-
\1 042-2587 -91 -1 62$1 .50Copyright 1991 byBaylor
University
Person, Process, Choice:The Psychology of NewVenture CreationKel
ly G. ShaverLinda R. Scott
Psychology can be distinguished from other behavioral sciences
by its emphasis on thebehavior of the individual person. Behavior,
in turn, is influenceJ oy ttre way in wnicn ineexternal world is
represented in the mind, and by the individual's exercise 6f
choice. Trrearticle examines the Possibility. thatrelatively
endirring attributes of the perion riltri "tr""tentrepreneurial
activity, describes the social cognitive processes involvetl in
conitructingrepresentations of the external environment, and
suglests which motivational variablesaffect behavioral choices.
Although past research on ';lf,e psychology of the
entrepren"ui',llag-not been productiv-e, a psychological approach
based bn'persons, process, and cnoiceholds promise for the
future.
rln best scientific form, researchers interested in the creation
of new business
ventures have attempted to construct definitions of
entrepreneurship. A philosophicalfoundation for the term is
provided in J. S. Mill's treatisebn political ".onorny 1iA+A;,in_
which the assumption of risk is asserted as a key ingredient of
"nt."pr"neuriai activity.Modern use of the term is usually credited
to Schumpeter (1934), who emphasizedinnovation and its role in
destabilization of economic systems. The discipline's attention'' n
shifted from the act to the actors, with McClellan-d's (1961)
assertibn that qualitiesar's(ociated with a high need for
achievement-preference for challenge, acceptance ofpersonal
responsibility for outcomes, innovativeness-characterized
suicessful-initiatorsof new businesses. This focus on personal
characteristics continued (e.g., Dunkelberg &looper, 1982;
Hornaday & Aboud, l97l; pickle, 1964; Timmons, lvoa1,
despite'its!{lures (e.g., Brockhaus, 1980; Brockhaus & Horwirz,
l9g6; carsrud, oim, a nooy,1986) and limitations (Wortman, 1987),
until Gartner (1988) challenged the whoieapproach by arguing that
the behavior of creating a new venture, not thJpersonality ofthe
founder, should be fundamental to the definition.- _
Predictably, there has been a rejoinder to the challenge
(carland, Hoy, & carland,J?8!), T-d papers listing the personal
characteristics of entrepreneurs continue to flour-ish (Mcclelland,
1987; solomon & winslow, 1988;winslow & Solomon, l9g9).
More-over, because the act of creating an organization cannot occur
in isolation, otherresearchers have attempted to specify aspects of
the social (Carsrud & Johnson, l9g9)and economic (Hornaday,
1990) contexi in which such innovation takes place. In theextreme,
the contextualist position denies that the personal characteristics
of organiza-tion founders contribute anything of importance tb the
success of the venture"(e.g.,ll.d.i".h, 1989), regardless of
whether the contribution originates in personality or
inbehavior.
In summary, a substantial amount of journal space has been
devotedWnter,199
to the attempt
ZJ
-
to define "entrepreneurship," but no consensus-has emerged.
Notwithstanding this lackof agreement on a fundamental term, there
has been tremindous growth in tnE stuay orventure creation. lndeed,
a recent software package (EntreBIB-Fc) catalogues n"Lryl4'000
relevant entries' we endorse Mitton's (198i) conclusion
that'entreprJn"u.ririp i.like obscenity: Nobody agrees what it is,
but we all know it when we see it.' In the faceof such-
controversy, prudence dictates adherence to op".utionul
definitions, ,o irrir,"remainder of this paper we shall be
discussing "new^venture creation,,, rather than"entrepreneurship.
"
FUNDAMENTALS OF A PSYCHOLOGICAL PERSPECTIVEllaving limited
ourselves to an account of "new venture creation,,, we believe it
isuseful to describe the essential ingredients of a psychological
approach to the study-ofany phenomenon. Most of you will recognize
thai we havJjump"o t o- one definitionalthicket to another. There
are psychologists who study the ;;sJJiu. pu-p,'-tr,"
"i""_trochemical process through which changes in ion
concentrations at the cell membraneproduce- action potentials in
nerve cells. There are psychologists who investigate mail-ematical
reasoning processes in toddlers. And there are psyciotogists who
eiamine theways in which entire communities mobilize in response to
calam"itier. rr urt"Jio gi;"their particular outlooks on
p_sychology, these various researchers might generate nearlythe
controversy found in definitions of entrepreneurship.Despite this
potential for dispute, psychologists have been able to agree on
certaincore ideas, and these are recounted in our introJuctory
textbooks to delimit the domainof inquiry (e.g., Shaver &
Tarpy, in press). As a disciiline, psychorogy is rarely aenneaas
the sum of the activities of its practitioners. Rather, it
is"oistinidsrrea r.om oitrersocial or behavioral sciences, such as
sociology, anthropology, uni".ono-i"s, uy-ii.emphasis on the
individual person as the level of analysii. wTitrin this
concentration onthe indivirlual Qterson), two of psychology's core
theoretical concerns have been thecontents of mind (the process
intervening between external world and observable be-havior), and
the exercise of free choice.
Emphasis on the PersonThe basis for the boundary distinction
is.the unit of analysis. Where anthropologists
emphasize cultural influences on actions, and sociologists
emphasize social structure andorganization, psychologists
concentrate on individuJs. A translation of the Greek rootsof the
discipline's title would be "the study of human spirit or soul.,,
Although therehave been diverse, if not to say contradictory,
descriptioni of what a human spiJt, soul,or mind might be like,
psychology has always recognized that whatever the discription,a
mind exists within a single individual. Consequeritly,
psychologists are predisposed tosearch for explanatory concepts
that can be located *ittrin the person.'For example,when attempting
to account fgr the aggressive behavior found in collections "f
;;t-mous individuals, the sociologist LeBon (1895) was comfortable
with a notion tt" te.*Jo"the Group Mind. " In the.h.""9: of social
psychologists, this group-based prrenomenonwas transformed into
individualized processes of be[avioral cdntagion (Wheeler, 1966)l'
A recent paper by Funder and Dobroth (1987) investigated the ease
of judgment of 100 personalitycharacteristics, and found that among
the l5 with the highe;t level of interjudge agreement were .'tends
toarouse liking and acceptance," "behaves in an assertivJfashion,"
"is a taikaiive individual," ,.rebelliousand nonconforming," and
"power oriented." These traits could easily be part of
Mitton'siescripil"", "rthe "compleat entrepreneur," so it may be
that we all do know ona *h.n',r" see one.
24 ENTREPRENEURSHIP IHEORY ond p|?ACT|CE
-
or deindividuation (Zimbardo, rg70). Not j.ust-group action, but
arso group inaction, isinterpreted by psychologists in
individualized-terms. wff a murder in New yorkoccurred before a
large group of onlookers-virtually non" oirhom
intervened-socialcommentators were content with explanations in
terms of ..uputhy,, (Rosenthal, ,J,&i).But experiments designed
as anatogues of the r"irirg r9o nrylilorogirt. to an explanationin
terms of a sequence of decisioni made by each firai"ii"1"i'ilatan6
& Darley, 1970).Thus the first element of a psychological
approach to new u"niur" creation is a focus onthe person.This view
obviously contrasts with the extreme contextualist description of
newventure creation.2 For.tfie psychologist, it is not enoughio
tnl* trrut a certain numberof people wil respond to tie."push"', of
being rurtoultJ,'o. to tn. ..puil,, of marketportunities by
starting businesses on the.ir"own. n?-,-r,..,"ti. pry.torogist
wants toRlow which few of the hundreds laid off will create n"*
u"ntui"s, or why not everyonesees-and capitalizes upon-the marker
opportunity. A word "a;;ifi i;;;;;r,history of the entrepreneurship
riterature,..this broad psychorogicar approach has beeninadequately
represinted by itudies oi"'trre.per_sonality of itre
entrepreneur.,, Thef*g:lv fruitless quest for ttre personality
profile or trre successful organization founderis what
psychologists.would call a persliotogicat "no"uuo.. Such searches
for transsi-tuational consistency in-p_ersonality traits went. out
of style in psychology over twenryyears ago, when Miichel (1968)
;;";J;^luriu"tyit at'b"h;;i;, shourd be regarded asthe consequence
of person-situatioin intfractions. withourrecounting alt of the
subse-quent theoreticar discussions (e.g., see Funder &
Dobroth,- tggl, Kenrick & Funder,1988)' suffice it to say that
psych-oiogy's emphasis on it" p".r"" as the unit of
analysisencompasses both techniques and processes well beyond ihe
identificatio. oi ,p"iii"personality variables presumed to Gad to
the foundi;g ;; n"*'orgunirations.
The Process of Representing the External WorldThroughout its
history, psychology has recognized the biological foundation
forhuman and animal behavi,or,-eiamin#the contents of mind, assumed
that there wereI ts to rationalitv. and noted the.importance of the
immediaie social context. But eachof-these elements of the
psyc.horogilul'upioach depends on the presence of at leasr
some"mental furniture" that originaLo ouiiioe.tt. orgunr.-."i"iti.
sense, psychorogyfollows in the Aristotelian tridition of
empiricism,"rather ttran tne platonic view thatsensory information
is mere flickering shadows on the cave wall. consequently, it is
notsurprising that the oris_inal truly psyciorogicar
ry,1"0 *",'plyrnopnyrjcs, the srudy ofthe relationship between
externai siimulation and the internal psychological experience-oj
the individual' specification of the process by which thetternal
world becomesrepresented in the mind is the first of psychology'.
