Makridakis,
1981; Schwenk, 1986). This confidence counters the perceived pressure to change. The more
the confidence is the less likely will be change. These relationships add further feedback
loops into the model.
The link between pressure to change and information processing capability (figure 5, the
inner loop) is represented as a bell-shaped curve as the literature dictates (e.g. Staw et al.,
1981). Being under high pressure, the organization would like to get out of the unfavorable
situation as soon as possible. Hence, the vigilance in information search and processing
would decrease, leading to an organization that has less processing capability both in terms of
capacity and quality. On the other hand, when the pressure is very low, the organization does
not feel the need to scan the environment and thus, has low processing capacity and quality.
As a result, the capability is low at low values of perceived pressure and it increases as the
pressure increases. But if pressure rises above a certain level then the capability starts to
decrease. Hence, a bell-shaped curve. It would be possible and desirable to separate these
positive and negative relationships between these two variables into two causal pathways. For
instance, it could be that when pressure increases the stress increases leading to lower
capability and when pressure increases the need for new information increases leading to
higher capability. However, since the literature is silent regarding these paths, we have
decided to use the bell-shaped curve that is used in the literature.
* Cognitive abilities are “the abilities of the decision unit to interpret information, generate options creatively,
calculate and make choices between alternative courses of action” (Smart and Vertinsky, 1977).
Change
ercentage
Commitment
to strategy
f Average
J performance gap
Perceived performance
\
Fitof Information
strategy handling capability
zs
Change in fit Information
+ overload
Changes 4
completed Intensity of search for
A confirming information
Changes :
ingivess Volume of
confirming
information
New changes
acs 4
Sead Confidence in
strategy
Figure 5: Effects of commitment and information overload on confirmation bias and confidence
Bias in internal information processing — Tendency to relate success to strategy — As
commitment affects the external information processing; it also affects the internal
information processing. Internal information processing refers to the information collection
that leads to the perceived success of the strategy used. As external information, internal
information can be biased too. If decision-makers are committed to their strategy then they
will believe that the strategy is successful. As Schwenk (1984) states, “Decision-makers who
believe that the company’s current strategy is successful may ignore information suggesting
gaps between performance and expectations”.
In the model, this tendency, named as the tendency to relate success to strategy, is
determined by the commitment to the strategy (figure 6). Higher commitment leads to a
stronger tendency to show that the strategy is performing well. Consequently, the perceived
performance is biased towards the expected performance.
We need to stress that perceived performance signifies two effects. In system dynamics,
we use perceptions to refer to time delays in collecting and processing information. Decision-
makers cannot know the instantaneous performance; hence, the smoothing of information is
incorporated into the perceived performance. However, the time delays are not the only
factors playing a role in determining the perceptions. Bias is also part of the equation. If the
decision maker is biased, for instance, in her information processing, so will be the perceived
performance. This bias is incorporated by the tendency variable, which determines the
relative weights the decision maker gives to the expected performance and the real
performance measure (which is the average fit of strategy). Hence, the perceived performance
is a weighted average of the expected performance and the average fit where the tendency is
the weight given to the expected performance (see equation 1). As a result, the more biased
the decision-maker is, the less will be weight for the real performance and the performance
will be perceived to be higher than it actually is’,
Perceived performance = Average fit of strategy *(1-Tendency) + Expected performance * Tendency
Equation I: The perceived performance
° Of course, if the real performance is already high (close to the expected performance level), the bias will not be
as recognizable.
| Commitment
Change PP) ‘to strategy
percentage Wi
‘ :
A>) Tendeuy tomate
FitoF =) success to strategy Expected
strateg a ee ia performance
Change in fit~* Perenved: =a
" performance
Changes
completed Average perfor
7 mance gap
Changes
in process Pressure
he changes ta chanoe
+ started —<—
=a
Figure 6: Tendency to relate success to strategy
Having explained the model structure, we will now move on to model analysis.
4. Model Behavior/Analysis
As reactive theories of change, such as the punctuated equilibrium model (Tushman and
Romanelli, 1985) dictate, we will look at organizations’ change behavior in reaction to an
environmental change. We will show that looking at the effects of single versus multiple
biases in strategic change will make a difference for the outcomes of change processes. We
will consider two different environmental change patterns (Brown and Eisenhardt, 1997):
Environment 1: One time (punctuated) change in the environment such as the sudden
emergence of a breakthrough technology. This is created with a pulse function in the fit
decrease due to environmental change. In the graphs, this type of change is referred to as
pulse.
Environment 2: Continuous, incremental changes in the environment such as constantly
shifting market opportunities. This is created by defining a fit decay function in the fit
decrease due to environmental change. The fit decays over time with an average lifetime of
200 months. In the graphs, this type of change is referred to as decay.
