Center for Information Systems Research
Massachusetts
Institute of
Technology
Alfred P. Sloan School of fvlanagennent
50 Memorial Drive
Cambridge, Massachusetts, 02139
617 253-1000
TOWARDS A BEHAVIORALLY-GROUNDED
THEORY OF INFORMATION VALUE
Michael
E.
Treacy
February 1981
(Revised August 1981)
CISR No. 7A
Sloan WP No. 1191-81
(9=5
Center for Information Systems Research
Sloan School of Management
Massachusetts Institute of Technology
ABSTRACT
Economic models of information value have
practice
theory and
of
assumptions.
major areas
the
in
theory
little
that
importantly,
more
stem
from
modification:
multiple
on
the
is
and
due
to
invalid
reviews
five
the decision process, human judgement
under uncertainty, the choice of actions, multiple
resolution, and
it
descriptively
This paper examines those assumptions
for
impact
This is due in part to difficulties in
MIS.
operationalizing these models, but
problems
had
decisions over time.
descriptive assumptions in the economic
theory
information
Incorporation of valid
will
move
toward a behaviorally-grounded theory of information value.
^^^\C^\
signal
the
field
INTRODUCTION
The theory of information economics has had little
discernible
impact
on the theory and practice of management information systems design and
large
evaluation, despite
provided
has
economic theory
communication
and
information
of
obvious
overlapping
models
of
[3'+]
the
and
[50]
of
information system. [35] It has provided models of
from among
interests.
The
transmission
and
value
the
the
of
optimal
an
choice
available information system components [35] [40] and of the
comparative informativeness of information
systems. [7]
Yet,
none
of
these models has had much influence on MIS.
Part of the explanation of this unhappy situation is that the theory is
difficult to operationalize, because theoretical models of
value require
large
numbers
impossible to estimate.
and Onsi[6],
of
parameters that are sometimes
input
Direct applications of the theory
Feltham[18],
Mock['41],
unsatisfactory in their results.
information
by
Bedford
and others have been awkward and
This has
led
some
to
abandon
unidimensional value model in favor of multiattribute approaches.
and
Epstein[30]
and
Ahituv[1]
have
each
presented
multiattribute schemes which maintain little of the
the
King
operational
substance
of
the
economic theory.
The major root of the problem with economic information
value
theory,
though, is not its inoperability, but that it is based upon invalid and
unrealistic assumptions
information.
about decision making and how managers utilize
Consider, as an example Blackwell's TheoremCY],
probably
result of information economics.
known
the best
The theorem directly
implies that a more disaggregated information system is always at least
as desirable as a less detailed
such
assumptions,
severe
as
Within
system. [23]
unlimited
processing, this result is valid, but
costless
and
in
more
a
confines
the
information
managerial
typical
validity vanishes and the entire point is lost.
setting the
information
of
limited impact
economics
our
on
of
Thus, the
stems
field
from
problems in the theory, rather than in the practice.
In this paper,
we shall expand this thesis and set directions
for
information value theory that is founded upon valid and
development of
realistic assumptions about managerial decision making behavior.
simply
not seek
the
operational
an
ultimate objective
move
to
is
information
of
model
We do
Our
value.
the economic theory towards what Keen
calls "a behaviorally-grounded conception of optimality that meshes the
of
pluralistic models
start with
a
review
decision
modifications of
decision
making
developments
in
the
of
descriptive,
with
science
process. "[29, p.
the
economic
31] We
of
theory
and a critical examination of underlying assumptions
information value
about the
optimization
of
analytic perspective
the
process
and
the
decision
Five
maker.
major
economic theory, suggested by descriptive models
of managerial behavior found in other fields of study, are discussed in
turn.
section
Finally, this
that
entire
illustrates
information value.
approach
the
utility
reviewed
is
of
a
in
a
behavioral
concluding
theory
of
ECONOMICS MODELS OF INFORMATION VALUE
Two distinct schools, one in economics and another in statistics,
have
with the value of information for more than twenty-five
been concerned
years. [24] Information economics and statistical decision analysis have
produced a series of similar
models
that
information
value
in
the
context of several restrictive assumptions about the behavior, ability,
and knowledge of actors using information.
The models
information
of
complexity of
value
vary
two
dimensions:
the information source and the number of decisions.
information source can be
a
single signal,
or multiple information systems and the
multiple.
along
Figure
1
a
decisions
can
be
single
or
illustrates this diversity and indicates selected
SINGLE
SINGLE
INFORMATION
SIGNAL
SOURCE
^^^^^— ^^^•"^•""^"
MULTIPLE
DECISIONS
The
single information system,
references for each type of model.
