Knee-deep in the big muddy : the effect of personal responsibility and decision consequences upon commitment to a previously chosen course of action

UNIVERSITY OF

ILLINOIS LIBRARY

AT URBANA-CHAMPAIGN

BOOKSTACKS

Digitized by the Internet Archive

in 2011 with funding from

University of Illinois Urbana-Champaign

http://www.archive.org/details/kneedeepinbigmud220staw

Faculty Working Papers

KNEE-DEEP IN THE BIG MUDDY:

THE EFFECT OF PERSONAL RESPONSIBILITY

AND DECISION CONSEQUENCES UPON COMMITMENT

TO A PREVIOUSLY CHOSEN COURSE OF ACTION

Barry M. Staw

#220

College of Commerce and Business Administration

University of Illinois at Urbana-Champaign

FACULTY WORKING PAPERS
College of Commerce and Business Administration
University of Illinois at Urbana-Champaign
November 25, 1974

KNEE-DEEP IN THE BIG MUDDY:

THE EFFECT OF PERSONAL RESPONSIBILITY

AND DECISION CONSEQUENCES UPON COMMITMENT

TO A PREVIOUSLY CHOSEN COURSE OF ACTION

Barry M. Staw

#220

...

Knee-Deep in the Big Muddy:

The Effect of Personal Responsibility

and Decision Consequences upon Commitment

to a Previously Chosen Course of Action

Barry M. Staw
Organizational 3ehavior Group
Department of Business Administration
and
Center for Advanced Study
University of Illinois at Urbana-Champaign

Abstract

It is commonly expected that individuals will revprse decisions or
change behaviors which result in negative consequences. Yet, within
investment decision contexts, negative consequences may actually cause
decision makers to increase the commitment of resources and undergo the
risk of further negative consequences. The research presented here
examined this process of escalating commitment through the simulation of
a business investment decision. Specifically, 240 business school
students participated in a role-playing exercise in which personal
responsibility and decision consequences were the manipulated inde-
pendent variables. Results showed that persons committed the greatest
amount of resources to a previously chosen course of action when they
were personally responsible for negative consequences.

Knee-Deep in the Big Muddy: The Effect o: Pcrsonnl Responsibility
and Decision Consequences upon Commitment
to a Previously Chnst i Course of Ac

Intuitively, one would expect individuals to reverse decisions or
to change behaviors which result in negative consequences. Yet, thei
seem to be many important instances in which persons do not respond as
expected to the reward/cost contingencies of their environments. Speci-
fically, when a person's behavior leads to negative consequences we may
find that the individual will, instead of changing his behavior, cognitively
distort the negative consequences to more positively valenced outcomes
(see, e.g., Abelson et. al., Aronson, 1966; Staw, 1975; Weick, 1966). The
phenomenon underlying this biasing of behavioral outcomes is often said
to be a self- justification process in which individuals seek to rationalize
their previous behavior or psychologically defend themselves against adverse
consequences (Aronson, 1968, 1972; Festinger, 1957).

No doubt, the largest and most systematic source of data on the justification
of behavior following adverse conseciences is provided by the literature of
forced compliance. Typically, in forced compliance studies an individual is
induced to perform an unpleasant or dissatisfying act such as lying to a
fellow subject about the nature of a task (e.g. Festinger & Carlsmith, 1957;
Collins & Hoyt, 1972; Calder, Ross, & Insko, 1973), writing an essay against
one's own position (e.g. Cohen, 1962; Linder, Cooper, & Jones, 1967; Sherman,
1970), eating a disliked food (Brehm, 1959), or performing a dull task
(e.g. Freedman, 1963; Weick, 1964; Pallak, Sogin, L Van Zante , 1974) . Negative
consequences result from carrying out each of these counterattitudinal acts
when no external rewards are present to compensate for the dissatisfying
nature of the experimental task (Collins & Hoyt, 1972). However, since it
is difficult for the subject in forced compliance experiments to undo the

2
consequences of his acts, it is predicted that the individual will bias
his a lc on the experimental task (or change his opinion on an attltudi-
nal ) so as to cogn negative lies resulting irom

his behavior. In short, Che tndtvicael is predicted to justify his previous

behavior or defend himself from negativt consequences through the perceptual

2

biasing of behavioral outcomes.