4.1 Base run
The base model represents the mechanism that will lead to the start of change in the face of
environmental change (change-generating mechanisms explained in section 3.1). In this
structure, it is assumed that no biases interfere with the decision of change. The model is
reactive in the sense that given a change in the environment, the organization would start
changing to restore the fit of its strategy. If we assume that there are no biases interfering
with the decision process, what would be the change behavior of the organization? The
theories predict that once the organization realizes the lower performance, it will start
changing. In figure 7 and 8, the behavior patterns under the two environmental conditions are
presented. As expected, in an environment with one radical change (Pulse base change), the
organization starts changing after a while and the change goes on till perfect fit is restored
(after which the change comes to a halt) (figures 7 and 8). Change does not start immediately
after an environmental change since it takes a while for the pressure to build up. In decay
base change, an organization in a continuously changing environment is simulated. Theories
of organizational change are not specific about the reaction of organizations in such
environments. Yet, Brown and Eisenhardt (1997) state that the incremental changes in the
environment render the discontinuous reactions from organizations ineffective and
organizations should learn to change continuously to be able to survive. This is what we see
in decay base change. The organization starts changing to counteract the on-going
environmental change and keeps on changing continuously (and incrementally) (figure 8) and
eventually balances its fit (figure 7). At equilibrium, the amount of change started is at the
magnitude of the fit erosion due to the environmental change. Hence, the resulting
equilibrium fit level is dependent on the speed of environmental change. The slower the
environment changes, the lower will be the gap between the equilibrium fit and the perfect
fit. Other than the speed of environmental change, the implementation time influence this gap
as well. If the change were implemented faster then the resulting fit would be higher.
Base Run - Fit of strategy Base Run - New changes started
100 4
85 295
70 19
35 os
40 02
0° 30 6 9% 120 150 180 210 240 270 300 ° 30 0 9 120 150 180 210 240 270 00
Time (week) Time (week)
performance units New change started : Pulse base ch
performance units New change started : Decay base chan
yerformance units!
performance units!
Figure 7: Base run —Fit of strategy Figure 8: Base run- change in fit
Under the assumption that there are no biases interfering with the decision-making process,
the organization restores its strategic fit. However, as the extant literature shows, biases and
heuristics do play a role in strategy making. Hence, we should consider their effects.
4.2 Effects of external information processing biases: Confirmation bias and confidence
We will first look at the effect of the confirmation bias and resulting (over)confidence. The
confirmation bias takes place when the organization collects confirming evidence due to
commitment to its strategy or as a reductionist technique to cope with the information
overload. The more confirming evidence an organization collects, the more confident it
becomes in its strategy and hence, opposes the change. The interplay between the confidence
and the pressure to change determines the amount of new changes started.
In Figure 10, the results for pulse and decay environments are given (compared to the
base runs presented in section 4.1, the simulation period is extended to see the overall
behavior patterns better). In the pulse case, we see that the organization does not change at
all. The reason is that the commitment to the strategy leads to the search for confirming
information that subsequently leads to confidence. The confidence withholds the organization
from changing (since only when the felt pressure is higher than confidence, change starts’).
Since every period with no change leads to higher commitment and higher resulting
confidence, the organization does not withdraw from its current strategy. The simulation
results show that the organization can go on with a low-fit strategy. However, we should also
7 The strength of confidence or rather, the strength of the effects of commitment and information overload on
confidence is very important here. If these relationships would be less strong then the confidence will be lower
constituting a lower threshold to change.
consider that in real life, this might not be possible. An organization that does not fit its
environment is bound to lose business and be defeated by its competitors. Hence, it is
possible that if the organization does not change, its fit decays further over time and the
organization eventually dies.
The model has been tested with different levels of initial commitment and amounts of
pulse. The initial commitment level does not have an effect on the results. A higher amount
of pulse, for instance a loss of 70% of the fit, does lead to a change reaction restoring the fit
back to its maximum level. This means that it takes a large amount of fit gap to generate
enough pressure to overcome the confidence the organization has to its strategy.
In the decay environment, on the other hand, we see that the organization goes through
periods of change rather than changing continuously. Before starting a change, the
organization waits till its fit is quite low rather than changing as the environment changes.
This is due to the high confidence caused by high commitment to strategy and low
information processing capabilities. In the face of no change, the commitment to strategy
keeps on building up and leads to the reinforcement of the confirmation bias and confidence.
The pressure, on the other hand, keeps on increasing in the face of decreasing fit and leads to
decrease in information processing capability, further information overload, and hence,
confirmation bias and confidence.
Confidence run - Fit of strategy
100
0 100 200 300 400 500 600 700 800 900 1000
Time (week)
Fit of Strateg
Fit of Strateg
‘ulse conf change =§ performance units
ecay conf change = performance units
Figure 10: Fit of strategy for an organization that is under the influence of confirmation bias. Two
environmental conditions are considered: one-time change environment (pulse conf change) and
incremental change environment (decay conf change).
4.3 Effects of internal information processing bias: Tendency to relate success to change
We will now look at the effect of internal information processing bias on the organizational
change process. As explained before, being committed to a strategy, the organization might
not perceive the performance indicators accurately and perceive the performance to be higher
than it actually is. The base run is a special case where the tendency is assumed to be 0,
meaning that the perception is not biased at all. In Figure 9, the behavior in pulse and decay
environments is given.