SINGLE
DECISION
the
MULTIPLE
INFORMATION
SOURCES
Each decision is framed as
action from
predefined
a
problem which involves the
a
set
alternatives.
of
choice
of
The utility of any
action depends on which state of nature occurs and the utility of
action-state pair
known.
is
an
each
The set of states is predefined, but the
decision maker, who wishes to maximize utility, has only
knowledge as to which will occur.
probabilistic
Information in the form of single or
multiple signals from single or multiple information sources is used to
probabilistic
refine the
requires that
decision
the
conditional probability
state of nature.
maker
have
knowledge
detailed
utility
information.
states occur.
Thus, it deals in expected quantities.
L.J. Savage, the originator of the fundamental axioms
was,
'ratioiial' person
analysis
decision
and
idealized
highly
"a
increase
theory
which
upon
of
the
of
behavior
flaws
acting
in
a
more
a
a
valid
descriptive
typical manager.
complex
model
the
of
of
value
Such a model must be founded upon
valid description of the managerial use
upon the
more
realistic,
We are attempting to prescribe the variables that must be
considered to produce
information to
a
with respect to decisions. "[47, p. 7] Our concern is
imperfections,
environment.
the
models are based, wrote
with the value of information to a typical manager, with all his
and
ante
ex^
any signals are received, actions are chosen, or
before
that his
an
is
It
measure, made
information economics
the
This gain in expected utility is
of action.
the definition of the value of the
of
each different signal given any
obtaining
of
Bayesian revision, which
through
The refined knowledge of states leads to an
expected
in the
knowledge
of
information,
sterile, prescriptive assumptions that define
rather
a
than
'economic man'.
AREAS OF MODIFICATION
These economics models of information value poorly describe
of information
in
inadequacies
discussion into five sections.
models, as suggested by
indicated.
We
roles
managerial decision making and hence poorly reflect
the obtainable value of information
The descriptive
the
a
conclude
of
managerial
typical
in
settings.
these models have b^een organized for
In each,
possible modifications of
the
reading of other related fields of study, are
with
some
remarks
on
the
difficulty
of
implementing such modifications and the efficacy of this approach.
1.
The Decision Process
According to economics
models,
effect on the decision process.
if decision
making
is
information
derives
from
value
its
This orientation is difficult to fault
interpreted broadly, for almost all managerial
activities which use information can be classified as some phase of the
intelligence-design-choice-review decision process. [55, p.
Simon
54]
writes:
finding
Decision making comprises four principle phases:
occasions for making a decision, finding possible courses of
action, choosing among courses of action, and evaluating past
These four activities
account for most
choices.
of what executives do. [56, p.
40]
....
Mlntzberg's study of the work of five chief executives reinforces
finding. [38, p. 250]
All
but
this
time spent in ceremonial activities (12
percent) and in giving information (8 percent) is attributable
or more phases of decision making.
to
one
Witte[66j formally tested for the existence of different phases in
decision process
equipment.
using
He divided
each
gathering, alternatives
choice.
evidence
The
sample of 233 decisions to acquire computer
a
characterized
periods and
the
decision
process
into
time
equal
ten
each activity in each period as information
development,
supported
alternatives
evaluation,
or
hypothesis that multiple phases
the
exist within the decision process.
The difficulty with the economics models is that they concentrate
upon
only one phase of decision making, the choice among alternative courses
of action.
They
assume that an occasion for decision making has been
found and that all possible courses
every conceivable
of
and
have
managers
for
But, by the
used
already
a
of information and expended the majority of their effort on
the problem. [56, p.
decisions are
Edwards
As
40]
profoundly
affected
and
Roxburgh[15]
identify
problems,
have
noted,
by information at the intelligence
and design phases of the decision making process.
to
consequences
of events have been determined.
course
time these assumptions are satisfied,
great deal
action
structure
alternatives,
Without
information
and
estimate
consequences, no choice is ever made.
Economic models of
information
value
must
be
expanded
to
include
consideration of these earlier phases, intelligence and design, if they
are to
accurately
reflect
the
benefits
of
information.
The final
phase, review, need not be explicitly modelled since it is usually part
of the intelligence phase of other decisions, and could be captured
such in a multiple decision model.
as
The phase theory of decision making implies not only that decisions are
comprised of different
follow
a
activities,
pattern,
set
implementation of
hypothesis
support the
progression
a
chosen
the
but
also
from
these
initial
Witte's
actions.
that
that
activities
recognition
evidence
does
to
not
the phases followed a clear progression.
Even when each decision was divided into subdecisions, no
support
for
the hypothesis was found.
Hintzberg, Raisinghani, and Theoret[39] also found evidence of
through
during
phases
decision
the
process,
twenty-five strategic decision processes
They suggest
cycling
that
used
is
different
in
as
a
decisions
comprehension cycles". [ 39
interrupts, created
by
t
seem
internal
of
organizations.
most
complex
and
involve the greatest incidence of
to
Evidence
265]
p.
study
means of comprehending and
clarifying complex decision processes and that "the
novel strategic
their
in
cycling
and
was
also
pressures
external
that
found
and by the
appearance of new options, caused cycling.