Recent empirical research has shown that there are two basic preconditions
for the biasing of outcomes within forced compliance situations. First, the
individual must have committed himself to behavioral consequences which are
irrevocable or at least not easily changed (Brehm & Cohen, 1962) . If it is
readily possible to reverse one's own behavior, then this course of action
may often be taken tc reduce negative consequences rather than any biasing
of behavioral outcomes (Staw, 1974). Secondly, the individual must feel
personally responsible for the negative consequences of his behavior (Carlsmith
Freedman, 1968; Cooper, 1971). That is, a person must perceive at least a moder
degree of choice in his behavior (Linder, Cooper, & Jones, 1967), and the possi-
bility of negative consequences should have been anticipated at an earlier
decision point (Brehm & Jones, 1970; Cooper, 1972).

Self- jus ti flcat ion in Investment Decision Contexts

iugh forced compliance studies have provided a great deal of data on
the biasing of behavioral outcomes, there remain a large number of situations
•vhich individuals may be co go beyond the distortion of negative con-
sequences to rationalize a behavioral error. For example, one societally import
context in which individuals may take new and concrete actions to Justify their
behavi lowing negative consequences is that of investment decision making.
Investment decision contexts are considered broadly here as situations in whi
resources are allocated to one decisional ative ithers, but in which
the level of resources can he increased or decreased at the ditcrc
decision maker.

When '. ve consequence! ncurr. investment context,

it is often possible for a decisi to gre I the commitment

of resources and undergo the risk of additional negative outcomes Jn order
to Justify prior behavior or demonstratt the ultimate rationality of an
original course of action. It follows, however, that committing additional
resources to a losing decisional alternative can also turn into a negative
cyclical process. That is, due to a need to justify prior behavior, a
decision maker may increase his commitment in the face of negative conse-
quences, and this higher level of commitment may, in turn, lead to further
negative consequences. Within the sphere of governmental policy making,
Just such an example of committing resources to a costly decisional alternative
was described by George Ball, the former Under Secretary of State, in some
early observations on U.S. involvement in Indochina.

...Once large numbers of U.S. troops are committed to direct
combat, they will begin to take heavy casualties in a war they
are ill-equipped to fight in a non-cooperative if not downright
hostile countryside. Once we suffer large casualties, we will
have started a well-nigh irreversible process. Our involvement
will be so great that we cannot--without national humiliat ion--
stop short of achieving our complete objectives. Of the two
possibilities, I think humiliation would be more likely than the
achievement of our objectives--pven after we have paid terrible
costs. (Memo from George Ball co President Lyndon Johnson,
July, 1965; source: The Pentagon Papers, 1971).

Obviously, many factors may have influenced governmental decision making in the

commitment of men and material to the ve.r in Indochina. But, the comments of

this high level official do underscore the need for research on the possibility

that important resource investment d ,ns may be influenced by the reluctance

of individuals to oast mistakes or a need to justify prior behavior.

Assessing Se 1 f - jus 1 1. f icat ion in Invfsta.ent it :'.s

An empirical test of self-jus ion in an investment decisiion context

would seem to involve an assessment of whe ' r not ney consequences

serve to increase individual -it to a decisional

4

However, an unambiguous test of sel f- Just i ficat ion would raorc

than the simple manipulation '-s and the an

sequent com: .0

account for che same emp ielat^onsh ; tment >n-

sequences. One such mechanism might be r.h. on makers to

maximize their own outcomes, sine- lely wht n negal I

consequences have been incurred that a new and larger commitment to a deci;
alternative will pay off in the future. A separate but related mechanism
which may also account for the effect of negative consequences on the commit-
ment of resources may be a "gambler's fallacy" that resources should always
be placed in a losing decisional alternative since "things are bound to get
better". Implicit in the notion of a gambler's fallacy is the perception of
long-run equality of investment alternatives and the non-independence of
outcomes over time (see Lee, 1971).

The separation of self- just if icat ion from alternative theoretical
mechanisms within an investment decision context may depend upon manipulations
conceptually similar to those used in previous forced compliance studies,
noted in several earlier studies (e.g. Collins & Hoyt, 1972; Calder, Ross.
Insko, 1973), the rationalization of one's behavior has been shown to be
significantly affected by the manipulation of prior choice and negative con-
sequences. Within an investment decision context, self- Justification may
similarly depend upon the levpl of personal responsibility one has had
determining a particular course of action and the outcomes resulting from
those actions. Thi riment described below was therefore designed to

teat self-Justification within an investment decision context by manipulating
these two independenr ibles and measuring t> 'he commit-

ment of resources to a previously chosen course of Through I

maximization o or a gambler's fallacy, one might expe^

5
sequences to cause an in. in the commitment of resources to a
llona] the simple consistency

actions over time .ilso expect Individual! to increase their

commitment to a d. -.a 1 alternative for which they .iave had some prior
choice. However, only self- Just if ication would predict an interaction of

^onal responsibility and decision consequences such that increases in
commitment world be even greater than the additive effects of these two
separate factors .