As can be seen, the tendency to relate success to strategy affects the change process. In
both cases, the tendency to relate success to the strategy leads to a delay in the start of the
change as well as a sub-optimal strategy once the change ends. The tendency makes the
organization perceive its performance better than it actually is. Pressure does not accumulate
fast enough and therefore, the change is started later in time. Even when the change process
begins, the organization initiates fewer changes than required due to low pressure. Fewer
changes mean longer duration before the fit can be restored and during this period,
commitment keeps on building, further reinforcing the tendency to perceive the strategy
successful. As a result, under both circumstances, the organization has low performance for a
long period of time and ends up with a low equilibrium fit’. This is especially problematic for
an organization in the decay situation where the organization is much too late in
counterbalancing the on-going decay in fit. (Note: In the model, the expected performance
that is used for the tendency is assumed to be the perfect performance of 100. Obviously, if
the expected performance were lower, the effect of tendency would be decreased.)
Tendency Run - Fit of strategy
100
85
70
55
40
0 100 200 300 400 500 600 700 800 900 1000
Time (week)
Fit of Strategy : Pulse tend change § ———————————_ performance units
Fit of Strategy : Decay tend change — performance units
Figure 9: Fit of strategy when the organization has a tendency to see its performance better than it actually is.
Two environmental conditions are considered: one-time change environment and incremental change
environment.
4.4 Combined effects of internal and external information processing biases
So far, we have looked at the biases occurring in only internal processing or only external
processing. However, in an organization, these two processes take place simultaneously.
Hence, so do the biases. In figures 11 a and b, the behavior patterns corresponding to the 4
different runs (no bias, only internal information processing biases, only external information
processing biases, external and internal information processing biases together) are
compared. Figures 11 a and b correspond to pulse and decay environments, respectively. We
see that under both circumstances, the joint effect of the internal and external processing
biases leads to the demise of the organization. In the pulse case, the organization does not
change at all and keeps on having a low-fit strategy. In the decay case, on the other hand, the
organization continuously loses fit and eventually dies.
The tendency bias generates a delay in the building up of the pressure. The confirmation
bias and confidence on the other hand generate a high barrier to the start of a change process.
During this combined delay, the commitment keeps on building further reinforcing both the
internal and external processing biases. The ever-increasing pressure, on the other hand,
causes information overload leading to further confirmation bias and confidence. The result is
an organization that cannot generate a change process. The combined effect of the biases
generates a worse situation than were we not to look at the biases or look at them in isolation
from one another.
* Once again, the implementation time has an effect on both of these measures. Lower implementation time
would mean faster recovery of the fit. Yet, with respect to the comparison of the behavior of base vs. tendency
cases, the value of the constant does not make a difference.
Overall comparisons - pulse
100
85
70
55
40
0 100-200» 300-400 500 00-700» 800-900-1000
Time (week)
Fit of Strategy : Pulse base change. §=§ —————————————_ performance units
Fit of Strategy : Pulse tend change. §=§ performance units
Fit of Strategy : Pulse conf change —— performance units
Fit of Strategy : Pulse whole change @// performance units
(a)
Overall comparisons - decay
0 100 200 300 400 500 600 700 800 900 1000
Time (week)
Fit of Strategy : Decay base change §=§ performance units
Fit of Strategy : Decay tend change performance units
Fit of Strategy : Decay conf change ——ES performance units
Fit of Strategy : Decay whole change I performance units
(b)
Figure 11: Fit of strategy compared in four different runs: no biases are assumed to occur (base run), only
internal information processing biases are considered (tend), only external information processing biases
are considered (conf), internal and external information processing biases are considered (whole). Two
environmental conditions are considered: (a) one-time change environment (pulse) and (b) incremental
change environment (decay).
5. Further Look into Commitment
In the previous section, we showed that looking at biases simultaneously rather than in
isolation and basing analysis on feedback loops generates interesting insights. Such an
analysis shows us the effects of the interactions of the biases.
When only external or only internal information processing biases are considered, the
organizational change process leads to lock-in to sub-optimal strategy in a one-time change
environment and to an ineffective manner of coping with environmental change in a
continuously changing environment. When all the biases are taken together, the organization
cannot recover to its full strength and in certain circumstances, it dies. These results would
suggest that since all organizations suffer from biases, there would be no organization to
perform at its full strength. This could indeed be true since the successful organizations are
those that outperform their competitors. A most successful company in a given environment
does not necessarily perform at its best but it does perform better than others in the same
environment. Hence, the fittest organizations survive. Looking at the results involving the
decay environment, we can further speculate that all the organizations in a continuously
changing environment would die. This, of course, is not necessarily true although as the
population ecology literature would advocate, it is possible that organizations die and new
ones replace them. Yet, there are organizations that survive in continuously changing
environments. There can be two reasons to why we cannot replicate this case. First of all, it
might be that such organizations have different change generating mechanisms than the
reactive one we have assumed in our model. The other reason could be the way such
organizations handle commitment to strategy controlling its adverse effects in difficult times.
The latter factor is an interesting one to look into since by managing their commitment,
organizations can cope with biases. In this way, they can also generate new change
mechanisms.
Analysis of section 4 demonstrates the importance of commitment to strategy in determining
the course of a change process. Commitment and its effect on the other biases can explain
why change attempts take too long to start (and why organizations do not change and die),
why organizations get locked into low performing strategies, and why change efforts do not
accumulate to bring an organization to a full-fit strategy.