The authors build their findings into
intelligence-design-choice
comprised of
two
model
an
and
decision
routines:
elaboration
pwsit
that
recognition
of
the
simple
intelligence
and
is
diagnosis.
Diagnosis is an optional routine used to clarify and define the issues.
Decision recognition
occurs
when
there
are sufficient signals about
either a crisis, a problem, or an opportunity.
problems by
of
stimulus was first suggested by Carter [10] in his study of
six strategic decisions within one
Cyert and
This categorization
March[12]
suggested
company.
The
earlier
theory
of
that decision recognition was always a
response to problems rather than to perceived opportunities.
Pounds[U3] has presented
identification, one
a
type
theoretical structure for analysing
of
decision
recognition,
as
comparing information about actual conditions against
of
a
chosen
'model' of how things ought to be.
a
process of
predictions
the
Tne models managers use
or explicit derivations from historical and planning data
are implicit
or models imposed by others or derived from outside
organization.
the
The design phase of decision making is not well understood.
March[12] posit
search for
problem
design
that
acceptable
largely
is
alternative
Cyert
and
matter of problem-directed
a
How
actions.
this
search
is
accomplished, though, is somewhat less than clear.
Mintzberg, Raisinghani, and Theoret[39] suggest chat design activity is
very different depending upon
whether
ready-made or
solution.
custom-made
a
decision
the
They
appropriate for ready-made solutions, but that
are
necessary
solutions.
for
the
description
of
the
Reitman[M6] and Alexander [2] have
note
more
design
further
sought
maker
that
search
is
models
elaborate
of
a
custom-made
detail
on
the
various forms of design activity.
In summary,
there
intelligence and
decision process.
intelligence and
is
design
The
design
general
agreement
activities
addition
processes
of
exist
some
in
the
literature
that
as important phases of the
consideration
of
the
to the model should provide a more
accurate evaluation of the managerial uses of information.
The
exact
each
nature of
phase
and
their
order
decision implementation appears to vary
Therefore it
may
from decision recognition to
among
of
classes
decisions.
necessary to build specialized information value
be
models, using the general information economics approach, for different
types of decisions or different
using
PoundsCUB]
model
of
of
roles
process
the
management.
problem
of
description of how managers utilize information
one could
build
finding
problem
for
The general approach to this model would be the same as in
economics, but
underlying
the
example,
as
a
finding,
of the value of information for monitoring.
model
a
For
assumptions
information
about managerial behavior
would be more accurate and would result in more valid, more useful, and
more usable theory.
2.
Human Judgement Under Uncertainty
There are two competing paradigms of how
judgement and
thought.
choice,
the
Bayesian
and
information
utilized
is
regression
the
in
schools of
The essential difference between the two is in the manner
of
assessment of the relationship between information and the states about
which one
drawing
is
inference.
The
Bayesians
propose the use of
conditional probabilities and Bayes' theorem to assess
the
impact
of
information
probability
of
obtaining.
proposed
upon
judgements
regression
The
by
prior
Brunswik
school,
[8][9],
uses
of
the
states'
formalized
in
correlations
the
of
lens
model
states
with
information signals to weight the importance of each information signal
in the final judgement.
human judgement,
the
After several hundred
psychology
studies
of
rivalry between the two schools remains intense.
10
Despite obvious conceptual overlap, attempts at unifying the two
views
have met with limited success. [57]
Mock and Vasarhelyi[42] have attempted to synthesize the lens model
with
processing
human information
economic model of information
the
accomplished
Their conceptual schema suggests that this may be
value.
of
substituting the lens model for the Bayesian model of signal
simply by
Hilton[22]
utilization.
integrating the
views
two
taken
has
different
information
an
in
a
processing
He has
into
a
extending the dimensionality of the utility function
by
made
be
course,
Of
to cover each feature.
dimensions can
model.
value
absorbed some of the features of human information
Bayesian view
towards
approach
fit
to
a
any
utility
with
function
enough
Thus,
the complex
information
utilization
situation.
resultant formulation is of dubious value.
Economics has adhered to the Bayesian view of
ever
joined
first
3avage[47]
since
formal,
subjective probability into a
making.
