MAKING CHANGE PERMANENT

A MODEL FOR INSTITUTIONALIZING CHANGE

INTERVENTIONS

Achilles A. Armenakis, Stanley G. Harris, and

Hubert S. Feild

ABSTRACT

tncrcasing global compclition has accclcratcd the ralc of orgaruzalionat chqcs. such
as rccnginccrinp. rcstrucrurinp. and downsizing. As 3 rcsulr. organizarionat k&r*
tind thcmsclvcs fxcd w~rh growing cynicism among cmployecs that the current wave
of chanpcs is no!hinp more than (tic prqrum o/~/w monrh that will pass as Ulosc that
prcccdcd it. WC ilddrcss the issue of how IO make changes pcrmancnt by providing ;1
model dcvclopcd from theory and rcxxuch on ogani7;itional change and from
successful prdcticss iniplcmcnlcd in numerous orpankations worldwide. ‘The modct
can scrvc ar Icas~ three purposes. First. the model can assist chqzc agents in planrung
for and sscssing progress toward insMulionatizing organk.tionat change. Second.
:hc model can help focus cffons of orgak&ionat scholars IC) study the change
process. Third, the model offers the basis for hypothesis testing regarding the success
or failure of change cffofls.

Research in Oqgnkzat~ Change and Ikvelopment,

Vdme 12, pylps 97-128.
Copyright 0 1999 by 5.41 Press Inc.
All rigbls of rtpmducuon in My form rescrwd.
ISHN: O-7623-06H-O

97
98 ACHILLES ARMENAKIS, SI’ANLEYHARRIS, ,md HUHER’TfEI1.D

Increasing global cornpetitIon and changing political idcolopirs world\vldc are

some of the causes for the accrlcratcd rate of organizational changes. Managers arc:
being required to change virtually cvcry aspect of the way organizations function.
Employrcs’ attitudes about work, their jobs, and their psychological contracts bvlth
their employer arc being changed. As the pace of change incrcascs, employees arc
continually faced with evidence that some changes arc simply passing fads or
quick-fix attempts, implemented with little commitment for their long-term suc-
cess. It is no wonder that announced changes are met with skepticism and a
p'oS~rtrrr-of-tll~,-monrh reception. In such circumstances, it makes perfect WISC to
wait to see if the organir~ltion is serious about the change bcforc going through the
machinations it requires. Clearly. such a reception to planned changes by those
ultimately rcsponsiblc for implementing them is detrimental to the change effort’s
timclincss and success. Thus. the change h;Ls littlc or no chance of being in.rfifrc-
ficmc~li:c~d; that is. becoming accepted, permanent. stable and/or normative.
Change efforts fail to become institutionalized for varied reasons. However, much
of the problem revolves around the failure to shcphcrd the chanpc effort through
the cntirc procccssof change from diagnosis to institutionalization. Moreover. some
changes are implemcntcd simply based on the desire to bc in fclshionor to crcatc
the impression that the organi&on is being proactive. ‘Ihc goal is image rather
than substanti\r change and such efforts generally lack thr orgamzation’s commit-
ment and follow-through to succeed.
But what about change efforts implemented with sincere intcntlons to improve
the organization’! We contend that two primary reasons those who arc responsible
for planning and implementing organizational change fail to follow through on such
change efforts are (a) their impatience and assumption that successful change
introduction and implementation guarantees institutionalization, and (b) their sim-
plc neglect of seeing change through to institutionalization. WC suggest that the
succc.ssrate for planned organizational changes could be improved by giving
change agents and students of organi/;ltional change a better appreciation of the
institutionalizing phase of the change process-by describing the numerous pieces
that must bc understood. acted upon, and integrated before an organizational change
can be successful.
Therefore. the purpose of this chapter is to shed some much-needed light on the
institutionalization of planned organizational change. After exploring the moaning
of institutionalization, we propose and develop an integrative model of the prtrcss
of institutionalization. In developing our model, we integrate scientifically rigorous
research findings with numerous practical cxamplcs taken from our cxpericnccs as
consultants and from the popular press. These practical examples arc excerpted
from organizAtional change cxpcriences of noteworthy companies like Whirlpool,
GM. Goldman-Sachs. Control Data, Ford. Merck, Xerox, Allied-Signal, GE. and
Chrysler. Thus. our intent is to propose a model a change agent can use to answer
the question “What must we do to facilitate the adoption and institutionalization of
change’?’ The model can also serve to guide diagnosis and modification of change
efforts. For change scholars, the model provides a framework to stimulate additional
research and theory on the dynamics of organizAtiona change and management of
the institutionalization process.