In the organizational change literature, organizational inertia is given as the main cause
behind these behavioral outcomes. Inertia is seen as the resistance of an organization to
change (Hannan and Freeman, 1984). Yet, inertia is a concept that has not been analyzed in
detail. The causes of inertia and the policies organizations can use to decrease their inertia
(when in trouble) are not very clear. There are interesting similarities between the concept of
inertia and commitment to strategy. It could be that the causes leading to commitment to
strategy are similar or the same as the causes leading to the accumulation of inertia. Hence,
links between the inertia literature and the escalation of commitment literature might prove to
be fruitful.
Different organizational theorists have different opinions of inertia. Some suggest that inertia
is resistance to all kinds of change and hence, organizations cannot change (e.g., Hannan and
Freeman, 1984). If the environment were to change, organizations with high inertia would
die. Others suggest that inertia is resistance to all other than radical change (e.g., Tushman
and Romanelli, 1985), hence, organizations can change radically if faced with high
performance gaps. The same dichotomy seems to exist in the commitment literature. Some
suggest that commitment to a course of action will disable the organization from changing
(e.g., various works of Staw and his colleagues). Others suggest that in the face of
performance pressures, the organization will withdraw from the current action and adopt
another (e.g., Northcraft and Wolf, 1984 and Lant and Hurley, 1999).
In our analysis, we have assumed that performance pressures do not affect the commitment to
strategy (following the line of reasoning by Staw et al.) or rather, we can say that even in the
face of low performance, the commitment keeps on building up. It is only when the strategy
changes that commitment declines. This approach to commitment leads to the death of the
organization. Such behavior patterns are supported by case studies of Staw and his
colleagues, where organizations committed to a course of action kept on supporting it, even
though the organization had very low performance results. These effects are rather similar to
those suggested by the organizational researchers advocating the detrimental effects of
inertia.
Although supported in the literature, failure due to commitment and inertia is not the only
behavior pattern possible. Some organizations do change successfully. Whereas the inertia
literature is silent regarding how inertia can decline to allow change, there is information
regarding how commitment might be declining in some organizations. There is a point of
view regarding how performance affects commitment. Some researchers support the idea that
in the face of negative performance feedback, commitment de-escalates and organizations
withdraw from action. Once again, these researchers have presented real cases where such
withdrawal took place.
We have incorporated this factor into our model to see the dynamic consequences of different
forms of commitment determinants. We have assumed that the fraction of decline in
commitment is dependent on the perceived pressure to change. Hence, when the organization
is under high pressure (meaning that the organization perceives a large performance gap) the
commitment decline is also high. Figure 12 represents the behavior of an organization in
which performance feedback affects the commitment to strategy is compared to one in which
performance pressures are of no effect. Figures 12 a and b refer to the pulse and decay
environment, respectively. As the literature suggests, the use of performance feedback to
update commitment to strategy enables change. When the commitment is decreased the
organization becomes less biased both in internal and external information processing tasks.
As a result, the organization gets a more “correct” perception of its environment and
performance. Thus, the change starts.
Yet, we should realize that the late reaction to changes in the environment still presents
itself as a problem. Moreover, in the continuously changing environment, due to this delay in
reacting to change, the organization cannot achieve continuous change.
Cause of Commitment Decline - Pulse
100
0 100° 200-300 400-500 G00 700 800-900-1000
Time (week)
Fit of Strateg
Fit of Strateg
uulse whole change performance units
ulse whole perf performance units
(a)
Cause of Commitment Decline - Decay
100
75
50
0 100-200-300 400 500-600-700 800-900-1000
Time (week)
Fit of Strateg
Fit of Strat
Decay whole change
Decay whole perf
performance units
performance units
(b)
Figure 12: Consequences of different determinants of commitment. Two environmental conditions are
considered: (a) one-time change environment (pulse) and (b) incremental change environment (decay).
At this point, the reader might be interested in knowing what this second formulation means
for our results of section 4. The comparison for different sub-models is represented in figure
13. As can be seen, considering the effects of individual vs. multiple biases does still affect
the consequences experienced by the organization. In both environments, there is a
considerable long delay before the change starts (hence, the organization has performance
problems for a long period, not a desirable situation for any organization) and in the
continuously changing environment, the organization is not able to generate effective changes
that will bring it in phase with the environment.
Overall comparisons - pulse
100
40
0 100-200-300 400-500. G00 700» 800-900-1000
Time (week)
Fit of Strategy : Pulse base pert. §=§ —————————_—— performance units,
Fit of Strategy : Pulse tend perf performance units
Fitof Strategy: Pulsecofpefif _—_—_—— performance units
Fit of Strategy : Pulse whole perf A performance units
(a)
Overall comparisons - decay
0 100 200 300 400 500 600 700 800 00 1000
Time (week)
Fit of Strategy : decay base perf. ——————————————_ performance units
Fit of Strategy : decay tend perf performance units
Fit of Strategy : decay conf perf performance units
Fit of Strategy : decay whole perf @EEEEEEEEEEND performance units
(b)
Figure 13: Fit of strategy compared in four different runs when the commitment is affected by the
performance pressures: no biases are assumed to occur (base run), only internal information processing
biases are considered (tend), only external information processing biases are considered (conf), internal and
external information processing biases are considered (whole). Two environmental conditions are
considered: (a) one-time change environment (pulse) and (b) incremental change environment (decay).