This
why
is
the
axiomatic
p(s),
decision
of
theory
and
economics information value models require
the
that the decision maker have knowledge of the
states occuring,
utility
of
concepts
and
of
probabilities
prior
of
the conditional probabilities of each
signal, p(y|s), for the derivation of p(sly), the probabilities of each
state occuring revised upon receipt of
signal
y.
formulation of the decision maker's problem, for it is
to produce
y,
directly
than to estimate
revision formula.
is
It
a
a
curious
simpler matter
subjective estimates of p(s|y) in the presence of
both
p(s)
and
p(y|s)
and
apply
the
Bayesian
11
Consider, for
conditions.
moment,
a
weather
forecast
The
is
forecasts
and
information;
it
tomorrow's
weather
corresponds
to
Tomorrow's weather condition (rain or sun) is the uncertain state.
think of your favorite weather forecaster and estimate the
forecast
that he
will
p{y|s).
An important difficulty immediately arises.
backwards to
states.
rain
normal
the
given
fashion
thinking
of
more
a
natural
information.
revision as
measures
it
This
a
probability
problem
The
about information and
forecast
a
signal,
then
natural
the
illustrates
example
an
rain,
of
notion
about
the
reliability
of
inadequacy
the
Bayesian
of
descriptive theory.
simplifies
maker
the
complications must be considered.
psychology literature that
systematic bias
the
in
There is
documents
estimation
probabilities
and
have
formulation,
model
and
of
Kahneman have identified three important
estimate
of
inference
Reformulation of the model to indicate direct estimation of
the decision
is
assessment, because it is chronologically
ordered (first an information
and
Now
will be sunny tomorrow,
it
The probability of sun tomorrow, given
p(s!y), is
state)
that
y.
a
p(s|y)
but
by
further
large and growing body of
theorizes
evidence
on
probabilities.
heuristics
demonstrated
how
Tversky
which
by
these
of
and
people
lead
to
systematic biasing of estimates. [27][62] [63] The 'prospect theory' they
have developed sheds considerable light on how outcomes are
framed
as
gains and losses in evaluating utilities and on the transient nature of
these values. [28] [64]
A
model
of
information value needs to include
consideration of these systematic biases, for they induce
subutilization of
information,
and
decrease
a
systematic
the obtainable value of
12
information.
3.
The Choice of Actions
Economics information value models
consequences
explore the
of
value
actions are chosen on
a
that
the
decision
maker
of every action, from their potential action
set, in every state of nature.
the expected
require
He chooses the action
outcomes.
which
maximizes
There is considerable evidence that
much simpler basis.
Simon was one of the first to question the maximum expected value model
He developed
of choice.
submitted it
as
well
known
idea
of
satisficing,
and
better description of individual behavior and as a
a
normative model
the
rational
of
behavior
conditions
under
costly
of
information gathering and processing. [51 ]t52][53][54] He suggested that
an action
determine
choice
a
rule
more descriptive of human behavior would be to
minimum aspiration level,
sequentially search
and
test
L,
for
a
decision
outcome
potential actions until an action a
and
'
is
found such that:
MIN u(s,a')
>
L
s
In this
formulation, u(s,a') need not be
only needs
aspiration.
to
know
whether
uCs.a')
accurately
determined;
is greater than L,
L and u(s,a') could be multidimensional.
one
the level of
Then, the action
choice rule need not be modified, but the chosen action a' must satisfy
the rule along every dimension.
This obviates the need for
among dimensions of the objective.
a
tradeoff
13
Cyert and March[12] extended this idea to the theory of
considerable work
has
continued
theory. [31] Stigler[60]
Many
activity.
of
value and serve as
has
these
this
in
explored
the
firm
and
bounded rational
economics
ideas could simplify
of
search
the
model of information
a
better description of decision making behavior.
a
Soelberg[58][59] studied the behavior of
making job decisions.
one acceptable
of
area
the
fifty-two
graduate
students
He found evidence that individuals had more than
alternative
choice
before
contradiction to strict satisficing
ending
behavior.
their
Soelberg
search,
in
developed
a
theory of decision making that combines the notions of maximizing along
the most
important
one
or
two dimensions of outcome and satisficing
along all others, to explain his findings.
The conflict between Simon's and Soelberg
resolved
can be
using
differentiation between
Mintzberg,
ready-made
's
theories of choice behavior
Raisinghani,
and
custom-made
Theoret's
solutions.
They
write, "The hypothesis with the strongest support in our study is
that
the
organization
solution.
.
solutions
.
In
designs
contrast,
typically
only
and
one
organizations
selected
them
alternatives". [39, p.
256] Soelberg
seeking and
from
choosing
fully-developed
among
's
from
that
chose
among
sample was
ready-made
a
of
custom-made
ready-made
number
decision
of
makers
solutions (job offers),
whereas many of Simon's conclusions appear to have germinated from
observations of
problems
involving custom-made solutions, such as the
widely referenced description of
the early
1950's.[11]
his
a
computer aquisition decision made in
14
Different simplified choice rules could also be modelled.