WHAT IS lNSTlTUTlONALlZATlON?
‘I’hc issues of pcrmancncc and stability are central to Lcwin’s ( 1947) unfreezing -
moving-freezing metaphor. From this metaphorical perspective. an institutional-
izcd change is one that is frozen and the process of creating that institutionalization
is freezing. Beer (1976) describes frecr.ing as follows: “the stabilization of change
at a new equilibrium state through supporting changes in refcrcncc group norms.
culture. or arganiAona1 policy and strucnm" (p. 939). As organiit;ltions arc
constantly undurpoing change and experiencing flux, talking about literal ~XVWUJ-
MWCL to dcscribc institutionalization is unrealistic. However, cxpcricnce tells us
that some changes have a longer life than others. Thcreforc, it stems reasonable to
talk about degrees of institutionalization as reflected in the duration of a state. 13~
how should such degrees be concel)tualizcA?

The Kale of CommiImcnI

Institutionalization is reflected in the presence of resistance against deLiatin&

from the current state. Rcsistancc to change is the same as commitment to the
current state. In his pioneering rcscarch on the sources of individuals’ normative
confomiity. Kslman (1958) opcrationalizrd commitment into the three dimensions
of compliance, identification, and intc:rnali~atIon.
Kclman defined mrnpliur~cr cornmitrrrmt insthat lvhich occurs because an indi-
vidual expects to receive specific rewards or avoids punishment by conforming.
Resistance to change due to organizational structure. resource limitations, fixed
investments, interor~;lniz;ltional agrcemcnts. threats to power. and economic and
intcrprsonal vested interests all reflect resistance due to cxtcrnal pressures. fears,
or constraints. The appropriatcnsss of the change, that is, its righrncss or ~~n~~ync~
for the organization, is not a concern. These sources of rcsis:ancc represent
compliancebased commitment to the system.
ld~ntificution c-ontmifmunr occurs &cause an individual wants to sstablish or
maintain a satisfying self-defining relationship to another person or group. The
individual adopts the induced behavior because it is asstaiatcd with the desired
relationship. but the content of the responses may bc irrelevant. Culture, group
cohesiveness. and other social system vested interest sources of resistance a)1reflect
identification-based commitments.
/~rren~ulizr~fiot~ c.onmitrtwnr occurs because the content of the induced behavior,
that is, the. ideas and actions of which it is composed, is intrinsically appealing and
seen as proper. Thus. the behavior is adopted because it is congruent with the
!ndividual’s ~1~s. Paradigms. fear of the unknown, habit. and orguniir;ltional
1m ACHILLES AKMENAKIS, STANLEY HAKRIS, and HUDEKT FEILD

culture sources of resistance

reflect internalized commitment. that is, people and
the system the present
b&eve slate IO be appropriate. In general, inlernalired
commifmcnt because of its unconscious, preconscious. or automatic nature is
viewed ;1sbeing a more powerful and persistent determinant of behavior.
Since Kelman’s research, numerous researchers have made extensive contribu-
tions to our understanding and use of organizational commitment. Kescarch indi-
cates foci of commilment cm bc particular entities (e.g., an organization, people,
values) to whom a person is attached (cf. Becker, 1992; Fishbcin & Ajzen, 1975;
George. 1990; Mathicu & Kohlcr, l9YO; Keichers, 1085). Moreover. Brckcr’s
research expanded Kelman’s typology IO include organizational identification,
organiyAIional internalization. supervisor-r&ted identity. supervisor-rclarcd inter-
naliyation, work group identity, work group internalization, and overall compli-
ancc. Hc found supervisor-related and work group-related commitments were
distinguishable from the more general organiirAIional commitment. This conceptu-
alilation points out Ihc importance of realizing that organizational mcmbcrs can
identify with and inIernali/e the values of, not only the organiir;lIion. but also Ihcir
supervisors and Ihcir work groups. As described below in more detail, we incorpo-
rate Ihc importance of chanpc agent (including supervisor) and organicuional
member (including \vork group) attributes in the proposed model of the change
process. Thus, Becker’s (1992) expanded conceptualization of commitment fits
neatly into the mcKlcl for institutionalizing change.
In a recent study of organization development (OD) practitioners, Church and
Burke (1995) found that practitioners thought OD should begin IO focus more on
system-wide organizational issues and less on the traditional OD foci of individual
and group processes and intcrpcrsonal relationships. In a sense. our focus on
institutionaliir~tion rrprcsents such a shift. However, wc prefer IO think of our
approach IO institutionalization xs bridging and interweaving the distinction be-
IH’CCII the system and individual OD concerns. Specifically, wc suggest that the
process of institutionalization at the system level is Ihc process of building com-
mitmcnt IO the changed state (or building resistance to changing from it) at Ihc
individual Icvel. To create compliance-based commitment, a change agent must tic
the change IO organizaionnl structure. inIerorganiir;lIional agrccmcnts, sunk costs,
and reward systems. In order IO crcatc identification-ba.sed commitment, a change
agent must tie changes IO association with their supervisor and mcmbcrship in their
work group.FurIhcrmorc, IO create inIemalil;ltion-bacd commitment re!lccId in
individuals’ paradigms. a change agent must tic changes to current employee beliefs
and values as Ihcy relate to the organi~;lIional culture.
In most circumstances. 3 change agent will probably avant IO creatr institution-
&aIion based on compliance. idcntilication, and internalization commitments. To
the cxtcnt a change agtx~ wants ;1 rccc’nl change IO be more easily changed
(unfrozen) in the future, insIituIionaliz;lIiori based on complirrnce may be sufficient.
In casts where il is clear that many, if not most, cmployccs will resist inlcrnaliit;lIion.
it may also be appropriate lo focus mainly on crcaling compliance commitment.
pa;ticulCarly in the short term. However, over time, behavior resulting from compli-
nncc-based commitment can become normative, that is. I/W MJ NY’W aho,~s done
rhing.~.The commitment perspective 011 institutionalization encourages the change
qcnt to consider ad plan for ths degree of institutionali/.ation desired.