6. Discussions and Conclusions
The analysis of the joint effects of multiple biases is missing in the literature. There is a
strong need for such a perspective since only by looking at the whole picture, further insights
can be gained as to how biases affect the organizational outcomes. In this paper, we have
looked into the joint effects of information processing biases on the strategic change process.
We have shown that looking at isolated effects of individual biases generates different
behavioral outcomes than looking at the simultaneous effects of multiple biases. From a
system dynamics perspective, it might be obvious that the behavior patterns change when
new feedback loops are added. Yet, showing the differences generated in behavior patterns is
very valuable for setting research goals for behavioral decision-making and organizational
change literature streams. As such, we are able to show that the analysis and research focus in
these areas should be shifted towards understanding the joint effects of biases and how to
overcome one bias in the presence of others.
We have also seen that as we added more biases into our analysis, the results got
progressively worse. For instance, the reaction time to start changing increased and
organizations got locked into low performing strategies. This raises the question of what
would happen if we were to extend this model to include other biases we did not look at so
far. Would the results get worse and worse? Even though we cannot answer this question
without doing the modeling, the analysis of section 5 could generate some insights.
Comparing the results where only the tendency to relate success to strategy bias (called
tendency bias from now on) was active to the results where all the biases were active, we can
spot two counter-intuitive results. First of all, under the influence of all the biases, the
organization is able to attain a higher fit than under the influence of only the tendency bias.
This might mean that the confirmation bias and the resulting confidence in strategy actually
take away some of the negative effects of the tendency bias. The fact that the organization
waits a while before reacting leads to high pressure, which generates the necessary amount of
change to reach a full-fit strategy in the pulse case. In the decay case, on the other hand, the
organization does achieve higher fit levels. Yet, these are temporary and the organization has
periods with lower-fit strategy as well. It would be interesting to investigate which of the two
situations is more beneficial to the organization. Is it better to have a continuous moderate fit
or to have periods of high fit alternated with periods of low fit?
We can also extend this discussion to include the possible effects of factors that are not
included in the model. In the situation where only the tendency bias is considered, the
organization keeps on changing continuously to counter-balance the environmental change.
In the situation where all the biases are active, the organization has periods of change and
periods of stability. Literature on success of change would support the idea that most change
processes end up being unsuccessful because the changes cannot be effectively implemented
due to the resistance of the employees (Nadler 1981). One of the causes of this resistance is
prolonged periods of instability: “Individuals in organizations can only stand so much
uncertainty and turbulence”. The overload of duration and amount of change going on may
create dysfunctional effects such as extreme defensive behavior and resistance to any
proposed change (Nadler 1981, Bennebroek Gravenhorst et. al. 1999). Hence, if we were to
include the effects of resistance to change and implementation success we could see that the
case with all the biases involved results in better performance than the case with only the
tendency bias.
Following this discussion, we can be in a position to propose that when more and more
biases are involved in a decision-making activity some of the adverse effects might cancel
each other out, leaving the organization better off.
Another insight we can generate from this modeling work is the similarity between the
concepts of commitment to strategy and inertia. It is possible that organizational inertia can
be explained in terms of commitment. How inertia accumulates and de-accumulates are not
well established in the literature whereas its effect on the change process is very fundamental.
The link between these two literature streams might generate improvements for the
understanding of the inertia concept.
There is a controversy in the commitment to strategy literature regarding the effects of
negative performance feedback on commitment. In order to investigate the behavioral
consequences of the two viewpoints, we have incorporated both into the model and compared
the results. If the commitment declines with negative performance feedback then the
organization can change whereas ignoring negative performance feedback causes the
organization to get locked into a low-fit strategy and even die. Given the drastic difference in
the organizational outcomes, it is desirable to focus further research efforts on identifying
how commitment really declines and why certain organizations choose one de-escalation
strategy over the other. Understanding these causes would generate a basis for identifying the
leverage points for effective and timely de-escalation of commitment. Since commitment is
beneficial in certain circumstances (such as stable or very volatile environments), it should
also be researched when commitment should de-escalate. Thus, we would propose that the
literature should focus its attention on how to manage commitment rather than generating
more and more cases in showing that commitment exists. Understanding how commitment
can be handled effectively could lead to major understanding of how change processes can be
carried out more effectively.
There are several limitations to the model and the analysis. As explained before, the aim of
this study was not to show how all the biases are related to each other but to show that
inclusion of the interaction of biases into the analysis does lead to different insights. Yet, a
further, more detailed modeling would enhance our understanding of the effects of biases in
decision-making. In this respect, our effort should be seen as a starting point.