For example,
one could model the practice of developing plans based upon assumptions
about most likely future scenarios.
the most
This is equivalent to
identifying
likely state of nature and choosing an action to maximize the
value of the outcome if that state obtains.
decision
The
rule
would
be, choose a', such that:
uCs'.a')
=
MAX u(s'.a)
a
where
p(s'!y') =MAXp(s|y')
s
Clearly, the appropriate model of action choice varies among classes of
decisions.
Research has provided some understanding of when and
different choice
strategies
are
used,
but
not enough to be able to
construct one integrated model of action choice.
information value
should
be
specialized
to
where
Therefore, models
particular
classes
decisions or types of managerial roles, so that the appropriate
of
of
action
choice model may be incorporated.
U.
Multiple Signal Resolution
There is little evidence that individuals resolve multiple and possibly
conflicting signals through
Bayesian
psychologists
a
have
complex Bayesian revision process.
developed
theories
about
Even
individuals'
misaggregation of multiple signals to explain the apparent conservative
revision of prior probability estimates. [5] [ 16][20][22][63][65]
15
The regression paradigm offers only a slightly
multiple signal
resolution.
better
description
of
and Dudycha and Naylor[14],
Suramers[6lj
studied the utilization of orthoganol information signals and concluded
that the lens model provided an accurate description of how individuals
utilize uncorrelated
information
Brunswik[9] suggested
signals
well
how
forming
judgements.
that intercorrelations among information signals
are the rule rather than the exception.
to evaluate
in
individuals
His lens model has
able
are
intercorrelations, by comparing an individual's
with regression weights.
been
used
to make adjustments for
weighting
of
signals
The evidence indicates that these adjustments
generally are not made very well. [3JE 17][21 ][25][49]
As with modelling
the
design
of
phase
decision
the
process,
the
direction to take in modelling multiple signal resolution is not clear.
Nevertheless, it
should
validity of
complex
the
be
to improve upon the descriptive
possible
Bayesian
revision
process
adopted
by
information economics.
5.
Multiple Decisions Over Time
How do managers deal with information over time?
assume that
future
all
beginning of
a
finite
decision
time
horizon.
information source
is
stream of
improvements
outcome
information signals
valued
optimally.
problems
as
[
the
gained
models
have been designed at the
18][ 19]
present
by
The economics
a
In
this
context,
an
value of the expected
decision
maker
using
The solution to this problem can only
be derived using dynamic programming, for one must consider the
impact
16
of each
signal
in
future
all
decisions
It is not the reuse of the
decision.
well as in the present
as
information source,
of
but
the
particular information signals that makes the structure of the model so
absurdly counter to intuitive notions of managerial behavior.
The economists
appear
have
to
been
trying
use
the
of
This might be accomplished
in decision making.
historical information
model
to
much more simply if we do not distinguish historical
information.
In
this way, the new information signals may embody historical information
and the overwrought complexity of the problem disappears.
The other major modification
of
the
decision
multiple
problem
Decision problems cannot
already been suggested in an earlier section.
be defined and enumerated at the beginning of any period of time.
must be
has
They
discovered, selected, or assigned with little forewarning.
We
best
be
have suggested that this
problem
issue
identification
can
described by adding an intelligence phase to the model.
CONCLUSIONS
We have examined the standard economic models of information value
suggested five
toward a
major
areas
of
behaviorally-grounded
suggestions are
not
processing,
without
information valuation.
theory
These
radical.
underlying assumptions about
revision
decision
abandoning
the
of
that
would move the models
information
modifications
making
general
and
and
value.
would
human
economic
Our
alter the
information
framework
of
17
The review of work in
the
of
theories
descriptive
major
five
decision
processing are not well understood.
revision
making
building
but there
a
be
that
information
Competing and conflicting theories
own
relevant
domain
of
This may lead some to be skeptical of success in
revised model and to retreat to
can
reveals
human
and
abound, each with its own proponents and its
managerial behavior.
areas
standard
economic
models,
no safe refuge in theory which is built upon a weak
foundation of invalid assumptions.
Rather than retreat, what is necessary is that
constructed for
particular
classes
of
information
theory
decision making or managerial
action.
Then, one need use only descriptive theory
domain.
The
relevant
to
that
resultant model will not achieve universal applicability,
but it may provide the theoretical underpinnings to the solution of
MIS
be
problem.
Specialization
operational difficulties
information value
have
should
encountered
been
applied.
also
whenever
alleviate
economic
of
the
models
of
some
After all, our implementation
studies have highlighted the advantages of tailor-made solutions.
time to tailor-make some theory.
an
It's
18
REFERENCES
Ahituv, N., "A Systematic Approach Toward Assessing the Value of an
1.
Information System," MIS Quarterly, V.
4 (1980), p.
61-75.