"or" ron.".n, that is fundamental toa psychological approach to new
u"niui" creation.Because psychology combines external circumstances
with intrapsychic processes,it can be defined by Gwin's
expression,
--
B : fiP, E),meaning that behavior is a function of both person
and environment (Lewin, in cart_wright, l95l)' Neither alone
constitutes a sufficient explanation for an
individual,s2.SeeGergen(l982)foranoutlineofthecontextua|istposition,no,,.no'o,,ff
i3:y"i11':iiifj :lfjT3::::.:"",,x,1,3:;*:.:ii.1:l1T:.u1,nc, are
negotiated within the interpersonarcontext" rhus, rike the
contextuarist position *i,r'r" "r* ,."ffi
.ffifiil1ffiJ'iln,rilHffJ,'fJ^jsearch for any transsituationally
enduring p"rronui
.n_ucteristics.
\Mnter,199,l
-
observable behavior. Although a person's cultural milieu and his
or her membership invarious social networks and organizations
certainly influence action, study of thesefactors in their own
right is beyond the purview of psychology. The
psychologicalquestion is how a person's cultural heritage and
social neiworks be"o*" ieir"r"ntJd inmind. Answers to this question
have covered the entire range of possibilities.
For example, the first general theoretical approach in
psycholbgy, known as struc-turali1m, attempted to specify the
irreducible elements of mind, the psychological build-ing blocks
from which all aspects of conscious experience were constructed
(Titchener,1898). In contrast, where structuralism concentrated
almost exclusively on mental ex-perience, behaviorism, at least as
first outlined by Watson, denied that anything usefulto psychology
could ever be learned from the analysis of conscious thoughl: "T-he
timeseems to have come when psychology must discard all reference
to consciousness; whenit need no longer delude itself into thinking
that it is making mental states the object ofobservation" (watson,
1913, p. 16l). For his part, Freud argued that uncon"sciousdesires
and fears, not conscious thought, were the principal motivating
forces behindbehavior. Not only did he believe this true in cases
bt ctinicat disturbaice, but also ineveryday life (Freud, [19011
1960). Finally, William James (1890) constructed a com-plete model
of the self by distinguishing the ongoing stream of consciousness
(the I)from the material and psychological objects of reflection
(the Me). As suggested byrecent exponential growth in what has
become known as "cognitive science,; modernpsychological theory has
rejected Watson's extreme behavioriim in favor of a
metathe-oretical stance that gives consciousness a fundamental role
in human action. Thus apsychological approach to new venture
creation must involve cognitive processes thatoccur within the
individual person.
This point is convincingly made by considering a behavior with a
long history in theanalysis of new venture creation-the assumption
of risk. Whether onJ reads tutitt otasks one's local venture
capitalist, one learns that the founding of a new
businessorganization involves significant risk. Yet reviews of
research have often failed to findri_s\laking propensity associated
with success in new venture creation (e.g., Brockhaus,1980, 1987).
Putting aside, for the moment, the possibility of flawed
measurementprocedures, there is an explanation for this
inconsistency in terms of the cognitiveprocesses involved.
Objectively, a certain proportion of new businesses fail, and in
apurely statistical sense, the failure rate for a particular type
of new business is a plausibleindex of the riskiness of ventures of
that sort. Now the personological appioach tobusiness entry has
assumed that everyone agrees on the level of risk involved, and
thatsome people (entrepreneurs) are constitutionally more willing
to take the chance than areother people (non-entrepreneurs). A
variety of well-known cognitive processes suggestsan important
alternative explanation: Perhaps those who found businisses do
nolittrtabout the risks in these statistical terms. For example, a
study of high-technologyentrepreneurs found that 213 of the sample
did not perceive themselves to be doinesomething risky (Corman,
Perles, & Vancini, 1988). One must have fear to
demonstraticourage. Rather than continuing the vain search for
personological risk-assumption, weshould begin to study the
cognitive processes involved in risk assessment. In ihort, weneed
to know how the business world is represented in the cognitions of
people who do,and people who do not, found new ventures. We shall
return to the isJue of risk-assumption later.
The Crucial Role of ChoiceBeyond the question of how the
external world becomes represented in the person
lies another of psychology's traditional disciplinary concerns.
this is the conciptural
26 ENTREPRENEURSHIP THEOI?Y ond PMCTICE
-
quagmire of determinism versus free will. Psy-chology has
Descartes (1952 [1641]) tothank for tlling the problem. In
cartesian dualism, iii. u.ov *"s a material substance,constrained
by the laws of physics, but the will was unextended and
incolporeal, utterlyfree of-physical limits. suCn i view leads to a
position of ..indeterminisp,,-1[s s1s1-cise of will cannot
be.predicted any more thin it can be controlled. Accepting
thisposition would put psychologists, and others *tro r"*"h foi
relularities in behavior, outof business. Fortunately foi
psychorogy, dualism founders o"n it, fuilu." ; ;p;;if; "plausible
mechanism for inteiaction between-the mind (will) and the body.The
extreme philosophical alternative to dualism is ttre maierialist
identity theory, amonism that equates.-l1o with brain (churchrand,
19gg). tn-principte (if not in facr,because of technological
limitations), ali mental events a.e, si.,lttuneously,
neurochem-ical e.vents in !h9 physical brain. Not surprisingly,
ia"niity trreory takes the form of---
lard" determinism. Stated in the extrem", ir ai-"ntd u"tiuity is
actually n"u-.h"*-ical activity in the brain, then that
neurochemical activity .urt-tiuu" had some precedingcause, and that
cause must have had some preceding ciuse, until ultimat"ry tf,.
."lr.r-sion reaches the joining of sperm and egg to form
,n&-u.y""i" human, and that joiningis the result of causes
external to the eir--bryo. In other *oidr, behavior is best
regardedas determined by forces outside the particular individual.
tn'psycrrology, the hard de-terminist position has been argued most
forcefuily by stinnJ.ie.g .,1blD, although
f'reu-d's biological determinism is no less antagonistic than
Skinner,s to a notion offreedom of choice.The debate between
advocates of free will and advocates of determinism is of morethan
academic inreresr.to. psychology. Some forms of p.y;i;;;rapy assume
that clientscan choose to change their actions. Fsychologists
ofr#eipe.t i"riiritt"v uuout t",nporu.yIapses in volition among
criminal defendantslPsychologiJts stuJy attitude change,
socialinfluence processes, and sources of resistance-to corifo.-ity
iressures. None of thiswould be possible unless one were to assume
that under nor-ut, everyday, .onJition.people can make choices. on
the other hand, determinists *gu" that unconstrained
choice all but eliminares the predictablity of behavior.
n"Jiiir,.'d;;.;;iii,;'i,assumed away, what business- do
psychologists have in searching roi regutaritie's inbehavior?
Fortunatel:r, psychologiris hau" been able to settle flr a position
known as.
- -,rft determinism. " This position follows the activity tnrory
oi'"ausality first specifiedby Thomas Reid (1863). The theory holds
that .uu*tity iti rru1nun behavior is funda-mentally different from
causality in physicar systems (which *. p.".u-"d to be com-pletely
deterministic)' specifically, iteid asserted that rhe notion of
determinism wasincompatible with the universaltruman experience of
deliberation, choice, and with allsystems of morality. Moral
judgments make sense only if people are tirought to ilecapable of
making choices for which they can later be h;ld d account.
capriciousnessand unpredicrability are.avoided
-br, qtr" presumption that menral events, lit" p-rry.i."revents,
are characterized by regularities that science can discover. Thus
the soft deter-minism-position agrees with indeterminism that
mental events need not have externalphysical causes, but agrees
with hard determinism that all behavior is caused by
,;;;-thing----either an internal mental event or an external
physical event. Needless to ,uy, unypsychological theories that
involve cognitive maniputation-oi-inro.-ation must alsosubscribe to
a soft determinist position on the issue oi free will. And this
includes nearlyall psychological theories thai attempt ro account
for human fi;;;: e;;;;fi;,choice is the third feature of our
approach to the creation of new ventures.A comprehensive
p.sychological portrait of new venture creation will ultimately
haveto show how the individual's cognitive representations of the
world get translated intoaction' To.accomplish this purpose, it
wili ultimately be necessary to consider generalorienting
dispositions (such as attitudes), motivationar pii*tpr"r (such as
subjectiveWinter,199'l
-
expected utility), and personal motives (such as achievement
motivation). It should benoted that choice for a psychologist is
not always the same as the rational decisionmaking inherent in
economic.theory. Important psychological models of social
behavior,such as exchange theory (Kelley & Thibaut, tg1u, do
aisume that people,s actions aredirected by a desire to maximize
their own outcomes. yet, there are significant excep-tions, perhaps
the most familiar of which is the idea ttrut g.ouf processes can
obscure thefacts and produce faulty decisions (Janis, rg72). A
th;;G;'p'sychological "pf.o*r, tothe study of choices involved in
new venture creation must-incorporate both the rationaland the
irrational features of decision makine.In the pages that follow, we
describe h6w consideration of person, process, andchoice can
illuminate the study of business founding. In;Jsection, we describe
oneexample in detail, and then briefly mention others ihat should
atso be considered. Noattempt is made to be.exhaustive; our
objective is only to present the possibilities ol apsychological
approach to new venture creation.
THE PERSON, NOT THE PERSONALITY
-
As noted above, psychology's focus on the person as the unit of
analysis should notbe construed as equivalent to iiearch for
personblogical variables that, regardless of thesituation, produce
behaviors that lead to tire creation of new ventures. Ear-ly in
psyctrot-ogy's history as an experimental science, woodworth (1938)
described the rela'tionshipbetween the environment and the person
by noting thaf the environment was the sourceof stimuli impinging
on the organiim, wtriitr then f,roduc"o u *"aru.able response.