Method
Subjects

The subjects of this experiment were 240 undergraduate students enrolled
in the College of Commerce and Business Administration at the University of
Illinois, Urbana-Champaign. Subjects had volunterred to participate in a study
on financial problem-solving as one means of fulfilling a course research
requirement. Upon arrival, the subjects were asked to work on the "A & '
Financial Decision Case" in which it was necessary to play the role of a
corporate executive Ln making some decisions about the allocation of research
and development funds.

As students in a business school, subjects generally were experienced in
working on i a cases in which an organizational or financial scenario is
presented and some action or set of actions are called for by the stude~
However, in order to maximize the involvement of subjects and to provide a
rationale for the study, the experimenter told each subject that the purpose
the case was to examine the effectiveness of business decision making under
various amounts of information. Each subject was told that the particular case
on which he would be working contained only a limited amount of informatio

that the information provi' >uld still be sufficient for a business school
8tu make "a good financial deciBi ejects wt re asked to do
heat Job they could on the cases and to place their names on each p I the c

mfl f-*>r i A 1

6

Th- A & S Financial Dec 1b Ion O

Th< cial decision case ui this scudy ^al

poration in . Th-

• r8 of sales and e. ) of the "Adwrns & '

Conpany", and a se» is pre .i in which the suh .• s asked to play

a major rol nancial decision-making . As stated in the <.-■ th«

profitability of the A & S Company, a large technologically-oriented firm,
has started to decline over several preceding years, and the directors
the company have agreed that oae of the major reasons for the decline in
corporate earnings (and a deterioration in competitive position) lay in some
aspect of the firm's program of research and development. The case further
states that the company's directors have concluded that 10 million dollars of
additional R&D funds should be made available to its major operating divisions,
but, that for the time being, the extra funding should be invested in only one
of the corporation's two largest divisions. The subject is then asked to act in
the role of the Financial Vice President in determining which of the two
corporate divisions, Consumer Products or Industrial Products, should receive
the additional R&D funding. A brief description of each corporate division is
included in the case mar ie subject is asked to make the financial

investment decision basis of th< itial benefit that R&D funding

will have on t' nings of the divisions. In addition to circling the

chosen division, subj. re also asked to write a brief paragraph defend

r allocs tirr s .

After compl' case and turning it in to

experimenter, s*. linist secon' ise which

■'.ed another financial investment decision. Par; financial

Decision Case presents the subject with the condition of Adams & Smith Com"'
in 19 Ltlal alio f rt se

funds. As state.d in Par: i ia agaii

7
for r« mpany la convinced that

Is an - In

, 20 million dollars has been nvi ipital

R it D funding, and the sub President, is again

asked pon its prop' tion. This time, however, the subjec

J to divide the R&D money in any way he wishes among the two major
corporate divisions. Financial data (e.g. salts and earnings) is provided
for each of the five years since the initial allocation decision and, as
earlier, the investment decision is to be made on the basis of future contr
bution to earnings. Subjects made this second investment decision by specifying
the amount of money that should be allocated to either the Consumer Products

Industrial Products divisions (out of a total of 20 million) and again wrote
a paragraph defending the decision.

Mani pulat ion of Consequences

Decision consequences were experimentally manipulated in this study through
the random assignment of financial information. One half of the subjects were
provided information that the division initially chosen for R&D funds sub-
sequently pf r formed better than the inchostr d vision, while one half were
given Information showing the reverse. For example, in the positive con-
sequences condition, subjects received financial data which 3howed that the
chosen division had returned to p ble levels while the unchosen division
continued to decline. In a pa. manner, subjects in the negative con-
sequences condition received financial data which showed a deepening decline
in the profitability of the chosen division but an improvement in the unchosen
division. The exact natirr- of the financial data provide subjects is shown
in Tables 1 and

1 and 2 abo

8

Mtinipt: 'ml Respo i ty

assigned to
n in w;
roadt >rmed t

>n cast described above in which subjects made an initial decision to
allocate R&D funds, discovered its consequences, and then made a secc
inv l decision. Howev ^ half of the subjects v so randomly

assigned to a low personal responsibility condition in which the ei

uncial decision case was presented in one section. In the low person
responsibility condition, subjects were asked to make the second allocation
decision without having made a prior choice as to which corporate division
was most deserving of R & D funds. Subjects in this condition received one
set of case materials which described the financial condition of the AdamF &
Smith Company as of 1972, the time of the second R&D funding decision. They
were told in the case that an earlier R&D funding decision had been ma
in 1967 by another financial officer of the company and that the preceding
officer had decided to invest all the R&D funds in the Consumer (or Industr^
Products division. The financial results of each corporate division (e.g. sal
and earnings data) were presented from 1957 to 1972, and, like other subjects,
persons in the low responsibility condition were asked to make the (second)
R&D runding decision based upon the potential for future earnings. In s1
the information presented to low personal responsibility subjects was
identical to that given her subjects except for the fact that the