2.
Alexander, E.R., "The Design of Alternatives in Organizational
2*4
A
Pilot Study," Administrative Science Quarterly V.
Contexts:
382-404.
(1979), p.
,
Armelius, K.
and
Armelius,
B.,
"The Effect of Cue-Criterion
3.
Correlations,
Cue
Intercorrelations and the Sign of
the
Cue
Intercorrelations on Performance in
Suppressor
Tasks,"
Variable
Organizational
Behavior and Human Performance
V.
17
(1976),
p. 241-250.
,
4.
Arrow, K.J., "The Value of and Demand for
Information," in C.B.
Radner (ed.). Decision and Organization North-Holland
McGuire and R.
Publishing Co., 1972, p.
131-139.
,
Beach, L.R., "Probability Magnitudes and Conservative Revision
5.
Subjective Probabilities," Journal of Experimental Psychology V.
,
(1968), p.
of
77
57-63.
6.
Bedford, N.M.
and Onsi, M., "Measuring the Value of Information
An Information
Theory Approach," Management Services Jan-Feb 1966, p.
15-22.
,
"Equivalent Comparisons of Experiments,"
Blackwell, D.
7.
24 (1953), p.
265-272.
Mathematical Statistics. V.
,
Brunswik, E.
The Conceptual Framework of Psychology
8.
Chicago Press, 1952.
,
,
Design
,
Carter, E.E., "The Behavioral Theory of the Firm
Corporate Decisions," Administrative Science Quarterly
413-428.
p.
and
,
of
University of
Representative
Brunswik,
E.
Perception and the
9.
Experiments, University of California Press, 1956.
10.
Annals
V.
of
Top-Level
16
(1971).
19
Cyert, R.M., Simon,
11.
H.A.,
and Trow,
Business Decision," Journal of Business
237-248.
.
D.B.,
V.
Cyert, R.M.
12.
and March, J.G., A Behavioral
Prentice-Hall, Inc., 1963.
13.
Demski, J.S.
,
"Observation
29,
4
N.
Theory
Information Analysis, Addison-Wesley,
of
on a
(1956), p.
the
Firm
,
1972.
14.
Dudycha, L.W. and Naylor, J.C,
"Characteristics of the Human
Inference
Process in
Complex
Choice
Behavior
Situations,"
Organizational Behavior and Human Performance, V.
(1966),
p. 110-128.
1
Edwards, C.
and
Roxburgh,
K. ,
"Analysis and Implications
Management Uses of Information," Operational Research Quarterly
243-249.
2 (1977), p.
28. N.
of
15.
,
V.
Edwards, W.
16.
"Conservatism in Human Information Processing," in B.
Kleinmuntz (ed.). Formal Representation of Human Judgement John Wiley
,
,.
and Sons, Inc.,
1968.
Einhorn, H.J., "Use of Nonlinear,
17.
Noncompensatory Models as a
Function of Task and Amount of Information," Organizational Behavior
and Human Performance, V.
6 (1971), p.
1-27.
Feltham, G.A., "The Value of Information," Accounting
684-696.
43 (1968), p.
18.
Feltham,
G.A.,
19.
"Information Evaluation," American
Association: Studies in Accounting Research //5, 1972.
Review
,
V.
Accounting
Gettys, C.F.
and Manley, C.W., "The Probabiliy of an Event and
Estimates
of Posterior Probability Based upon
its
Occurence,"
Psychonomic Science, V.
11 (1968), p.
47-48.
20.
Hayes, J.R., "Human Data Processing Limits in Decision Making," in
Bennett
(ed.),
Information System
Science
and
Engineering:
Proceedings of the First Congress on the Information Systems Science
McGraw-Hill Book Co., 1964.
21.
E.
,
20
Hilton, R.W., "Integrating Normative and Descriptive
Information Processing," Journal of Accounting Research
477-505.
(1980), p.
22.
,
Theories
V.
18,
N.
of
2
Information
Value:
Determinants
of
"The
R.W.,
Hilton,
23.
27, N,
Synthesizing Some General Results," Management Science V,
1
,
(1981), p.
57-64.
24.
Hirshleifer, J., "Where
American Economic Review, V.
are
63,
we
N.
in
2
a
theory
(1973)t p.
of information?"
31-39.
and Blanchard, W.A., "A Study of the Effects of
Hoffman, P.J.
Varying Amounts of Predictor Information on Judgement," Oregon Research
Institute Research Bulletin, 1961.
25.
of
Assessment
R.M.,
"Cognitive Processes and the
Hogarth,
26.
American
the
Probability Distributions," Journal of
Subjective
271-289.
350 (1975), p.
Statistical Association, V.
70, N.
and Tversky, A., "On the Psychology of
Kahneman, D.
237-251.