F,lctors Affecting Institutionalization

\Vhile littlc \vork exists on the process of facilitating institutionalization, efforts

hat-r been made to determine the types of chanpss that are most easily institution-
alized and the types of orgarG.ational factors that arc most conducive to institution-
alizing change. In general. interventions that arc received positively b)
organizational members are more easily institutionalizd than those rcccivcd ncga-
tively. For example, Tornatzky and Klein (1989,) found that Innovations that (a)
produced a relative advantage, (b) were more compatible with an organiitation, (c)
were relatively less complex, (d) were Iowcr cost, and (e) could be implemcntcd OII
;I trial basis w.cre more likely to bc adopted and institutionaliled.
Kcgarding organiAona1 factors, Damanpour (199 I) analyzed the tindings of 23
smpirical studies that investigated thr role of I3 content. contcstual, and process
factors on institutionaliration. Among the 10 factors found to bc ;usociatcd with
innovation were (a) functional diffcrcntiation (i.e.. content), (h) technical knowl-
et& resources (i.c., context), and (c) communication (i.e., process). The logic to
this research focus is that successful change may depend more otl the fit between
content. context. and process considerations than the nature of the change.
Although research on the attributes and organizational factors provides intcrest-
ing explanations for the success and failure of change efforts. practical application
of the findings ih limited. What options arc opml when relative advantage is not an
obvious attribute of the intervrntlon? What options arc open to those organiir.ations
that do nor fit the profiles? Finally, even if an nitervention offers compctlt~vc
sdvantage and an organization fits a successful profile, the tindings do not provide
insights into change processes and dynanucs. Xrc thcrc other furces !hat ma!
L*ontribute to Jnstitutionali/.atic,n: ‘To nddrcss such qucsfions, \vc propose a mtrlcl
to aid in understanding and in ultimately facilitating the institutionali%ilti(,n of
c hgc.

THE MODEL
Our process ~iidrl burlds off Lcwin’s (1947) stages of change and social learning
theory (Bandura, 1986). The model. depicted In Figure 1, comprises the following
constructs: ‘rhr three generic stages of change, the change rrlcssagc and its camp
1lc111s. coInnlitIncIlI (as explained above). the attributes of the change agent and the
org;Azitional membership. reinforcing strategies, inctitutionalization, and assess-
IIlL’IlI.
 ACHILLES AKMENAKIS, STAkLEY HAKRIS, and HUREKT FEILD

! MscafXYCr M+TWPAKENCSS SELF.EFFICN?Y P%hUPM PEaSCaAL I

SJPPOR’ VALENCE (
I
+

STRATEGIES
!