We believe that we have shown some important information processing biases. However, the
model should be expanded and improved to validate the results of our analysis. For instance,
we have not included the processes that cause the accumulation of commitment. Literature
streams such as escalation of commitment give a more detailed account of the factors that
affect the commitment to a strategy. Including these causes explicitly would increase the
validity of the model and could generate further insights.
The drifting goals structure could also be an interesting extension. Erosions in required
performance (and even expected performance) would lead to further delays in change
processes, and possibly, to further deterioration of the resulting strategic fit.
From an organizational change point of view, factors affecting the success of change
processes, such as implementation success and the appropriateness of the newly chosen
strategy are very important. Such success factors could also be affected by the biases that are
included in this model. For instance, the appropriateness of strategy depends on the
information processing capabilities of the organization. A decline in the quality of
information processing could lead to the implementation of inappropriate strategies that could
further deteriorate the fit of the organization. Extending the bias with such factors and their
relations to the biases would be desirable.
We have assumed that the tendency to relate success to strategy causes the perception of
performance to be higher than the actual when commitment is high. It is also possible that
under low commitment, the organization perceives its performance to be lower than the
actual. As Schwenk (1984) points out: “ Those who believe it [company’s current strategy] is
failing may overweight information on gaps between performance and expectation”. Such a
bias could have both positive and negative effects. It could, under environmental change
conditions, decrease the delay in reaction. It could, on the other hand, also lead to
unnecessary change when the organization is actually performing well.
Even though the model has the above-explained limitations, we believe that given the
interesting results it shows, this is a promising area of research that should be pursued in the
future.
References
Bazerman, M. H. 1998. Judgment in Managerial Decision Making, John Wiley & Sons, Inc.
Bennebroek Gravenhorst, K. M., R. A. Werkman, and J. J. Boonstra. 1999. The Change Capacity of
Organizations: General Assessment and Exploring Nine Configurations. 9" European Congress on
Work and Organizational Psychology: 29-54.
Brown S. L. and K. M. Eisenhardt. 1997. The Art of Continuous Change: Linking Complexity Theory
and Time-paced Evolution in Relentlessly Shifting Organizations. Administrative Science Quarterly
42: 1-34.
Caldwell, D. F. and C. A. O'Reilly. 1982. Responses to Failure: The Effects of Choice and
Responsibility on Impression Management. Academy of Management Journal 25(1): 121-136.
Child, J. 1972. Organizational Structure, Environment, and Performance: The role of strategic choice.
Sociology 6: 2-22.
Choudhury, V and J. L. Sampler. 1997. Information Specificity and Environmental Scanning: An
Economic Perspective. Management Information Systems Quarterly. 21 (1): 25-54 .
Conlon, E.J. and J.M. Parks. 1987. Information Requests in the Context of Escalation. Journal of
Applied Psychology 72: 344-350.
Drazin, R. and A. H. van de Ven. 1985. Alternative Forms of Fit in Contingency Theory.
Administrative Science Quarterly 30: 514-539.
Einhorn, H.J. and R.M. Hogarth. 1978. Confidence in Judgment: Of the Illusion of Validity.
Psychological Review 85: 395-416.
Hambrick, D. C. 1982. Environmental Scanning and Organizational Strategy. Strategic Management
Journal 3: 159-174.
Hannan, M. and J. Freeman. 1984. Structural Inertia and Organizational Change. American
Sociological Review 82: 929-964.
Hogart, R. M. 1987. Judgment and Choice: The Psychology of Decision. Wiley.
Hogarth, R. M. and S. Makridakis. 1981. Forecasting and Planning: An Evaluation. Management
Science 27(2): 115-138.
Kiesler, S. and L. Sproull. 1982. Managerial Response to Changing Environments: Perspectives on
Problem Sensing from Social Cognition. Administrative Science Quarterly 27: 548-570.
Lant, T. K. and A. E. Hurley. 1999. A Contingency Model of Response to Performance Feedback.
Group and Organization Management 24(4): 421-437.
Lant, T. K. and S. J. Mezias. 1992. An Organizational Learning Model of convergence and
reorientation. Organization Science 3: 47-71.
Levitt, B. and J. G. March. 1988. Organizational Learning. Annual Review of Sociology 14: 319-340.
Nadler, D. 1981. Managing Organizational Change: An Integrative Perspective. The Journal of
Applied Behavioral Sciences 17(2): 191-211.
Northcraft, G.B. and G. Wolf. 1984. Dollars, Sense, and Sunk Costs: A Life Cycle Model of Resource
Allocation Decisions. Academy of Management Review 9(2): 225-234.
Pfeffer, J. and G. R. Salancik. 1978. The External Control of Organizations: A Resource Dependence
Perspective. New York: Harper & Row.
O'Reilly III, C. A. 1980. Individuals and Information Overload in Organizations: Is More Necessarily
Better? Academy of Management Journal 23(4): 684-696.
Schwenk, C. R. 1984. Cognitive Simplification Processes in Strategic Decision-making. Strategic
Management Journal 5: 111-128.
Schwenk, C. R. 1986. Information, Cognitive Biases, and Commitment to a Course of Action.
Academy of Management Review 11(2): 298-310.
Smart, C. and I. Vertinsky. 1977. Designs for Crisis Decision Units. Administrative Science Quarterly
22: 640-657.