4 (1973), p.
Psychological Review. V.
80, N.
27.
Kahneman,
Decision Under
28.
Prediction,"
An Analysis of
and Tversky, A., "Prospect Theory:
263-291.
2 (1979), p.
Risk," Econometrica, V.
47, N.
D.
Keen, P.G.W,, "The Evolving Concept of Optimality," Studies in the
29.
31-57.
6 (1977), p.
Management Sciences V.
,
B.J.,
W.R.
and Epstein.
King,
30.
Information," Management Datamatics V.
,
of
Value
the
171-180.
(1976). p.
"Assessing
5,
N.
4
Engineering
March, J.G., "Bounded Rationality, Ambiguity, and the
587-608.
V.
18
of
Economics,
(1978),
p.
of Choice," Bell Journal
31.
32.
Inc.,
March, J.G.
and Simon, H.A., Organizations
,
John Wiley and
Sons,
1958.
C.P.
"Problems in Information Economics." in
Marschak,
J.,
New Directions in Basic
Management Controls:
Bonini, et al
(eds.).
38-74.
Re3earch7~McGraw-Hill Book Co., 1964, p.
33.
21
Marschak, J., "Economics of Inquiring,
Communicating,
American Economic Review, V.
2 (1968), p.
1-18.
58, N.
Deciding,"
3^.
Marschak, J., "Economics of Information Systems," Journal of the
35.
American Statistical Association, V.
66, N.
192-219,
333 (1971), p.
J.
Marschak,
and Miyasawa,
K.
"Economic Comparability
36.
of
Information Systems," International Economic Review V.
9 (1968), p.
,
,
137-174.
McGuire, C.B.
and Radner, R.
(eds.). Decision and
37.
North-Holland Publishing Co., 1972.
38.
Mintzberg, H.
,
Managerial
The Nature of
Work
,
Organization
Harper
and
,
Row,
1973.
Mintzberg, H.
Raisinghani, D, and Theoret, A., "The Structure of
'Unstructured' Decision Processes," Administrative Science Quarterly
V.
246-275.
21 (1976), p.
39.
,
,
,
10.
Mock,
Journal of
41.
Mock,
T.J.,
"Comparative Values of Information
Structures,"
Accounting Research: Selected Studies, 1969, p.
124-159.
"Concepts
T.J.,
Accounting Review,
V.
Information Value
765-778.
of
and
Accounting,"
46 (1971), p.
Mock, T.J.
Information
and Vasarhelyi, M.A., "A Synthesis of the
Economics and Lens Models," Journal of Accounting Research V.
16, N.
414-423.
2 (1978), p.
42.
,
43.
Pounds,
W.F.
Management Review,
"The
,
V.
11,
Process
N.
1
of
Problem
(1969), p.
Finding,"
44.
Raiffa, H.
Decision Analysis:
Introductory Lectures
Under Uncertainty, Addison-Wesley Publishing Co., 1970.
,
Raiffa,
H.
45.
and Schlaifer,
Theory, M.I.T.
Press, 1961.
R.
,
Industrial
1-19.
Applied
Statistical
on
Choice
Decision
22
Reitman, W.R., "Heurlstlo Decision Procedures,
Open Constraints,
and the Structure of Ill-Defined Problems," in M.W.
Shelley and G.L.
Bryan (ed.). Human Judgement and Optimality John Wiley and Sons, Inc.,
46.
,
1964.
47.
Inc.,
Savage, L.J., The Foundations of Statistics
John Wiley and
,
Sons,
1954.
48.
Schlaifer, R., Probability and Statistics for Business
McGraw-Hill Book Co., 1959.
Decisions
,
Schmitt, N.
49.
and Dudycha, A., "A Reevaluation of the Effect of Cue
Redundancy in
Multiple-Cue
Probability
Learning,"
Journal
of
104 (1975), p.
Experimental Psychology. V.
307-315.
Theory
C.E.
The Mathematical
Shannon,
W.,
50.
and Weaver,
Communications. University of Illinois Press, 1949.
Simon. H.A., "A Behavioral Model of Rational
51.
Journal of Economics, V.
99-118.
69 (1955), p.
of
Quarterly
Choice,"
H.A..
Structure
of
Simon,
"Rational Choice and
the
52.
Environment," Psychological Review V.
129-138.
63 (1956). p.
the
.
53.
Simon, H.A., Models of Man, John Wiley and Sons, Inc.,
1957.
H.A.,
54.
Simon,
"Theories of Decision Making in Economics and
V.
Behavioral Science," American Economic Review
49
(1959),
p.
255-283.
,
55.
Simon, H.A., The Shape of Automation. Harper and Row,
Simon.
H.A..
Prentice-Hall, Inc.,
56.
New
Science
(revised).
1977
The
of
1965.