Thisstimulus-organism-response (S-o-R) view constrasted with
Watsonian behaviorism,sclaim that the o contained nothing of
theo^retical importance. Most modern pt.h;i;-gists subscribe to
some version ol the S-o-R modei, but all iecognize that onty
ttrestimulus and the response can be observed directly. Any and all
of ttt" *g"nir-i"variables (the various o states) must be
inferredfrom the reiationships observedi"t*""nclasses of stimuli
and classes of responsei. what this means, u-ong other things, is
thatno particular o-explanation is any more "r:11_, than any other.
Some psychologists will,therefore' infer enduring personality
variables from observed relaiionships"uet*e"ns.timuli and
responses, whereas othei psychologists will instead infer
cogniiive, attitu-dinal, or motivational variables.
Having-learned-sometimes the hard wiy-ttre timitsof inference in
their graduate training, most psychologists regard with a healthy
skep-ticism any claims that a particular S-R conneition haJbeen
riediated """tu.iGy-L-fusingle-sort organismic variable.
Risk-taking PropensityThis skepticism is first directed at
methodology, and that is precisely the reason thatpsychologists are
wary of research claiming to iirow that one personility variable
oranother is characteristic of people who found new ventures. A
specific example illus-trates the point.On the reasonable
assumption that the creation of new ventures involves
risk,students. of entrepreneurship began to investigate whether
founders where higher-inpropensity toward risk than were people who
did not start their own businesses."Unfor-
SnatefY' the vast majority of this research (reviewed by
Brockhaus, 19g0, and later byBrockhaus & Horwitz, 19g6) has
employed ,o*. u".iion or tne ror""'*Jw"iil.i(1964) Choice Dilemmas
Questionnai.e (coQ) as the measu." oirirt-tit ing p.op";;itt.The
measure was originally developed by Stoner (1961) in a master's
thesis at the MIT28 ENTREPRENEURSHIP THEORT ond P|?ACTICE
-
school of Industrial Management, and having.been appropriated by
psychologists, themeasure has now found-its way back into buiinesr
."."*.tr. rne boe consiits or-tzshort vignettes in each of whictr a
protagonist is attempting to decide betieen u ,ur", uu,dull,
alternative and a m9r9 intereiting, but risky atternative. [n much
of the research onnew venture creation' subjects have read these
vigneftes only once, and have indicatedwhat probability of success
(of the risky alternativi) they *outo need to have in order
roencourage the protagonist to take the chance. The average of the
required probabilitiesof success is then interpreted as indicating
the risk-takitr"g p.tf.nrity of the subject.
-Following this tack, however, is very likely to invoiue ai
least three fundamentalmethodological problems. First, the
judgmentr ihut the subjects are, in fact, directed toma\e fo1
a.nother person (the protagonisi), are being inconLctiy
interpreted'as identical' the choices those same subjectJ would
make foi themselves. Now there are timesF?nt in psychological
research, the dependent variable is so heavily loaded with
socialdesirability-that subjects are asked "whai most people would
do," in the expectation thatanswers will reveal the subjects' own
tendencies.-But such reiearch typically containsindependent
variables that have been manipulated by the .*p".i*"nt.., and the
onlyclaims made concem differences between experimental conditions,
not the degree towhich one self-selected group of people (buiiness
founderg, as opposed to anotherself-selected group (managers),
poisesies the tendency rn queition.
The second problem is that in risk-taking propensiiy resiarch,
the cDe is being putto a use for which it was not designed. es
emptoyed by Kogan and wallaci (l96aiindby legions of
other.psychologisti, the CDQ was ttre vehi"cle to measure changes
rnexpressed levels of riskiness that occurred as a consequence of
group discussion."In thel-arse1 society at the time, the peaceful
civil rights demonstraionj were occasionallydeteriorating into
group.violence. As part of pJychology's attempt to understand
thisproblem, Kogan and Wallach wondered whethei
dinusloir'orresponsibility among mem-F* :{ a smallgroup led the
group to engage in hostile and agglessive behavior that
noindividual would have undertikenon his-oiher own. To "*uiTn" the
effects of discus-sion, researchers interested in what was then
called the "risky shift" would first haveindividuals respond to the
CDQ in private, then would have the group discuss each itemt-
ronsensus, and finally would have individuals rate the items-again.
The CDe neverhtderwent the sort of reliability and validity testing
that would 5e required foi a real"personality
measure," because the dependent variable of interest wis the
change inriskiness expressed from before to after discussion. In
short, the psychometric propJrtiesof the CDQ are unknown (although
in the normal course of evenis it was discovered thatsome of the
items produced conservative, rather than risky, shifts). To use
such ameasure as the index of a relatively enduring personality
trait is a serious methodoloeicaleror.
Third, recent reearch on what is now known as the "choice shift"
(reflecting the factthat on some items the actual shift was in the
conservative direction) makes a very strongtheoretical case against
using the cDe as a measure of risk-taking propensity. s*",evidence.
ever.supported the notion that diffusion of responsiUifity wai the
.uu." ofchanges in riskiness from before to after discussion.
Indeed, a number of studies showedthat face-to-face discussion was
not necessary, provided t'hat participants were showntranscripts of
what other people had said. Theie findings led investigaio.s to
concentrateon the information that was normally exchanged duing a
typical discussion of eachvignette. For each of the twelve items,
subjects think up
"a u#"ty of arguments in favor
of the risky alternative, and a number of arguments agiinst that
same alternative. Eachargument can then be rated for
persuasiveness, and theie ratings can be used to calculatethe
relative persuasivengls of the argument pools in favor of,"and
opposed to, the riskyalternative. Items for which the perzuasiu"nis
of the favoraure pooi was greater turn"i
29
----
Winter.1991
-
out to be those on which there was normally a risky shift. Items
for which the opposingpool was more persuasive turned out to be
those on which there was normally a con-servative shift. Moreover,
Burnstein and Vinokur (1977) showed that whether an indi-vidual
participant would shift in the risky or conservative direction
depended on therelationship between that person's own argument pool
and the group pool discoveredthrough discussion or some other
exchange of arguments. Thus, at the individual level,it is the
person's information, not his or her personality characteristics,
that determinesthe direction of shift. One needs no better reason
than this to discard the CDQ as aputative measure of a personality
trait.
Locus of ControlSeveral other personological variables are
frequently enumerated as part of the
"personal" characteristics of the creator of a new venture. For
example, in a recentanalysis of the entrepreneurial process,
Bygrave (1989a) presents an embellishment ofMoore's (1986) model
that includes need for achievement, internal locus of control,
andtolerance for ambiguity, in addition to risk-taking propensity.
Only two of these vari-ables-locus of control and achievement
motivation-have received wide attention inthe entrepreneurship
literature. In contrast to the case of risk-taking propensity, both
ofthese variables have been assessed with instruments designed to
measure individualdifferences.
To investigate expressed desire for control, many investigators
have used the Rotter(1966) lnternal-External Locus of Control scale
(e.g., Ahmed, 1985; Cromie & Johns,1983; Timmons, 1978;
Venkatapathy, 1987). Doing so, however, is at least a
tacticalmistake. As Rotter (1990, p. 491) himself has recently
stated,
In studying locus of control, because we were dealing with a
broad construct in-tended to study behavior in a variety of
situations, we wanted to sample manydifferent situations without
making the total score more dependent on one kind ofsituation (such
as school achievement) than on another (such as political
involve-ment).
Given this objective, it is not surprising that factor analyses
of the I-E scale haveroutinely revealed multiple factors, typically
distinguishing beliefs about personal con-trol from beliefs about
political systems or interpersonal relationships (e.g.,
Collins,1974;Levenson, 1981). Moreover, as Strickland (1989, p.4)
notes, elements of Rotter'ssocial learning theory "would demand
that reinforcement value and the intricacies of thesituation in
which behavior is occurring be taken into account for the most
preciseprediction. " Thus the global scale may not make valid
predictions in the specific settingof new venture creation; a
domain-limited scale such as that proposed by Paulhus (1983)would
be a preferable alternative. In fairness it should be said that the
I-E scale hassuccessfully predicted behavior in a variety of
interpersonal and health-related settings,and appears related to
perceived control, optimism, and physical well-being (see
Strick-land's 1989 review). On the other hand, past success, as
they say on "Wall StreetWeek," is no guarantee of future
performance.
Achievement MotivationOf all the personological measures
presumed to be associated with the creation of
new ventures, need for achievement has the longest history.
Indeed, McClelland's
30 ENTREPRENEURSHIP THEOI?V ond PI?ACTICE
-
(1961) initial assertion that need for achievement (nAch) is rhe
psychological moderatorbetween Protestantism^and- economic growth.
can tegriimar"iy' u" regarded as havingoriginated the search for
the "p"..onility characteistics of the ,u.-..rrrurl"i;;;;.-neur'"
Remarkably, although thit quest is now thought quixotic,
achievement motiva-tion remains perhaps the oniy.personological
variable"whor" urro"lution with new ven-ture creation appears
convincing.
Achievement motivaJion was originaily defined in.Murray's (r93g)
system of per_sonality' In philosophical terms, Munay's was.an
identity trr"*y of personality, holdingthat personality "may.be
defined as the-governing organ, o. iup'e.oroinate institution,
ofthe body. As such, it is isolared in the biain. N9irai"n,
n"p".Jonurity,, (Munay, r951,p.267). consequently, aneed was a
force "in.the u*in."gion;;(Murray, 193g, p. 123)that organized
perception' intellectual activity, uno u.fion.t rn. specific need
for-_.rievement (nAch) was defined as
To accomplish something difficult. To master, manipulate, or
organize physicalobjects, human beings, or ideas. To do this as
."pioiv, *o as in"depeno'e;irt;,possible' To overcome obstacles and
attain a high ttunJu.o. ro excel one,s self. Toriv-al and surpass
others. To increase self-regariby ;h. ;;;..tsful exercise of
talent(Munay, 1938, p. 164).