case's scenario bt in time (1972 rather than 19b"

necessitated making the second in* it decision without having participa

n earlier cholc-

Depend

The dependent variable utilize in this study was the individuals'
commitment to a previously it alternative. This variable

was opera t ional ized bv the amount of money subjects allocated on the
second R&D fu the corporate division chosen earlier

(either chosen earlier by the subject or the other financial offic
mentioned in the case). The amount allocated to the previously chosen
alternative could range between zero and 20 million dollars.

Summary of Treatment Groups

Of the 120 subjects in the high personal responsibility condition,
64 initially chose the Consumer Products Division as the best investment
for R&D funds, while 55 initially chose the Industrial Products Division.
(One subject was unable to make a choice between Consumer and Industrial
Products and therefore had to be excluded from further analyses). Since
jects self-selected themselves to prior choices and then financial in
mation was randomly assigned, four cells were created by initial choice and
financial information. However, as shown in Table 3, these four cells can
be collapsed into two primary treatment groups of positive decision conse-
quences and negative decision consequent

Insert Tab;> '5 jbout here

Of the 120 subjects assigned to the low personal responsibility con-

dit were also assigned to each of the four cells described above

For example, thirty were given cases in which another financial officer had
chosen the Consumer Products Divifi _ontinued to decline; thirtv

10
: which anot1 iancial o had ch
Products start.' improve; thirty in which

another financial • r had chosen the Industrial Products Division and it

and, thiv -ked on cases in which IndustriJi Prodis
was chosen a A^ e four cells can be collapfc

into two treatment groups of positive and negative decision consequences com-
prising 60 subjects in each.

The final form of the design of this experiment was a 2 x 2 factorial
in which personal responsibility and decision consequences were the manipulated
independent variables. As stated earlier, the amount of money invested in
the previously chosen corporate division (previously chosen either by the
subject or the other financial officer mentioned in the case) was the
dependent measure utilized in the study.

Results

Preliminary Analysis

A preliminary analysis was conducted to determine whether the object
of a subject's prior choice (Consumer Products-Industrial Products) or
the exact form of financial information (Ctli or C*lt) affected the amount
of money allocated to the previously chosen alternative. If there were

effects of either of these two variables then it would not be possible
to collapse the eight cells shown in Table 3 into a 2 x 2 analysis
variance. As can be seen from t; ; Table A, there were no ma

effects of either the objec :e (F < 1.00, d_f ■ 1/231, n.s .) or

exact form • mcial in ion (P < 1.00, df » 1/231, n.s .) .

rt Ta1 j bout '

raonal Responsibility and Decision Consequences
S: no main effects of the ob choice and financial

informa x 2 analysis of vari is conduc nal

11

ion consequences were th<

Table 5 shows that there significant mai, al

decision consequences, an.l i si int interaction of
ient v

Insert Tab

Under high personal responsibility conditions, subjects allocated an
average of irs to the corporate divisions they had eai 1 h r

chosen for extra R&D funding. Under low personal responsibility conditions,
subjects allocated an additional 8.89 million dollars to the corporate
divisions previously chosen by another financial officer. Under positive
decision consequences, subjects allocated an average of 8.77 million to the
previously chosen alternative, while 11.20 million was allocated under
negative consequences.

Interaction of Personal Responsibility and Decision Consequences

When subjects (personally) made an initial investment decision which
declined, they subsequently allocated an average of 13.07 million dollars
to this same alternative in the second funding decision. As shown in
Figure 1, the amount invested in the previously chosen alternative was
greater in the high personal responsibility - negative consequences condi-
tion than in any of the other three experimental conditions. Although
this result could have been expected from two significant main effects of
personal responsibility and consequences, th< rrence between the high
personal responsibil, /e consequence condition am

cells was of such magnitude as to produce a significai, •< raetion.
Furthermore, a close analysis ol shows the only significant

dif a among any of the four ex ntal conditions were between th*3
high responsibility - negative consequences cell a,. er th- >er-
mental conditions. For example, c .ences did not have a sign!'

12
■ under low p 1 responsibility conditions (t-1.20; df-llfl; n .B .) ,

intly <ifiect results under positive
sequences conditions (t^l.13, d_f 118, n.a .) .