Decision ,
Management
S.
"Comparison of Bayesian and
Slovic, P.
and Lichtenstein,
Regression Approaches to the Study of Information Processing in
V.
6
Judgement," Organizational Behavior and Human Performance
649-744.
(1971), p.
57.
.
,
23
Soelberg, P.O., A Study of Decision Making:
58.
Thesis, Carnegie Institute of Technology, 1966.
P.O.,
Soelberg,
59.
Management Review, V.
"Un programmed
9 (1967), p.
Decision
Job
Choice
Making,"
Ph.D.
,
Industrial
19-29.
Stigler, G. , "The Economics of Information," Journal of
213-225.
Economy, V.
69 (1961), p.
60.
Political
Summers, S.A., "The Learning of Responses to Multiple Weighted
61.
64 (1962), p.
Cues," Journal of Experimental Psychology V.
29-34.
,
Tversky, A.
and Kahneraan,
62.
Numbers," Psychological Bulletin,
D.
"Belief
,
V.
76,
N.
2
the Law of Small
in
105-110.
(1971), p.
D.
Tversky, A.
and Kahneman,
"Judgement under Uncertainty:
63.
1124-1131.
Heuristics and Biases," Science. V.
185 (1974), p.
,
Tversky, A.
and Kahneman,
Psychology of Choice," Science,
64.
D.
V.
,
"The Framing of Decisions
453-458.
(1981), p.
Wheeler, G.
and Beach, L.R., "Subjective Sampling
65.
and Conservatism," Organizational Behavior and Human
36-46.
(1968), p.
,
and
the
211
Distributions
Choice V.
3
,
Witte, E.
"Field Research on Complex Decision-Making Processes 66.
The
Phase Theorem,"
International Studies
Management
of
and
156-182.
Organization. V.
2 (1972), p.
2, N.
,
BASEMENT
Date Due
®^'?'
y.r
MAY071W
M13-QS
SEP
mk
4
m
DEC
20
I^^^^^
^990
8 isi
1
Wl
19^6
1
5 'if
jwtoaiMi
na 171
^p
07'88
Lib-26-67
ACME
BOOKBLND.'NG CO., INC.
NOV
1
1983
100 CAMBRIDGE STREET
CHARLESTOWN. MASS.
I
Hn28.M414 no.ll84- 81
Van Breda, Mic/lnterpretmq inflation
D»BKS
P.».BKS...
13
74201,3..,..
no,1185-
HD28.IVI414
)»5KS
15 aa?
^
81
0/MAXBAND a versatile
non^efi^
o«BKS
John
Little,
gOI?:'fi33
DOl
TDflO
3
HD78-M414 no.1193^ 81
Roberts, Edwar/lnfluences on
:
74201S
111
3
Toao 002 000 bsa
HD28.M414
p
innovalio
0D133
no,
1194- 81
Rotemberg, Jul/Sticky prices in the Un
742181
.D»BKS
0013304.C
-
lilllllllllllllllllllllllllliy.l.lililil
TOaO DOl TTb 2hQ
3
HD28M414
no.1l86- 81
Rotemberg, Jul/Monopolistic price
D»BKS
"
741994
art/u
0013^641
m
TOaO 001 TTb 03^
HD28.IVI414
Ralph.
Katz,
no.1187- 81
/An investigation
D»BK5
742019
III
Hill mil
I
nil!
I
Ill
into
t
0013356Q
mill
11
I
mill
11
TOaO DD2 DD7 307
3
HD28.M414 no.1188- 81
Beckhard, Rich/Challenges and issues
742004
D»BKS
0013557.6
TOfiO
3
002 031
i
TflM
HD28.M414 no.ll89- 81
Rotemberg, Jul/Wlonetary policy and cos
741997
D*BKS
PU 1.3.10.44
,.
Toao 002 000 a3i
3
HD28,M414 no.1l90- 81
Medoff, James /Involuntary termination
00]21O,5"
.D»BKS
742002.
.
TOaO 002 001 037
3
HD28.IV1414 no.l191- 81
Treacy, Michae/Toward a behavioral the
QOl.35.529,
D^BKJ
7420dp
II
I
III
ill
III
111!
I
III
II
Jl
TOaO 002 DM2 2Tb
3
HD28-IVI414 no.l191- 81 1981
Treacy, Michae/Towards a behaviorally
D*BKS
743101
Q,0,1.3W.2.7.
TOaO 002 DM2 2b2
3
HD28.M414 no.ll92- 81
Von Hippel, Er/lncreasing innovators'
11
3
00133580
D*BKS
74261G
II
II
l|l|!i
I
i|!|i
11!
'|l
III
I"
mill
111 illlN Till
Toao 002 007 saa
3
TOaO 002 ODD 7b5