Starbuck, W. H., A. Greeve, and B.L.T. Hedberg. 1978. Responses to Crises. Journal of Business
Administration 9: 111-137.
Starbuck, W. H. and F. J. Milliken. 1988. Executives Perceptual Filters: What They Notice and How
They Make Sense. In D. Hambrick (Ed.). The Executive Effect: Concept and Methods For Studying
Top Managers: 35-66. Greenwich, CT: JAI Press.
Staw, B. M. 1976. Knee-deep in the big muddy: A study of escalating commitment to a chosen course
of action. Organizational Behavior and Human Performance 16: 27-44.
Staw, B. M. 1981. The Escalation of Commitment to a Course of Action. Academy of Management
Review 6(4): 577-587.
Staw, B. M. and J. Ross. 1987. Behavior in Escalation Situations: Antecedents, Prototypes, and
Solutions. Research in Organizational Behavior 9: 36-78.
Staw, B. M., L. E. Sandelands, and J. E. Dutton. 1981. Threat-Rigidity Effects in Organizational
Behavior: A Multilevel Analysis. Administrative Science Quarterly 26: 501-524.
Sterman, J.D. 2000. Business Dynamics: Systems Thinking and Modeling for a Complex World. Irwin
McGraw-Hill, Boston.
Sutchliffe, K. M. Organizational Environments and Organizational Information Processing. In: (Eds.
Jablin, F. and L. Putnam) Handbook of Organizational Communications.
Tushman, M. and E. Romanelli. 1985. Organizational Evolution: A Metamorphosis Model of
Convergence and Reorientation. In L. L. Cummings and B. M. Staw (Eds.). Research in
Organizational Behavior 7: 171-222. Greenwich, CT: JAI Press.
Ungson, G. R., D. N. Braunstein, et al. 1981. Managerial Information Processing: A Research
Review. Administrative Science Quarterly 26: 116-134.
White, R. E. and R. G. Hamermesh. 1981. Toward a Model of Business Unit Performance: An
Integrative Approach. Academy of Management Review 6(2): 213-223.
Appendix: Equation Listing
Perceived performance = Average fit of strategy * (1-Tendency to relate success to strategy) +
Expected performance * Tendency to relate success to strategy
~ performance units
~ Perceived performance is the performance as the organization feels it. It has two
aspects. It depends on the average fit of the strategy. On the other hand, the performance is biased
towards expected performance depending on how committed the organization is to its current
strategy.
Average fit of strategy = SMOOTH(Fit of Strategy,4)
~ performance units
Commitment decrease=0 * Commitment to strategy * Commitment decrease fraction +
MIN(Commitment to strategy, Commitment to strategy * Change percentage)
~ commitment units/week
~ The amount of decrease in commitment per week. The formulation
MIN(Commitment to strategy,Commitment to strategy * Change percentage)" is used to simulate the
situation where commitment decreases only as a result of change in strategy. This formulation is used
for the analysis of section 4. The second formulation, "Commitment to strategy*Commitment
decrease fraction" is used to simulate the situation where commitment decreases as a result of
performance pressures. This formulation is used for the analysis of section 5.
”
Change in fit = Change completed
~ performance units/week
~ This is the amount of change in fit as a result of implemented changes.
Tendency to relate success to strategy = Table for tendency(Commitment to strategy)
~ Dmnl
~ Signifies the bias in performance measure. It is equivalent to the weight given to the
expected performance. |-tendency is the weight given to the objective performance measure of fit.
Pulse amount = 50
~ performance units/week
Fit decrease due to environmental change = 0*Fit of Strategy/Fit wear off time + pulse amount*
PULSE( 5,1)
~ performance units/week
~ The formulation “Fit of Strategy/Fit wear off time” is used to simulate the decay
environment and the formulation “pulse amount* PULSE( 5 ,1)” is used to simulate the pulse
environment
New change started = (Required performance - Perceived performance-Change in Process) * IF Then
Else ((Perceived pressure to change -confidence in strategy)>0, Perceived pressure to change-
confidence in strategy, 0)
~ performance units/week
~ Amount of new changes started per week.
Effect of perceived pressure to change on information processing capacity = Table for effect on
information processing capacity (Perceived pressure to change)
~ Dmnol
Commitment decrease fraction = Table for commitment decrease fraction(Perceived pressure to
change)
~ l/week
Perceived pressure to change = SMOOTH(Pressure to change, 26)
~ pressure units
Performance correction = (Performance gap-Average performance gap)/Delay performance gap
~ performance units/week
Exerted pressure to change = Table for pressure by performance gap(Relative performance gap)
~ pressure units
Change completed = Change in Process/Implementation time
~ performance units/week
~ Amount of change that is completed (implemented) per week.
Commitment increase = (1-Commitment to strategy)/"Commitment build-up time"
~ commitment units/week
~ The increase in commitment per week. Commitment increases continuously.
Confidence in strategy = Normal confidence * Intensity of search for confirming evidence
~ pressure units
~ The total amount of confidence in strategy. It can be seen as the pressure to not
change since it works as a threshold against change.