Depsite its focus on the whole individual (Murray introduced the
term..personologi-cal"), Murray's theory retained the psychodynu.i"
r..g., ei.*oian) assumption thatunconscious processes guide
behavioi. For our purposeJ, this assumption leads to
themethodological expectation that the assessment oi ne^"ds must
involve projective, as wellas objective, tests. His projective
Thematic^ Apperception teit (TAT) quickly'becamesecond only to the
Rorschach in breadtn orapptlcation, uJ*as the originuirn.uiu..used
in the assessment of achievement motivation t..g., Ly McClelland,
Atkinson,Clark, & Lowell, 1953).
. .Some 25 years later, Fineman (1977) would be able to describe
22 measures ofachievement motivation, only a few of which were
projective. More recently, Johnson('^90) reviewed eight measures
that have been invotvea in examining ttr",liuiio;rhlpt''aveen
achievement needs and new venture creation. Two of these-the TAT
and theMiner Sentence Completion Scale (MSCS Form T; Miner,
lgglFare projective mea_sures' Two are subscales of larger
personality inventories-the Edwards personal pref-erence Schedule
(EPPS; Edwards, i959) andthe personality n"."u..t Form_E
(pRF_E;Jackson, 1974). The remaining four-the Lynn Achievement
Motivation questionnaireILAMQ; L1nn, 1969), the Melirabian
Achievement Scale (MAS;Mehrabian, l96g), theSentence completion.fT,
rscr' Mukjerhee, 196g), and the work and Family orien-tation
lnventory (woFo; spence & i{elmreicrt, iqs:Fui. specific to
achievemenrmotivation, aqd_freouSltrv wepjesigned to overcome what
werl regarded as probremsin the original TAT measure.a Fineian's
(1977) review examineo 7g cases in which3' Given the tendency to
consider achievement motivation as a personological variable
independent of:J,T;11"1':lT:1i::::;'j::'j,Tj3r*l1T :qi:l,l
i'npo'tuni.r"'"ffi'Mu,,uv', theory were press(plural and singular
are the same form). These were attributes or p-p.ni.r
ot.-i.-"iiui..,, "1 ;Hii;:iacted to impede or facilitate the
goal-directed behavior of the individual. lVloreover, Murray
distinguishedi!frn{:;i:
(the objects as thev reallv are) ftom beta press (the objects as
they are p.i..iu"J'ry ir,"4' The most pervasive problem was the
failure of rAT-like measures to predict achievement behavioramong
females ' Given the dated, highly sex-role stereotypediature of the
pictlures actually used in the TAT,this is understandable. Indeed,
futuie users of the original ref pi.t."r"""gi, r" * required by
editorialboards to insist on a gender-based limit for
generalizaiion ot..r"'u..tr inu"r"i"g tt" original TAT
pictures.
Wnter.'t99,| ..,1
-
correlations among measures had been computed, and in only 28Vo
of these instanceswere the correlations positive and significani.
Johnson's (t99b) review, specific to theentrepreneurship
literature, added no studies in which there had been crois-scale
com-parisons, but did find a relationship between achievement
motivation, however mea-sured, and entrepreneurship, however
defined, in 20 of 23 studies. The reasonableconclusions Johnson
draws from this are that thorough specification of the targef
g.oup,careful operationalization of achievement motivation, ind
detailed descripti"on
""f ti"
environmental context are needed before future research can make
substantial progress.From a psychological perspective, of course,
"careful operationalization" of achieve-ment motivation would
necessarily include attention to lssues of scale reliability
andvalidity-psychometric niceties all too often overlooked in the
past literature on n"*venture creation. Notwith,standing the
various conceptual and methodological issues yetto be resolved
(Frey, 1984), achievement motivation remains the perso-nologist's
blstcandidate in the attempt to account for new venture
creation.
Method and ContextMany of the lessons inherent in research on
the person variables associated with
entrepreneurship have been noted before (e.g., Gartner, tqgg;
Low & MacMillan, lggg;Sexton & Bowman, 1985; vanderwerf
& Brush, lggg), but three bear repeating here.First, don't send
a scale to do an inventory's job. If your goal is to ideniify th"e
rela-tionship between a particular personological variable ind niw
venture creation, use apsychometrically respectable instrument
designed for "trait" assessment rather than forsome other research
purpose. Second, don't confuse a single score with the
answer.Adhere to the principle of multiple operationism (campbJll
& Fiske, 1959), whichdemands that conceptual variables be
measured in more than one way. Such researchtactics are essential
not only to assess the relative performance of the multiple
measures,but also to discover whether the underlying variible is
unidimensional oi multidimen-$oryI. Third, don't ignore the
context. Is the instrument itself domain-specific (like thePaulhus,
1983, Spheres of Control scale's three subscales) or domain-general
(iike theoriginal Rotter, I-E Scale)? A mismatch between scale and
assessment environment willproduce confusion. Even if the scale and
the domain match, the assessment outcome islikely to be compromised
if the demand characteristics of the setting (Orne, 1962)
andpressures toward socially desirable responding are ignored.s For
both methodologicaland conceptual reasons, then, psychology's view
of new venture creation "on"entLt",on the person in his or her
situational context.
THE PROCESS OF SOCIAL COGNITIONNot even the most resolute
advocate for enduring "personality differences between
entrepreneurs and nonentrepreneurs" would argue that a complete
map of the humangenome will reveal a specific gene that can
separate new venture founderi from everyoneelse. Variations in
entrepreneurial activity within the individual over time,
betiveenindustries, and across political boundaries are too obvious
for a purely genetic expla-
5. Social psychologists routinely assume that research
participants are actively anempting to discover thetrue purposes of
the study, and the demand characteristics are the sum total ofthe
"u". uiuilubl" (whetheror not they might lead the subject in the
correct direction). Moreover, researchers need to be aware of
thesocial. desirability implications of questions they ask.
Finatly, any interchange between researcher andparticipant should
avoid evaluation apprehension, the subject's tendency to resfond in
a ..typical" fashionto avoid appearing "different" (Rosenberg,
1965).
ENTREPRENEURSHIP THEOTT/ ond p|IACT|CE
-
\nation to be prausibre. Environment is_ crucial. consequentry,
the psychorogical per_spective on new venture creation should
""".i4* rr.iutiiui'"*,"-ar environment be-comes represented in
experience. Because the creation of-a Lusiness venture is,
funda-mentally' a social.activity, we shall concentrate on
processes of social cognition. Theseinclude aspects of
attention,.rnarrtoty, "ategorization, and-inference (Fiske &
Tayror,l99l)' but the approach can be illusirated riith two
r"r."-i^"o"tent areas-hypothesistesting and attributions for
success und fuilur".
Cognitive HeuristicsIn the process leading to creation ol-1.new
venture, ..opportunity recognition,,isoften described and treatJd
u, u f..y lnlredrent:
Entrepreneurs see ways to put resources and information together
in new combina_tions. They not.onry iee.thl system as.it is, u",
"r-ii'n'rlit be. They have a knackfor looking at_the usuar
and,."i'g tr,"il.d, ;;;:'ordinary and seeing theextraordinary'
consequently, they.in ,pot opportunities that turn the
commonplaceinto the unique and unexpelted (Mitton, 19g9, p.
l2).
Environmental circumstanc.gs do not parade past like so many
conventioneers, some ofwhose badges say "inreresting
prospecr,;'*i".Lur;;;;;y ..dud.,,Rather, oppor_tunity recognition,
like beaut-yj is in trre t;-;*L",i.i;d;i..HJ* ao"Jlil;
;5i.""process the incoming informaiion? To whai internar rr""o*J.
rs rr compared? No re-search on new venture creatio.n.directly
address;;;il;;;, our rt. social cognitionIiterature suggests some
possibilities.
For example, consider the question, "what is the likelihood that
Restaurant X willfail in its first year of operation?"
soptristicatea "";p;;.lilip ."r"*hers recosnizethat this is a
controversial question, u'uiit" definitional i.;;tili;;k" i;;;
*""rtlt"ryto be lost on "ordinary folki" outside the academi.
";;;;;6. so instead of answeringth'question with queitions ald
quatiRcaiions, th9s9 people a.e litety to give a numericale'-,nate.
But because people are nor good intuiti";
.;;iJ;;;ns, this judgment underuncertainty is apt to be
influenced by"at least three-
iiiiiir'i'n"uristics-availabitity,representativeness, and
anch_oring (Kahneman, slovic, & Tversky, 1gg2; Kahneman
&Tversky, 1973; Tversky & Kahiei""", f qzi, e p"rron
*nolur, read about anotherrestaurant's closing in the morning paper
wlr live a'trifi;.;;ri-"te of failures rhan willa person who has
not seen suchl^story in u rong iiti.-i,rt.'uuuilubility heuristic).
Aperson for whom Restaurant X is typicaf of successfur
"rtiri.n.ents will make a lowerguess about failure than will u
p"..- for whom tne nestauranlx ,"r"*ules failures
(therepresentativeness heuristic)-. Flnally, a perceiver *rro
mo*s-itru, ,n "" local restaurantshave failed will make a smaller
estimate than a perceiver who has been told that 10,000restaurants
have fa'ed nationaty (the anchoring heuristict. - *
why should researchers inteieiteo in new venture creation care
about such cognitiveheuristics? For one thing, they prouio" a
social "ogniti;-ultLatiue explanation forphenomena
frequentry_int"rp."i"a in ott., theoreticait".-r. er'one example,
scherer,Adams, carley, and wiebe (19g9) .tro*"d-that prese;;; "r""
enrepreneurial parentwas associated with increased expecrancy or
"nt"rinfu;
.;;;;*uriar career. Schereret al' made a strons.case that.the
p*"nt provided-a.ot" r6o.t whose activities andbehavior r"..