Insert it here

Discussion

Interpretation of the Effects

The main effect of decision consequences upon commitment to a previously
chosen alternative could be explained by a maximization of gain hypothesis.
Either through the objective re-appraisal of action-outcome contingencies
following negative consequences or through a "gambler's fallacy" that the
probability of gain i9 increased by prior failure, individuals could have
decided to increase their investment of resources. However, it is interest-
ing to note that, although a maximization of gain hypothesis provides an
adequate explanation of the main effect in analysis of variance terms, its
explanatory power is somewhat weakened when individual cell means are con-
sidered. Specifically, while maximisation can account for the effect of
decision consequences under the high responsibility condition, it is less

clear why there was no significant effect of consequences under the low

4
responsibility condition.

A related interpretive problem also weakens the consistency of choice
explanation of the ma: of personal respo. . t For examp
it may well be true that to consistency in choice decisions, indivi-
duals will allocate more m alternative that was per-
sona arlier point li ,g., under high responsibil
than one chosen previously by sonv r low responsibility).

.!ual cells of t iysis of variance I amined
(see P appears that I of personal

is not y consisten y. Only und> ces

was ence between tho high ;ind low responsibil.

conditions, a I thou j was a nonsir ;nt trend under posit:. n-

■

Thus, from the data of this study, it is not unreasonable to conclude that
the primary effect of responsibility and consequences was that individuals
invested a substantially greater amount of resources when they were per-
sonally responsible for negative consequences. The significantly greater
commitment of resources under this one experimental condition clearly accounted
for the interaction of personal responsibility and decision consequences.
However, a close examination of Figure 1 also shows that the substantial
difference between the condition of high personal responsibility - negative
consequences and the other cells could also underlie the statistical significance
of the two main effects. As a result, the data from this study provide even
somewhat stronger support than expected for the hypothesis that Individuals
who are personally responsible for negative consequences will increase the
investment of resources in a previously chosen course of action.

Sel f- just i f ication Versus Se If -percept ion

Frequent 1 n a s- on process is experimental! ' ed

its outcroppings at lit to separate -e derived from se'

perception theory (Ben, 1967, 1972). The dis on between self- Juitif ic
and self-perception is also impor crpretation of the present

study and shoul> ,nsldered in some depth.

14
In e^ >f st-1 f- 1uat i f lcat ion veriufl sr 1 t -percept ion

Ives around dual formula of the process of rationalization. On

the one hand, s sLification (Aronson, 1968, 1972) or dissonance

theory (Fes* • . 1957) posits that individuals possess a potent need

to restore the "appearance" of rationality to their own behavior. As a
result, the theory predicts that individuals will cognitively re-evalua:
decisional alternatives after an important choice (e.g. Walster, 1964;
Knox and Inkster, 1968; Vroom, 1971) or actively distort the

characteristics of a behavioral task (e.g. Festinger & Carlsmith, 1957; Weick, 1966)
On the other hand, self -perception theory posits that individuals retrospectively
restore rationality to their behavior by simply inferring the causes of their own
actions within a social context. Self-perception theory predicts that
individuals will re-evaluate their behavior so that it conforms to their
own notions of how one might feel or behave if he were acting rationally.
Thus, like self- justification, the retrospective analysis of behavior which
comprises self-perception theory can also account for the re-evaluation of
alternatives following a decisional choice (see Kelley, 1967, 1971) or
changes in the perception of the characteristics of a behavioral task (see
Calder & Staw, in press; Deci, 1971, 1972; Salancik in press; Staw, 1975).

It is possible that a self -perception analysis can al9o be usefully
applied to the effects of personal responsibility and decision consequences
within an investment decision context. For example, when individuals personally
select a course of action which results in negative consequences, they may

-ospectively infer that their prior choices were especially meritorous in that
they required some suffering and, as a result, they may subsequently choose to
invest even greater amounts sources in the losim rnative. This

caus is pla

i disposition
r ive p: -ing a se
per vsis is the is substantial body of

ce which ah. o avoid the self-

attributi. Iead6 to negative consequences

or results in personal failure (see Weiner, Frieze, Kukla , Reed, Rest, and
Rosenbaum, 1972). Thus , it would seem very unlikely for individuals to
attribute greater internal causality (and therefore invest more) in a
previously chosen alternative which has led to negative consequences.
In contrast, it would seem more likely for individuals to take concrete
actions to reduce negative consequences for which they are responsib!
or at least to attempt to reduce those negative outcomes which cannot
be attributed to an external source. This latter interpretation is con-
sistent with a self- justification notion that individuals actively seek to
maintain or restore the appearance of rationality to a previously chosen
course of action.