Table for commitment decrease fraction( [(0,0)-
(1,1)],(0,0),(0.155294,0.0284698),(0.254118,0.0782918),(0.369412,0.209964),(0.451765,0.359431),(
0.5,0.5),(0.550588,0.679715),(0.6141 18,0.822064),(0.672941,0.903915),(0.748235,0.950178),(0.804
706,0.97153),(0.85647 1 ,0.985765),(0.910588,0.989324),(1,1))
Normal confidence = 0.5
~ pressure units
~ Normal amount of confidence an org feels towards its strategy.
Table information overload(
[(0,0)-
(2,1)],(0,0),(0.225882,0.0284698),(0.348235,0.0676157),(0.494118,0.117438),(0.64,0.181495),(0.771
765,0.24911),(0.884706,0.35943 1),(1,0.5),(1.11529,0.615658),(1.20941,0.701068),(1.44471,0.85053
4),(1.56235,0.914591),(1.71765,0.975089),(1.83059,0.992883),(2,1))
Initial fit = 100
~ performance units
Initial commitment = 0.5
~ commitment units
Commitment to strategy = INTEG (Commitment increase-Commitment decrease,initial commitment)
~ commitment units
~ Signifies the total amount of commitment the organization has to its strategy.
Commitment ranges from 0 to 1. Commitment of | means that the organization is totally supporting
its current strategy.
Fit of Strategy = INTEG (Change in fit-Fit decrease due to environmental change, initial fit)
~ performance units
~ Signifies the fit of organization's strategy to the strategy demanded by the
environment. Fit of strategy is a sign of organization's performance.
Acceptable performance gap = 0
~ performance units
Relative performance gap = (Average performance gap-Acceptable performance gap)/100
~ performance units
~ Signifies the difference of the existing performance gap from the acceptable
performance gap. The difference is divided by 100 for scaling purposes.
Table for tendency(
[(0,0)-
(1,1)],(0,0),(0.141176,0.02491 1),(0.24,0.103203),(0.341176,0.217082),(0.416471,0.352313),(0.5,0.5),
(0.618824,0.626335),(0.724706,0.697509),(0.863529,0.736655),(1,0.75))
Expected performance = 100
~ performance units
~ The total expexted performance of the strategy. The expected performance is assumed
to be equal to the maximum fit a strategy can have.
Intensity of search for confirming evidence = table commitment(Commitment to strategy)+table
information overload(Information overload)
~ Dmnol
Table commitment(
[(0,0)-
(1,1)],(0,0),(0.117647,0.0177936),(0.197647,0.0462633),(0.275294,0.088968),(0.331765,0.163701),(
0.390588,0.263345),(0.432941,0.348754),(0.5,0.5),(0.5341 18,0.622776),(0.571765,0.72242),(0.6211
76,0.80427),(0.6941 18,0.875445),(0.748235,0.914591),(0.804706,0.953737),(0.870588,0.985765),(1,
1)
Available information = 2
~ Amount of available information
Normal information processing capacity = 0.5
~ Amount of information that can be handled under normal pressure.
Information handling capability = Normal information processing capacity * Effect of perceived
pressure to change on information processing capacity
~ Amount of information that can be handled
Table for effect on information processing capacity(
[(0,0)-(1,2)],(0,0.04),(0.0447059, 1.08185),(0.0611765,1.40214),(0.1,1.75),(0.145882,
1.90036),(0.2,2),(0.3,2),(0.4,1.94306),(0.447059, 1.91459),(0.5,1.83),(0.6,1.57),(0.661176,1.37367),(0
-7,1.18),(0.727059,0.982206),(0.757647,0.7829 18),(0.778824,0.697509),(0.8,0.58363),(0.851765,0.3
91459),(0.894118,0.263345),(0.931765,0.170819),(1,0.12))
Pressure change = (Exerted pressure to change - Pressure to change)/Pressure build up time
~ pressure units/week
Performance gap = Required performance-Perceived performance
~ performance units
Information overload = Available information-Information handling capability
Average performance gap = INTEG (Performance correction, 1e-006)
~ performance units
Change in Process = INTEG (New change started-Change completed, 0)
~ performance units
~ Change in Process signifies the amount of ongoing changes in the organization. It is
measured in terms of the amount of fit gap to be closed as a result of the on-going changes.
Change percentage = ABS(ZIDZ(Change in fit, Fit of Strategy))
~ l/week
Delay performance gap = 4
~ week
Fit wear off time = 200
~ week
"Commitment build-up time" = 60
~ week
Implementation time = 26
~ week
Pressure build up time = 8
~ week
Pressure to change = INTEG (Pressure change, 0)
~ pressure units
~ Amount of pressure the organization feels given the performance gap. This pressure
dictates the percentage of the gap that should be closed in the fit of the strategy.
Required performance = 100
~ performance units
Table for pressure by performance gap(
[(0,0)-(1,1)],(0,0),(0.1247,0.04626),(0.2282,0.1139),(0.3176,0.2598),(0.3929,0.4235)
(0.4471,0.5979),(0.52,0.79),(0.5906,0.9324),(0.7,1),(0.8,1),(0.9,1),(1,1))
Back to the Top