".ul"uted through " rlr.r'"rrodering.onriJ"nt *ith Bandura,s
(1977)social learning theory. But the-child.en oient epreneurial
parents have more than a role
Winter.199,t 33
-
model, they also have information that is unavailable to
children whose parents did notstart or purchase a firm. The way to
tell the two explanations apart, of course, is toexamine the
contribution to child's entrepreneurial interest of parent's
venture success.Unfortunately, the study was only able to measure
the child's perception of parentsuccess. So the social cognition
alternative cannot be ruled out. The larger conceptualpoint is that
the social cognition processes are no more-and ns ls5s-2n inference
fromthe data than is the notion of vicarious learning from an
effective role model. Each is anO-variable that accounts for the
observed relationship between the S of entrepreneurialparents and
the R of child's expectancy of an entrepreneurial career.o
Success, Failure, and PersistenceCognitive heuristics guide
judgments under uncertainty. When outcomes are known,
other processes of social cognition enter the picture. Before
proceeding, it is importantfor us to specify more carefully what we
do and do not mean by an "outcome." In arecent application of chaos
theory and catastrophe theory to entrepreneurship, Bygrave(1989b)
notes that organization founding is a discontinuous change, and
this view oforganization creation as the singular event to be
explained is inherent in other accountsthattake quite disparate
theoretical positions (e.g., Aldrich, 1990; Gartner, 1988). Weagree
that the actual founding of an organization is a discontinuous
change, but do notbelieve that it is the only "event" to be
explained. Nor is it likely that the founders ofnew business
organizations would agree that they have only done one thing.
Rather,founders would probably argue that they have done many
things in order to get to thefinal one.' Thus we believe that the
discontinuous creation of a new venture is precededby a series of
prior discontinuous events, each of which can be considered a
separate"outcome." Such outcomes might include identifying a market
niche, designing aproduct or service to fill an identified need,
and establishing the company. Note thatthese are only a few of the
"outcomes"; they are not intended to exhaust the possibil-ities.
Moreover, it is not necessary that the researcher be able to
specify all of theoutcomes in advance. Social psychologists would
be as comfortable with the founder'sown list of outcomes as they
are with people's natural cognitive categories (Rosch,1978) or
people's chunking of behavioral events into discrete sequences
(Newtson,1976). No matter how the sequence from initial idea to new
company is segmented, thesocial cognition approach argues that the
explanations potential founders offer for priorsegments will affect
the likelihood of the final discontinuous change.
Of particular interest are the attibutions potential founders
form to explain suc-cesses and failures in business-related
activity. Attribution processes were first describedby Heider
(1958), who noted that the essential elements of task performance
wereability, intention, effort, task difficulty, and luck. When an
action is intentional, itssuccessful performance depends on the
combination of the two internal elements (abilityand effort) with
the two external factors (task difficulty and luck). In each of
these pairs,
6. The same is true for any account that depends on the
assessment by an individual, or a group ofindividuals, of the
probability of an uncertain outcome. Thus, for example, when
population ecologistsargue that "potential founders may be
frightened by relatively high or increasing death rates"
(Aldrich,1990, p.9) they are making an inference about a
psychological moderator (fear) that psychologists wouldargue
follows in time the gathering of information. Parsimony alone would
suggest that purely informationalmoderators such as cognitive
heuristics should at least be ruled out before higher-order
moderators areinvoked.7. Whether we are correct in this supposition
is an empirical question. From a psychological perspective,of
course, what really matters is the definition provided by the
founders, themselves, rather than byentrepreneurship theorists
34 ENTREPRENEURSHIP THEORY ONd PMCTICE
-
!the first element is stable (a person's ability does not change
dramatically from one timeto the next, and the difficulty of a
given iask is a constait relardless or *ni.t p".ronattempts it).
The second element in each pair is variable 1a persJn's effort
varies, "'Jin.capriciousness of luck is notorious).
Building on thes_e distinctions, weiner, Russell, & Lerman
(r97g) developed anattributional model of persistence in
achievement tasks, and the model can be adapted tothe founding of
an organization. In a technological start-up, for example, before
therecan be a new organization, the founder-to-be riust at minimum
develop ano tesi pioio-typ:s, conduct appropriate market research,
create the standard financial p-p"ii-r,and construct a business
plan suitable for securing venture capital. Rarely is ealtr orthese
activities completed to the founder's satisfactlion on the first
pass. Consequently,such a founder will typically havg multiple.
failures to explain, andwill have to persistin the face of those
failures in order to reach the ultimate oui".iiu.. In weiner,
Russell,-
'd Lerman's terms, the potential causes of each of the failures
could be lack of ability,lack of effort, insurmountable difficulty
of the task, bad luck, or,".t, *r,* id;;;*,as the intentional
barriers erected by people who might not want to see the
venturesucceeo.
. 3" .u*gus potential causes can be described by three
theoretical dimensions:stability (stable or variable), locus of
causality (internal or "*,.*uiidunJ-i",."ri"""ri,y(whether or not
production of the effect involves the founder's intentions, the
intentionsof other people in a position to influence the outcome,
or some combination of the two) .These three dimensions are not
entirely independent, because intentionality on the partof the
founder would be necessary for there io be an attribution to the
variable-internalcause of "effort. " People simply do not
exertthemselves by accident. The model *ju.,that the positions of
the various elements on these three dimensions produce one of
twokinds of internal effects. By itself, the presumed cause's
location on the staUitity Jl-mension determines expectancy shifts.
If the failure is attributed to bad luck (external,variable), or to
insufficient internal motivation (internal, variable), the
perceiver canreassure himself or herself that in the future things
will be different. In contiast, the locusand intentionality
dimensions combine to produce the affective reactions oi pride
orshame' Specifically, slram9 results only if the outcome is
attributed to an intentional, andinternal, failing' Turning back to
the positive, observable, behavior of persistencei,
the-.nders-to-be can be expected to persist after failure only if
they believe that their stable
capabilities are equal to the task, that the variable elemlnts
of the equation (their owneffort and the contributions ofothers)
are subject to their intentional control, and that theemotional
consequences of repeated failed attempts are less negative than the
emotionalconsequences of declining to try.
This attributional model was originally derived as an
explanation for achievementmotivation, and it has been found to
predict achievement behavior in a wide varieiy ofsettings (much of
this research is reviewed in Weiner, 1986). Moreover, the
principlesinherent in the model have been generalized beyond
contemporary American culture tosocieties with dramatically
different economic_ind social .yit..., such as the people'sRepublic
of china (Stipek, weiner, & Li, l9g9). Thus the model can
easily be tians_planted to the study of new venture creation. A
first effort in this regard has recently beenreported by Gartner,
Gatewood, and Shaver (1991). Briefly, theJe investigators
askedpre-venture clients of a Small Business Development Centeito
explain wtryitrey wantedto enter business . In addition, each
participant was asked a compaiable . . why' ' questionabout two
kinds of purchases-the last automobile purchased, and the last
piece orsports
I .*:pi,: the similarity in the terms, artributional locus of
causalityindividual difference variable of "internal-external locus
of control. "
Winter.1991
rs conceptually distinct from the
-
or. fitness equipment purchased. The participants' free-response
answers were thensubjected to an attribrrtional coding
thaicategorized each uni*". according to whetherit was
internal/external, stable/uu.[b1., and ieflecteo u "uur" or a
reason (a version ofintentionality). Analysis of the results
showed, for example , that luvo of the businesschoices were
explained in internal terms, whereas r.*", itun'35vo ofthe choices
in theother two categories were explained in internal terms.
aritrougn more research in thisarea clearly needs to be done, these
preliminary findingt inoilur" that the attributionalmodel can
profitabry be appried to the study of n"* u"lnt* .."ution.
Representations of External Circumstancescognitive heuristics
and attributional principles are part of the psychological
accountof ways in which the external environment becomes
ryPreseil in the mind of a personwho might become the founder of
a new venture. For this reason they represent theinternal process
presumed to mediate between the stimuli (suctr as
..opportunities")presented by the external environment and lhe
responses fruch ur ..organization found-ing") that are critical
dependent variables in the study of new venture creation.
varia-tions in the process of social cognition can herp. "*pruin
*rry not ail peopre high inachievement motivation (or any othe.
personotogicai variable'reliably associated withnew venture
creation).actually found organizations. Additionally, theories or
,*i"rcognition must be considered c-ompetition-for uny otrr".
.^pru*tory principles that findtheir .way through the heads of
organization founderr. o" ttr"i. face they are no moreplausible, or
less plausible, than oiher o-variables such as sociat tearning,
role expec-tations, or prior organization deaths. The challeng" roi
rutu." ,"r"ar"t, on new venturecreation is to find convincing ways
to distinguish
"on" ru"r, "*planation from the others.Unfortunately, the
experimental iesigns that are a staple of psychological research
maynot always be appropriate in the study of new venture
creation.
when facing the task.of disentangring competing o-variables,
psychologists doexperiments that occasionally stand thJorilinal
p.opolition on it, t"iJl r..-";;*;,rather than searching for
personological difierenc""i";;;;-,-i"uy or risk-taking propen-sity,
shaver, williams, and Scott trqqol used a bogus test
oi.nLp..neurial abilities toconvince subjects that they either
were, or were not,-lik"ly candidates for the ro*oingof new
ventures. overall results showed several effects for this belief.
people whothought they had entrepreneurial characteristics
"*p."r*o!.*ter creativity and achieve-ment motivation' Moreover' on
items dealing *ittr uurln"rr .isk (but not on itemsdealing-with
personal risk), males who believeJthey *".".ni."p.eneurial selected
morerisky alternatives on some cDe items and on "o-p*uure i[ms
designed by scott(1e90).These studies .uI: ,*g important poinrs. At
the empiricar lever, they show that asocial cognition
variable-beliei in oneis entrepreneurialiendencies-affects
responsesthat may be related to success in new venture creation. At
the conceptual level, theybolster our concern about
the."personality characteristics'; apfroach. when scores onmeasures
that have been used in personological research canii manipulatedby
instruc-tions in.a brief experimental setiing, one"must wonder just
how ..enduring,, are thepersonality characteristics these measures
purport-to identify? we recogni ze thatexper-imental designs, which
uue common in psyinological re"search, may not arways beappropriate
in the study of new venture^cieation] on the otte. tano, ,o-"
{uuri-experimental designs (Campbell & Stanley, 1966; Cook
a-CarnpUef l, l91g)can serveas excellent substitutes. lt jhe very
reastj we would echo receni calrs (Gartner, 19g9;Low &
MacMillan, 1988) for proqpective research on organi"aiion
founding.