Self- just! f lcation and the Escalation of Commitment

As we have ndividuals are personally responsible for negati

consequences , they may decide to increase the investment of resources to
prior course of a* allows that this 9ame process of escalation may

also occur in many di- contexts In which additional effort, ai
resources are committed to an unsatisfactory policy alternative. Thus,

search should focus on the critical factors underlying the escala'
of resources, both in terms o resources committed and the num-

ber of times an increase in resources will be made to a decisional alternative-.

t varlab f may ount of loss

already incu '4, for discuss,

of t ietnam Dollar" phenom. the p< jd efficacy of the
resources being cc < .g. the ability of R & D expenditures to
increase future profits), the nature of the decision making entity (e.g.
Individual decision maker vs. group decision making body), personal
characteristics of the decision maker (e.g. self-esteem, tolerance for
ambiguity), and the evaluative consequences of the situation.

One conceptual note which could prove useful in future studies of the
escalation of commitment is the distinction that, within investment decision
contexts, there may be two separate sources of self- Justification. First,
an individual may desire to demonstrate rationality to himself or restore
consistency between the consequences of his actions and a self-concept of
rational decision making (Aronson, 1968). This may be a rather ubiquitous
phenomena as has been demonstrated by research on cognitive dissonance and
other consistency theories (see Abelson, Aronson, McGuire, Newcomb, Rosenberg,
& Tannebaum, 1968). Secondly, the individual may attempt to demonstrate
rationality to others or to prove to others that a costly error was really
the correct decision over a longer term perspective. This second form of
self- Justification would seem to be most important in organizational contexts
where a decision maker may be uncertain of his own status within a social hierarchy,
in gov .tal policy I one in which a decision maker teay be anxious
t his political standing among constituents. No doubt, these two forms
of self-Just if icatlon could both be viewed as face-saving activities
(Goffman, 1959), wit! istlnction of an internal versus external orienta-

n on the part of I .on maker. However, while the first form of
self- Justification tat aaed on a general human need to be consistent and

rect (Fes- , 1957; White, 1959), the second form may relate to Individual
res for social approval (Crownc and ■ •, 1964) . Futur arch she

be

Just and ! I within

3 .

18

1. The .)! wishet. his gr /illian Brighton for

his help in thi <il materials, t R. Oldham,

. and G r comments on a'

aion of this manuscript, and to The Center for Advanced Study at the
University of Illinois, L-rbana -Champaign for the facilities necessary to
complete this study.

2. An active controversy exists over the theoretical interpretation of the
data from forced compliance studies (see Bern, 1967, 1972; Jones, et. al.
1968; Ross & Shulman, 1973). However, the issue of self-Justification
versus self -perception will be addressed in a later section of the paper.

3. Ina2x2x2 analysis of variance there was a corresponding main effect
of personal responsibility, 3n interaction of prior choice and financial
information (same as main effect of decision consequences), and a triple
interaction of personal responsibility, prior choice, and financial
information (same as interaction of responsibility and decision consequences).
It is possible, of course, to postulate (post-hoc) that the valence of

future outcomes was less for subjects under low rather lhan high responsibility
conditions, and, thus, the motive to maximize gain was correspondingly
weaker in low rather than high responsibility conditions.
5. Other at ctly) demonstrate the escalation of commitment

usin^ the door1' ique (e.g. Freedman and Fraser 1966) can

be inter; r an increase in the perception of internal causality

following n commitment or by an individual need to justify prior

behavior.

19
es

i, R. P.; Aronson, B.J M< N'ewcomh, R. M.; Re; M. J.,

& Tennenbaum, P. H. | ; Cogn h is tency , Chicn

Rar.

Aronson, F. . The i insu! it justification: An analysis

some conflicting data. In S. Feldman (Ed.) Cognitive Consistency.
M York: Academ s , 1966.

Aronson, E. Dissonance theory: Progress and problems. In R. Abelson,

Aronson, W. McGuire, I. Newcomb, M. Rosenberg, & P. Tannenbaum (Eds.),
Theories of Cognitive Consistency. Chicago: Rand Mc.Nally, 1968.

Aroi The Social Animal. San Francisco: W. H. Freeman and Company,
1972.

. D. J. Self-perception: An alternative interpretation of cognitive
dissonance phenomena. Psychological Review, 1967, 74, 183-200.

Bern, D. J. Sel f -perception theory. In L. Berkowitz (Ed.) Advances in

ntal Social Psychology, Vol. 6, New York: Academic Press, 1972.
W. & Cohen, A. R. Explorations in Cognitive Dissonance. New York:
Wiley, 1962.

Brehr, I. R. A., T": ct on dissonance of surprise conse-

quences. Journal of P< and Social Psychology, 1970

Calder, B. J., Ross, M. , & Insko, C. A. Attitude change and attitude

•ttribul choice, and consequences. Journal

oj_ ity and Social Psychology, 1973, 25, 84-100.

Calder, B. J. -w, B. S. The Be If -perception of intrinsic and extrinsic
motivation. J_ and Social Psychology, in press.

Carlsmitb Freedman, T Bad decisions and dissonance: Nobody's

perfect. In R. McGuire, T. Newc< Rosenbe:

P icy . Chicag.

20

Cohen, A. R. An experiment on small rewards for discrepant compliance and
;itude ch In J. W. Brehm & A. R. Cohen (Eds.), Explorations in
v.nitive Dissonance y, 1962.

Cooper Perso I dissonance: The role of foreseen

^sequences: An integration and extension of the "forced compliance"

literature. Journal of Experimental Social Psychology, 1972, 8, 558-594.
Crovme , D. P. and Marlowe, D. The approval motive; studies In evaluative

dependence . New York: Wiley, 1964.
Deci, F. . L. The effects of externally mediated rewards on intrinsic motivation.

Journal of Personality and Social Psychology, 1971, _18, 105-115.
Deci, E. L. The effects of contingent and noncontingent rewards and controls

on intrinsic motivation. Organizational Behavior and Human Performance,

1972, B, 217-229.
Festinger, L. A Theory of Cognitive Dissonance. Stanford: Stanford University

Press, 1957.
Festinger, L. & Carlsmith, J. M. Cognitive consequences of forced compliance.

Journal of Abnormal and Social Psychology, 1959, 58, 203-210.
Freedman, J. L. Long-term behavioral effects of cognitive dissonance.

Journal of Experimental Social Psychology, 1965, 1, 145-155.
Freedman, J. L. & Fraser, S. C. Compliance without pressure: The foot-in-

the-door technique. Journal of Personality and Social Psychology, 1966,

4, 195-202.
Goffnan, E. The presentation of self in everyday life. Garden City, N.Y.:

Doubleday, 1959.
Jones, R. A., Linder, D. E., Kiesler, C. A., Zanna, M. , & Brehm, J. W.

Internal states or external stimulus: Observer's attitude judgements

and the dissonance-self-persuasion controversy. Journal of Experimental

Social Psychology, 1968, 4, 247-269.

Kelley, H. H. Attribution theory in social psychology. In D. Levine (Ed.),
ska Symposium on Motivation, Lincoln: University of Nebraska Press,

19h
Kelley, H. H. Attribution in social interaction. New York: General Learning

Prt'ss, 19
Knox, R. & Inkster, J. Postdecision dissonance at post time. Journal of

Personality and Social Psychology, 1968, 13, 319-323.
Lee, W. Decision theory and human behavior. New York, Wiley, 1971.
Linder, D. E., Cooper, J., & Jones, E. E. Decision freedom as a determinant

of the role of incentive matnitude in attitude change. Journal of

Personality and Social Psychology, 1967, 6, 245-254.
Pallak, M. S. Sogin, S. R., Van Zante , A. Bad decisions: Effects of

volition, focus of causality, and negative consequences on attitude change.

Journal of Personality and Social Psychology, 1974, 30, 217-227.
Pentagon Papers , The New York Times (based on investigation reporting of

1 Sheehan), New York: Bantam Books, 1971.
Ross, M. & Shulman, R. F. Increasing the salience of initial attitudes:

Dissonance versus self-perception theory. Journal of Personality and

Social Psychology, 1973, 28, 138-144.
Salancik, J. R. Interaction effects of performance and money on self-perception

of intrinsic motivation. Organizational Behavior and Human Performance,

in press, 19
Sherman, S. J. Effects of choice and incentive on attitude change in a

discrepant behavior situation. Journal of Personality and Social
-hology, 1970, 15, 245-252.
Staw, Attribution of the "causes" of performance: A general alternative
ition of cross-' search on organizations. Organ! zat ■

Behavior and Human Performance, in press, 1975.

., B . S . Att itudinal "anging a major

Journal of Personality

.' Soc- ohology, 742-7 51.

Staw, B. S. Intrinsic and Excrinsic Motivation. New York: G< Learning

Press, in p 1975.
Vroom, V. H. Organizational choice: A study of pre- and post-decision

processes. Organizational Behavior and Human Performance, 1966, ^, 212-225
Walster, E. The temporal sequ. decision processes. In L. Festinger

(Ed.), Conflict, decision, and dissonai Stanford: Stanford University

Press, 1964.
Weick, K. E. Reduction of cognitive dissonance through task enhancement and

effort expenditure. Journal of Abnormal and Social Psychology, 1964, 68,

533-539.
Weick, K. E. Task acceptance dilemmas: A site for research on cognition.