36 ENTREPRENEURSHIP THEORY ond pMCT|CE
-
TIIE EXERCISE OF CHOICEAt the risk of belaboring the obvious,
new ventures are not created by role models,support networks, or
the itate of the population ".o..ying capacity. New venruresemerge'
and take $e form they do, uecauie iriin" laiu"rit" "t oi"". made by
individualpeople' Not surprisingry,
rhe,rr"a'airii"e"pr,i"fr""t;;';i".p.ychorogicar
approach tonew venture creation is its focus on thE
i--"-oiut.*li,J..o.r,o of choice, From theperspective of the
potential founOer, two questions u*..11i"uf, ,,Can I make a
differ_ence?" and "Do I want to?': The ilrior tt"r" in;r* o"rl"i".o
contror, whereas thesecond requires consideratio" "a;;r;;"tional
processes.
-,.erceived ControlIn Mitton's q.9s2) view, "Entrepreneurs
prefer to take and hold unmistakablecommand" (p. l3). They know
rt.y'.rn make a differeie, and they berieve they,personaty, can
exerr.o,nqgr ou.. pJopt. u"o ""*tr.^iioi'i"u.n the mosr
,.compreat,,entrepreneur, however, berieves rdi;; o1 she .un
.on,.ll lverything. perceived con_trol-unlike the generaliz"o
.*p".tuniy ro, rnt.-;iiffi Jf .ont.or, or the need
forffilfiiifffl::" tn' rvpe e'L!r'""1"' pattern (Friedman &
Rosenman, le74F-isMuch of the recent psychological writing on
issues of contror among adults hasconcentrated on the behaviorat
efficts of losiig ioir"i"a.""g orher things, this ma_terial
inctudes psyclrolog,"ur i.*-t^u-*;,i:"_"d";t;il;ni;l.uno tr,"
effecrs of controlon intrinsic motivation' rtyttotogi.uireactance
i, ir,. .notiuutional state aroused whenexisting freedoms are
rest;cteo oi ttrieate",:d_(B:.ilr'iiio, g*tm & Brehm,
lggl).what makes ir interesting i. ;il;;';;" is produced uy the
demand rhat ,,you must,,do somethins vou *ouri tuu".to..n';";" '";
; ; ; ;#:The key issue is ross offreedom to rfioose,^n".lg_"
id;ffiq::
_: something distastefur. where the ross ofcontrol persists over
time, learned help'lessness- may ue- tt"-."rutt (seligman,
1975).Rather rhan continuing 1o. figh;; ;; ;;;:';" simprv
sir";6:lccording to the reformu_' 'd version of rearn;d
h"fit.;,;.ia9o,
,i,J a?r",".iJ"s psycrrorogical effects areg*atest when fa'ures
are attributed to i1i;-;i,;;il;i,.i"s, and are expecred
rogenerarize to other situations reu*.o"n, Setigman,-a1.iro"r",
lgTg). wtrere terp_lessness is often a consequence of inaccurate
at;ibutions ro-thJself, intrinsic motivationcan be reduced bv
correct attributions ol fontror ,o .*r.,nuirorles (Deci & Ryan,
r9g0).specifically' "utn rewards ro..u"..rrruiturt p".ro.rnu".i .r"
t""e debilitating effectson intrinsic motivation uno
r"rra.[..]nur.n, ir trrai-reeai;i ,, seen as controring,rather than
compe"i::.{il;ti;;;i} Ryan, tt*i. ilil risk of oversimplifyingeach
of these perspectives, the aiswer to tn. qu"rtion'.iCu" irnu*e a
difference?,,willbe affirmative-onrv if the pe;;;;;;;;r;"., the
choice t* ". r,"r. to make, b) has someinitial success thai can-be
aftriilJ'i;';;ry, permitting (c) intrinsic interest in theproject
to be maintan"a' onlt ufu* iiir',rr. way in whicfr the potential
founder rhinksabout reality, not the externar
..irity ru.tr, ,nu, determines the ourcome.
Outcomes and ExpectationsBeing abre to make a difference is not
the same as wanting to do so. The foundersof new ventures give a
variety of ."ur*i'*f,." urf,J*fi,'ffi estaUtished their
busi_nesses. Many of these have i.""nrr/!."" ;;;;o"r"rlli?r88), and
incrude (in
Winter.1991
-
descending order) desire for autonomy, interest in personal
achievement, dissatisfactionwith current job, desire to make money,
and unhappiness in current career. In both thepsychological
literature and the organizational literature, such motivations have
typicallybeen described in terms of some form of expectancy theory
(e.g., Atkinson,'1964;Feather, 1982; Fishbein & Ajzen, 1975;
Vroom , 1964). These theories share a commonform. Each predicts
that some motivational outcome will be determined by the
summedproducts of individual evaluative elements and individual
subjective probabilities thatthe associated evaluative elements
will, in fact, be achieved by the aition being con-sidered.
There are, however, two difficulties associated with applying
expectancy theory tonew venture creation. First, there is the
question of how particular eiements-many ofwhich are actually
negative in the organizational context-should be scaled. For
exam-ple, take "job dissatisfaction" from Cromie's (1988) review of
reasons for startins anew business. How might such an evaluative
element be turned into the action ofcreating a new venture?
Specifically, should it be counted as a negative (on the
grounds,for example, that it represents a failure of the link
between performance and ourcomes)?Or should it be counted as a
positive (on the grounds thai it acts as a stimulus to trysomething
else)? If it is the latter, what reason does the individual have to
believe thatfailure of a performance-outcome link in one context
means that there will be success ofthe same link in another
context?
Portions of psychological exchange theory (Thibaut & Kelley,
1959) suggest analternative way to think of things like
dissatisfaction with one's current job. Beginningwith the
assumption that individuals seek to maximize their own outcomes
-in any
relationship, Thibaut and Kelley identify two standards against
which such outcomesshould be judged. The first, called simply the
comparison level (CL) is the average ofpast outcomes from similar
circumstances. To the extent that one's present job is belowthe
average set by one's past experience, job dissatisfaction will be
the result. Thesecond standard, called the comparison level for
alternatives (CLu,,) is the best currentlyavailable alternative to
the present circumstances. A person whose current
occupation;loutcomes exceed both CL and CLur, will be content, one
whose outcomes are below CL,but above cl-u,,, will be unhappy, but
dependent on the current job. The differencebetween people who
consider job dissatisfaction as a fact of life, and those who use
itas a reason to go out on their own may be as simple as whether
"creating my ownbusiness" is one of the viable alternatives
considered.
The second difficulty with applying exchange theory to new
venture creation is morefundamental. To see the problem, assume
that all elements can be scaled, and thatdespite the uncertainties
inherent in new venture creation, the subjective probabilitiescan
also be specified. Expectancy formulations then predict the
resulting choice. Butthey do so by treating a highly desirable, but
low probability outcome as equivalent toa much less desirable, but
highly probable outcome. In a recent discussion of riskvchoice,
however, Lopes (1987) makes a persuasive case against precisely
this assump-t ion.
Traditional study of risky choice in experimental psychology has
presented subjectswith rrva-alternative "gambles," that frequently
take the form "would you preie. a1007o chance of winning $3,000, or
an Sovo chance of winning $4,000?" Although thesecond choice has a
higher expected utility ($3,200;, most subjects choose the
firstchoice, suggesting that people are "risk-averse" (Kahneman
& Tversky, 1979). suchfindings are exceedingly difficult to
generalize to the creation ofa new venture, becausethe real world
rarely presents just two choices. In a program of research designed
toovercome this limitation, Lopes has presented subjects with six
alternative "lotteries"with approximately equivalent expected
values. After the participant chooses which
38 ENTREPRENEURSHIP THEORT ond PMCTICE
-
lottery to play, a random device indicates the ..winnings.,, One
of the loftenes isriskless, arways producing rorn" puyoii, *hereas
,h;';'rh*ile conrain the possib.ity ofazero payoff' The lotteriet
ut iho*n to^subjects u. pi.tu..r, with dollar amounts ofwinnings on
the ordinate una nu,nu".. or ",i'"r."1:"ut#iluur" for each dolrar
amounrshown on the abscissa. In addition io ,r,".ri.tt".,
rru.iy,-in"re is a ..short shot,,, a"peaked" distribution, u
.".tunguiu. Oi.r.tUutio;;;'ffi;j"t aistrtUution, and u ;;longshot"
which has mostly ,".o oiiotu-payoff our.o-o.-iilJ'p."f"."n.es of
risk_aversesubjects follow this ordlr in Oescenlirig fashion.From
analvsilgl ugtrr trt. .*pi.*"i p.eferences and free-response
expranations forchoices, Lopei (1987) h;;;;.r;"p;it*o-ru"to.
th;t";;i.ky choice. The first facroris a personal preference tor
r.cuity uolu, potential. Mathematically, security seekersweigh the
worst outcomes more ireavily, ;h;;;; o"*"i", seekers weigh the
bestoutcomes more heavily' In a departure from traditionui
"*p.fiunc.y theory. Lopes arguestl" hese weights include uotrr
ir,unger'in tt. .rii.""ipi"iiuilities and changes in thevartles of
the ourcomes to which thiie probabiri;;;;';";i"'.,r.6. The
secondfacror isaspiration level' and is defined bt 6; iituation-
rn. r"iii"g provides informarion onwhat might be a reasonabre
expect"til"; ,t. ,|".#; u"rr.rn"uriu" being considered iscouched in
a conte-xt of other .nti..r uuuilable;'anJl".g., ,rr","gic goars
can alter thetactics employed' Preference r- ,".uriiy versus
potential is considered a personologicalvariable' whereas
aspiration r.u"r ir rii'.rational; the rinur .i.r.y choice is
thought io bedetermined by the interaction u"l*""r.in. two- once
again, behavior relatei' to newventure creation is produced by the
interaction u.r*""nloron uno environment.
Expectancy and UncertaintvIt may be no accident
3u,,.i!,:.":cy theories have been heavily appried within
theorganization riterature (e.g., Loike & Latham, l9g0),;;t
iess u.oaoly used in rheIiterature on new venture creation. The
array of possibie et.*"nt. that could be valuedis more likely to be
specifiable in an ffiiratronri "";;;;irh; ir is in an
entrepreneuriarventure' After all' new venture creatio'-n is
nothing ir not innovation, taking the unfore-seen with the
foreseeable. Moreover, even if ttie elemeni, "ouro be specified,
theirvalu's might not be.,tlose ex.pecled !r-rh.."r"u..t".r-
nnoiinauy, compared to thereli'e certaintv of the organizational
climate, the entreprenzurial climate provides fewclues to the
probability e#maier iir"i ririrro ue attactre'J," *"t individual
evaruativeelement. In the examination "r .n"i".,'", wet as in the
,iuov or process, the psycho_logical perspective incorporates both
personal and situational variables.CONCLUSIONS
The study of new v:nqrr: creation began with some reasonable
assumptions aboutthe psychologicar characteristics or
"*t."p."neurs.,, Through the years, more and moreof these
personolosical characte.irti"r"r,ui. been discarJ"f,^j.u""ted, or
at the very.,?i,T;"llxlt;"jlx u"'n '"u'u."al"ii".ru"rv. ir,"
i"r,irii^ been a tendencv top,ortanr ;,o" i ur n"
t*o',i.3llffi;,Ifi ij JH#[T:|#,!;:* : _:irc u m s tan c e 1 are i m
-important; finance is importanti;;;;;;;. asency assisrance
,,lH?:XXlllTilT:'l"tj;these will, alone, create a new ventu.". Fo.
that we need a person, in whose mind all ofthe possibilities come
together, who believes that innovation is possible, and who has
themotivation to persist until the job Jd*;. i.rron, process, and
choice: for these we needa truly psychologicar perspeciiv. on
n"*l"nture creation.
Wnter,199139
\
-
REFERENCES
Abramsor, L. y., serigman, M. E. p., & Teasdarg,J D (197g).
Leamed herpressness in humans: cri_tique and reformulation. Journal
of Abnormal psychotogy, 57, 49_74.Ahmed, s' u' (1985)'
"4.1:l.I^-!*ing propensity, locus of conrol, and entrepreneurship.
personality andIndividual Dffirences, 6, 7Bl -7 82.Aldrich' H'
(1989). From taits to rates: An ecological perspective on
organizationalfoundings. paprpresented at the Gateway Conference on
Entrepreneurship, St. Louis, MO.'Aldrich, H' E' (1990). using
an.ggolosical perspective to study organizational founding rates.
Entrepre-neurship Theory and practice, t4(3),7-24.Atkinson, J. w.
(1964). An introduction to motivation. princeton: van
Nostrand.Bandura, A. (1977). sociar rearning theory. Englewood
Cliffs, NJ: prentice Hall.Brehrn, J. w. (1966). A theory of
psychological reactance. New york: Academic press.Prehm' S' S''
& Brehm, J. w. (1981). Psychological reactance: A rheory of
freedom and control. NewYork: Academic hess.
Brockhaus, R' H., Sr. (1980). Risk taking propensity of
entrepreneurs. Academy of Management Journal,23 ,509-520.
Brockhaus, R. H., Sr. (1987). Entrepreneurial folklore. Journal
of Smatl Business Management,25(July),l -6 .
Brockhaus, R. H., Sr., & Horwitz, p. s. (1996). The
psychology of the entrepreneur. In D. L. Sexton &R' W. Smilor
(Eds.),The art and science of entrepreieirship,-fp.25-48.
Cambridge, MA: s.lli"g...Burnstein, E', & vinokur, A. (1977).
Persuasive argumentation and social comparison as determinants
ofattitude polarization. Journal of Experimental sociai psychology,
13, 315-332.Bygrave, W. D. (1989a). The entrepreneurship paradigm
(l): A philosophical look at its research method-ologies.
Entrepreneurship Theory and practici, I4(l), j-26.
Bygrave, w' D. (1989b).-.The entrepreneurship paradigm (II):
Chaos and catastrophes among quantumjumps. Entrepreneurship Theory
and practice, I4(2), j_30.
campbell, D' T., & Fiske' D. w. (1959). Convergent and
discriminant validation by the multitrairmuftimethod matix.
Psychological Bulletin, j6, gl-105.
Campbell, D' T', & Stanley, J. C.' (1966). Experimental and
quasi-experimental designs for research.New York:
Houghton-Mifflin.
carland, J. w., Hoy, F., & carland, J. A. c. (l9gg). "who is
an entrepreneur?,, is a question wortnasking. American Journal of
Smalt Business,,l3(Spring), 33_39.Carsrud, A' L' ' &
Johnson,-R. w. ( 1989). Entrepreneurship: A social psychological
perspective. Economic& Regional Development, I, 2l-31 .
Carsrud, A' L., olm, K. w., & Eddy, G. G. (1986).
Entrepreneurship: Research in quest of a paradigm.In D' L. Sexton
& R. W. Smilor (Eds.), Ihe art and science of entreprineurship,
pp. Xl-ttt.CumUriOi",MA: Ballinger.
cartwright' D. (Ed.). (1951). Field theory in social science:
selected theoretical papers by Kurt Lewin.New York: Harper &
Row.
40 ENTREPREMURSHIP THEOI?Y ond PMCTICE
-
\Churchland, P. M. (l9gg). Mauer and consciousness (rev. ed.).
Cambridge, MA: MIT hess.collins' B' E' (1974)' Four components of
the Rotter Internal-External scale: Belief in a difficult world,
atiijrll',)l;,^rt-ff:1Bti. world, and u poritfoaiv responsive
wortd. Journat of personatity and sociat
,t#l;Ja?;"fr,tffitaT|l;,jtJ, r.' (rs7s) Quasi-experimentation:
Desisn and anatysis issues for fietdcorman' J'' Perles, B',
&.vancini, P. (1988). Motivational factors influencing
high-technology
entrepre-neurship.JournalofSmaltBusinessManagement,z6
-
t-
Gartner, W. 8., Gatewood, 8., & Shaver, K. G. (1991).
Reasons for starting a business: Not_so_simpleanswers to simple
questions. In G. E. Hills & R. w. t-aForle fnor],'Research at
the markitingtentrepreneurship interface, pp. 90- 101. chicago:
office of Entrepi*"uria 'd,uoies, University of lllinois
atChicago.
Gergen, K' J' (1982)' Toward transformation in social knowledge.
New york: Springer-Verlag.Heider, F. (1958). The psychology of
interpersonal relations. New york: Wiley.
ii,\fr^\;l; A', & Aboud' J. (1971). characteristics of
successful entrepreneurs. personnet psychotogy,
Hornaday' R' w' (1990)' Thinking-about entrepreneurship: Afuzzy
set approach.presented at the Meetingof the Academy of Management,
San Francisco.Jackson' D N' (1974)' Personalit-v Research Form
manual. Goshen, Ny: Research psychologists press.James, W. (1890).
principles of ps,vchology. New york: Holt.Janis' I' L' (19''.2)'
victims of groupthink: A psychological study of foreign policy
decisions andfiascoes.Boston: Houghton Mifflin.Johnson, B' R'
(1990). Toward a multidimensional model of entrepreneurship: The
case of achievemenrmotivation and the entrepreneur.
Entrepreneurship Theory "na pr"itiii-,-ii(3),39_54Kahneman' D' '
Slovic, P,, &Tversky, A. (1982). Judgment under uncertainty:
Heuristics and biases. NewYork: Cambridge University press.
Kahneman' D" & Tversky' A. (1973). on the psychology of
prediction. psychological Review, g0,237-25t .
Kahneman, D', & Tversky, A. (1979). Prospect theory: An
analysis of decision under risk. Econometrica,47 ,263-291
Kelley' H' H' ' & Thibaut, J. w' ( 1978). Interpersonal
relations: A theory of interdependence. New york:Wiley.
Kenrick' D' T" & Funder, P c..^(19-8q) hofiting from
controversy: Lessons from the person-situationdebate. American
Psvchologist, 43, 23-34.Kogan, N., & Wallach, M. A. (1964).
Ris/c taking: A studlt of cognition and personaliry. New york:
Holt,Rinehart. & Winston.
lataf' B- & Darley, J. M., Jr. (r9i0). The unresponsive
bystander: why doesn,t he herp? New
york:Appleton-Century-Crofts.