In S. Feldman (Ed.) Cognitive Consistency. New York: Academic Press, 1966
Weick, K. E. Amendments to organizational theorizing. Academy of Management

Journal, 1974, _17» 487-502.
Weiner, B., Frieze, I., Kukla , A., Reed, L., Rest, S., Rosenbaum, R. M.

Perceiving the Causes of Success and Failure. New York: General Learning

Press, 19

23

Table 1

Consumer Products Contribution to Sales and Earnings
of Adams & Smith Company 1

Fiscal #
Year Sales

*
Earnings

1957 624

14.42

1958 626

10.27

1959 649

8.65

1960 681

8.46

1961 674

4.19

1962 702

5.35

1963 717

3.92

1964 741

4.66

1965 765

2.48

1966 770

(-12)

1967 769

(.63)

First R&D Funding Decision as of 1967

Manipulated Improvement

Manipulated Decline

Fiscal
Year Sales Earnings

*
Sales

Earnings

1968 818 .02

771

(1.12)

1969 829 (.09)

774

(1.96)

1970 827 (.23)

762

(3.87)

1971 846 .06

778

(3.83)

1972 (est) 910 1.28

783

(4.16)

~ond R&D Funding Decision as of 1972

n millions of dol]
1 Parenthese

i earnings

Table 2 2k

Inch Contribution to : md Earnings

of Adams £ Smith Company

Fi.s
Year

*

*
Earnings

1957

670

15.31

1958

663

10.92

1959

689

11.06

1960

711

10.44

1961

724

9.04

1962

735

6.38

1963

748

5.42

1964

756

3.09

1965

784

3.26

1966

788

( .81)

1967

791

( .80)

First R &

D Funding Decision i

as of 1967

Fiscal

Manipulated

Improvement

Manipulated Decline

Year

Sales*

Earnings*

Sales*

Earnings'

1968

818

.02

771

(1.12)

1969

829

(.09)

774

(1.96)

1970

827

13)

762

(3.87)

1971

846

.06

778

(3.83)

1972 (est) 910

1 . 28

783

(4.16)

Second

D Func

as of 1972

* f doll

1 P« net losse arnings

25

Table 3

Schematic Analysis of the Cells

to Which Subjects Were Assigned

Under Both High and Low Responsibility Conditions

HIGH PERSONAL RESPONSIBILITY

Financial Information

C+I +

C + I +

Consumer
Products

Initial
Choice

Industrial
Products

Positive

Negative

Consequences

Consequences

(n=32)

(n=32)

Negative

Positive

Consequences

Consequences

(n=27)

(n=28)

LOW PERSONAL RESPONSIBILITY

Financial Information

C+I +

C+It

Consumer
Products

Ini-
Choice

Industrial
Products

Positive

Negative

Consequences

Consequences

(n-30)

(n=30)

Negative

Negative

Consequences

Conseguences

(n-30)

(n=30)

26

Table 4
Amount of Money (in millions) Allocated to
Previously Chosen Alternative by Level of Personal Responsibility,
Object of Prior Choice, and F Lai Information

Personal

Responsibility

Prior

Che I

Financial Information
C+ 1+ C| Ii

9.36

12.56

Consumer

Products

positive

negative

consequences

consequences

High

13.46

9.00

Industrial
Products

negative
consequence s

positive
consequences

8.22

9.22

Consumer

Products

positive

negative

consequences

consequences

Low

9.65

8.48

Industrial

Products

fative

positive

;uences

consequences

21

Table 5
Analysis of Variance of Effects of
Personal Responsibility and Decision
Consequences upon Allocation of Resources
to a Previously Chosen Alternative

Source

df

MS

Personal Responsibility (P)
Decision Consequences (D)
Interaction (P x D)
Error

1

282.36

14.40

<.001

1

351.57

17.93

<.001

1

109.12

5.56

<.019

35

19.61

Figure caption: Amount of Money Allocated
to Previously Chosen Alternative by Personal
Responsibility and Decision Consequences

28

29

o

->

■

K
U
O

c

0)

>

u
<a

c
>-.

c
o

6

'.-

D

o

-

>

u
a.

13.00
12.50

12.00
11.50

11.00
10.50

10.00, .

9.50,.

9.00

8.50

8.00<k
7.50

ty Conditions

ty Conditions

Positive
Consequences

Negative
Consequences

Decision Consequences

tv