LeBon, G (1895)' Psychologie desfoules. Paris: olean. (Trans. as
The crowd.London: T. Fisher Unwin,1896. )
Levenson' H' (1981)' Differentiating among intemality, powerfut
others, and chance. In H. M. trfcoun(Ed.), Research with the locus
of conrrol instruct: Assissment methodi, vol. r, pp. 15-63. New
york:Academic Press.
Locke, E' A', & Latham' G. P. (1990). Work motivation and
satisfaction: Light at the end of the tunnel.Psychological Science,
l, 240-246.Lopes, L' L' (1987)' Between hope and fear: The
psychology of risk. ln L. Berkowitz (Ed.), Advances inexperimental
sociar psychorogy, vol. 20, pp.2s5-)95. Neriyork: n.ui"*r" n
rr.Low'M'8"&MacMil lan, I
.c.(1988)'Entrepreneurship:Pastresearchandfuturechallenges.
JournalofManagement, 14, 139-16l
42 ENTREPRENEURSHIP THEORY ond pMCT|CE
-
!Lynn, R. (1969). An achievement motivation questionnaire.
British Journat of Psychology, 60, 5Zg-534.McClelland, D. C.
(1961). The achieving society. Princeron: Van Nosrrand.McClelland,
D. C. (1987). Characteristics of successful entrepreneurs. Journal
of Creative Behavior, 2l
.
219-233.
McClel land, D. C., Atkinson, J. W., Clark, R. A., & Lowell
, E. L. ( 1953). The achievement motive. NewYork:
Appleton-Century-Crofts.
Mehrabian, A. (1968). Male and female scales of the tendency to
achieve. Educational and ps-vchologicalMeasurement, 28.
493-502.
Mill, J. S. (1848). Principles of political economy with some of
their applications to social philosophy.London: John W. Parker.
rdner, J. B. (1985). Sentence completion measures in personnel
research: The development and validationof the Miner Sentence
Completion Scales. In H. J. Bernardin & D. A. Bownas (Eds.),
Personality assess-ment in organizations, pp. 145-176. New York:
Praeger.
Mischel, W. (1968). Personalit-t and assessmenl. New York:
Wiley.
Mitton, D. G. (1989). The compleat entrepreneur.
Entrepreneurship Theory and Practice, 13(3),9-20.Moore, C. F.
(1986). Understanding entrepreneurial behavior. In J. A. Pearce,
II, & R. B. Robinson. Jr.(Fls.), Academlt of Management Best
Papers Proceedings. Forty-sixth annual meeting of the Academy
ofManagement, Chicago.
Mukjerhee, B. N. ( 1968). Achievement values and scientific
productivity. J ournal of Applied P sychology,52, 145-147 .
Murray, H. A. (1938). Explorations in personality. New York:
Oxford.Murray, H. A. (1951). Some basic psychological assumptions
and conceptions. Dialectica, 5,266-292.Newtson, D. (1976).
Foundations of attribution: The perception of ongoing behavior. In
J. H. Harvey,W. J. Ickes, & R. F. Kidd (Eds.), New directions
in attribution research, vol. l, pp. 223-248. Hillsdale,NJ:
Lawrence Erlbaum Associates.
Om'e, M. T. (1962). On the social psychology of the
psychological experiment: With particularreferenceto demand
characteristics and their implications. American Psychologist, 17,
ii6-i83.
Paulhus, D. (1983). Sphere-specific measures of perceived
control. Journal of Personaliry and SocialPsychology, 44, 1253-1265
.
Pickle, H. B. ( 1964). Persornlity and success: An evaluation of
personal characteristics of successful smallbusiness managers.
Small Business Research Series No. 4. Washington, DC: U. S.
Government PrintingOffice.
Reid, T. (1863). The works of Thomas Reid (W. Hamilton, Ed.).
Edinburgh: Machlachlan & Stewart.Rosch, E. H. (1978).
Principles of categorization. In E. H. Rosch & B. B. Lloyd
(Eds.), Cognition andcategorization. Hillsdale, NJ: Lawrence
Erlbaum Associates.
Rosenberg, M. J. (1965). When dissonance fails: On eliminating
evaluation apprehension from attitudemeasurement. Journal of
Personality and Social Psychology, 1, 28-42.
Rosenthal, A. M. (1964). Thirty-eight witnesses. New York:
McGraw-Hill.
Rotter,J. B.(1966).General
izedexpectanciesforinternalversusexternal
locusofcontrolofreinforcement.Psychological Monographs, S0, Whole
No. 609, l-28.
Winter,1991 43
-
Rotter, J' B' (1990). Internal versus external control of
reinforcement: A case history of a variable.American Psychologist,
45, 489-493.Scherer, R' F', Adams' J' S., carley' s. s., &
wiebe, F. A. (1989). Role model performance effects ondevelopment
of entrepreneurial career preference. Entrepreneursnip'ri""i and
practice, I3(3), 53-71.schumpeter, J. A. (1934). Theory of economic
development: An inquiry into proJits, capital, credit, inter-est,
and the business cycle. New york: Oxford University press.
scott, L' R. (1990). Entrepreneurs and risk taking propensity:
Can both be created in the individual?Unpublished honors thesis,
College of William & Mury.'Seligman, M. E' P' (1975).
Helplessness: On depression, development, and death. San Francisco:
Free-man.
Sexton' D. L.' & Bowman, N. B. (1985). The entrepreneur: A
capable executive and more. Journal ofBusiness Venturing, I,
129-140.
Shaver, K. G., & Tarpy, R. M. (ln press). psychology. New
york: Macmillan.Shaver, K' G', Witliams, S. L., & Scott, L. R.
(199o).. Entrepre-ne.u.rial beliefs, creativity, and
risk-taking:Personality or situation? Unpublished manuscript,
College of williu. tni'*Skinner, B. F. (1971). Beyond freedom and
dignity. New york: Knopf.Solomon' G' T'' & winslow, E. K.
(1988). Toward a descriptive profile of the entrepreneur. Journal
ofCreative Behavior, 22, 162-l i l .
Spence, J' T', & Helmreich, R. L. (1983). Achievement
related motives and behaviors. In J. T. Spence(Ed')' Achievement
and achievement motives: Psychotogical "na toriotogiiit approaches,
pt t-74: S."Francisco: Freeman.
Stipek' D', weiner, B., & Li, K. (1989). Testing some
attribution-emotion relations in the people'sRepublic of China.
Journar of personarit-t and socir psychorogy, so, ros-iio.Stoner,J'
A' F' (1961). Acomparisonof
individualandgroupdecisionsincludingrls&. UnpublishedM.
A.thesis, Massachusetts Institute of rechnology, School oi
lndustrial Manage'ment.Strickland, B' R. (1989). Internal-extemal
control expectancies: From contingency to creativity.
AmericanPsychologist, 44, l-12.
Thibaut, J. w., & Kelley, H. H. (1959). The sociar
psychorogy of groups. New york: wiley.Timmons, J' A' (1978).
characteristics and role demands of entrepreneurship.
AmericanJournal of SmaltBusiness, 3, 5-17.
Titchener, E' B. (1898). The postulates of a structural
psychology. philosophical Review,7,449-465.Tversky, A', &
Kahneman, D. (1974). Judgments under uncertainty: Heuristics and
biases. Science, Ig5,tt24-tt3t.
Vanderwerf, P' A', & Brush, C' G. (1989). Achieving
empirical progress in an undefined,field. Entre-preneurship Theory
and practice, I4(2),45-59.V-enkatapathy, R. (1984). Locus of
control among entrepreneurs: A review. psychological Studies,
29,97- t00.
Vroom, V. H. (1964). Work and motivation. New york:
Wiley.Watson, J. B. (1913)' Psychology as the behaviorist views it.
Psychological Review,20, lsg-l.lj.
44 ENTREPRENEURSHIP fHEOt?y ond p|?ACT|CE
-
weiner' B' (19g6)' An attributionar theory of motuatton and
emoilon. Newy::1. 8., Russer. D.. & Lerman, D. ,e78)
^;:":",',,.":^-^::l:"
^'* york: Springer-Verrag
fly:v. w. Ickes, d^. I Kidd (Eds.). N;; !::::::'e consequences
of causal ascriptions. rn J. H.Hillsdale, NJ: Lawrencgg.rUaum
arriiul.r.,.. dtrections in anlibution research.
""l.,l..ip,.si_Xrwheeler, L. (1966). Toward a theory of behavioral
contagion. psychorogicar Review,73, 17g_rg2.Winslow..E K.. &
Solomon. G. T. (t9g9). FurtherJournal of Creative nii)rior, zs ,
tis-l6i'.
' -' -'Lr development of a descriptive profile of
entrepreneurs.Woodworth, R. S. (193g). Experimentul psychology.New
york: Holt.Wortman. M. S.. Jr. tt9g7)- Entrepreneurship:
An..-,.archinthefield' Journatof Managemen,,',t,r l i l-Zli i .t
ingtvpologyandevaluarionof theempiricar
i*ii,il.in.l"r";J]?lti; lhe^hu1gn choice: Individr,ii1.,reason.
and order versus deindividuarion237-307'.iffi;, u;iil,i *11S,*rj
#J'* (Eds ). 1vp6;a,; ;r;;:;^ on motivation, te6e. pp.
Kelly G. Shaver is professor of psychology at the College of
William & Mury.Linda R' scott is a student at T. c. wilriams
Schoor of Law, university of Richmond.
..^_ ,T":ujlrors thank Elizabeth J. Gatewood and Lversion of
this paper. J' v4rewuo(l ano Lanny Herron for theirhelpfur comments
on an earlier
\Mnter,199t
45