I - __.- --..
!Ir ’
1

t---
fO;wAL:UTION

I_._.
AcT:V.TIfS

- -.I

.--
1 1
___-.--.--
DiFFGW

I- PRACTICES
---. r
RITES 6

1 CEAEb(O’~IES 1
L---J

- - - -.- I
i I

I,

1
[-3
Finwe 1. Institutionnlizine Chanec
 The Stages of Change

Most models of the change process arc built around L&n’s (1947) stags ot
unfreezing. moving, and freezing. Paralleling Lewin, Bridges (1991). in his Hark
un transitions. frames the process in terms of endings, transitions, and new bcgn-
nlngs. In Figure I, we have used labels consistent with rcccnt change litcraturc (0
rlcscribe the three stages ;LSrcadincss, adoption, and institutionalization. Rrtrdir~c~ss
is the cognitive statr comprising beliefs, attitudes. and intentions toward a change
effort. When readiness for change rxists (cf. Armcnakis. Harris, 8c Mossholder.
1993), the organization is primed to embrace change and rcsistancc is reduced.
Organi?A;ional members will embrace the change and the adoption stage begins. If
nrganizational mcmbcrs are not ready, the change may be rejcctcd. and organiza-
tional members may initiate negative rractions. such as, sabotage, abscntecism. and
output restriction. A&@n is the act of behaving in the new way. on a trial basis.
That is, the change can still be rejected. As discus4 earlier. insrirurio,zr~li;LJIion IS
rctlectcd in the dcyrcc of commitment to a new way, that is, the post-change state
of the system.

‘The Change Mcss,~gc

ht the core of our model is the rrtessczgerequired to build commitment to a change

effort. All efforts to introduce and institutionahz change cm be thought of ;LS
sending a rncssagc to organizational members. The introduction of change creatas
a great deal of uncertainty and confusion. I3scntially. the purpose of the change
~nessagr is to crcatc cenain core sentiments in members of the organization h;
answering a set of five key questions they have about the change. ‘Ihc first question
is “Is change really necessary?” The question is answered by the discrcpnq
component of the message. Discrepancy refers to information regarding the need
for change as reflected in the discrepancy between the current and ideal state in the
organiir;ltion. The second key question raised by organizational members is “Is the
specific change being introduced an appropriate reaction to the discrepancy’?” The
ul’l”T’l’ricrf~,nr.s.scomponent of the change message provides the response to this
question. A third qucslion generated during a change is “Can I/we successfully
implcmcnt the change?” The eficcrcy component of the change mrssagc answers
this question by providing information and building confidence regarding the
individual and group’s ability to successfully implement the change. X fourth
question has to do with organizAtional support for the change and reflects skcpti-
cism resulting from previous yrr,Srutn-of-rhe-monrh. half-hcartcJ change intcrven-
tions. The purpose of the principal srcppon message component is to provide
information and convince organizational members that the formal and informal
leaders are committed to successful implcmcntation and institutionalization of the
change. Finally. organiLationa1 members will want to know *‘What is in it (the
change) for me?” By clarifying the intrinsic and extrinsic benefits of the change.
 ACHILLES AKMENAKIS, STANLEY HAKKIS, and HUREKT FElLI)

the persond valence component of the change message addresses this question.
Imbedded in the concern for personal valence is the intrinsic desire for change
fairness and justice. As Cobb, Wooten. and Folgcr (1995) note, employee pcrccp-
tions of justice during periods of organizational change encompass assessments of
the fair distribution of positive and negative outcomes, of the fairness of change
procedures. and of the appropriateness of change agents’ treatment of them.
Perceptions of justice are particularly important for encouraging the typ of
extra-role behavior generally rcquircd of change efforts (Cobb et al.. 1995).
The degree to which organizational members receive adcquatc answers to their
core questions is a prime determinant of the nature of their ultimate commitment
to the change. While the emphasis in this chapter is on the institutionali/.ation of
change, it is important to note that the change message also has implications for the
creation of readiness for change (Armenakis et al., 19%) and its adoption. In fact,
to the degree that the core questions are answered adcquatcly in early stages of
change, sentiments central to institutionaliition may already be established.
The role of these five messages in generating positive change momentum is
rxemplificd in an investigation conducted by Nutt (1986). Nutt studied the change
implementation tactics of hospital executives in 9 I case studies. The most succcss-
ful tactics (labeled intervention and participation) described the change agent ;LS
demonstrating early support for the change (principal support), communicating the
need for change (discrepancy and appropriateness), and involving organi/.ational
members throughout the change process (efficacy and personal valence). The least
successful tactic (labeled edict) described the change agent as not discussing change
plans with organizational members, not justifying the need for change, and using
control and prsonal power to mandate adoption.
Before examining the strategies that can be employed to send the five key
mcssagc components, it is important that we briefly examine attributes of the
change agent that affect the persuasiveness of the change mcssagc and charac-
tcristics of the organizational members who are the targets of the change that affect
their receptivity to the mcssagc.

Change Agent Attributes

Anyone involved in initiating. implementing, and supporting a change can be

considered a change agent. Initially, this change agent may be the head of the
organization (i.e.. global change agent). People in all leadership positions will
ultimately be expected to support and help drive the change throughout the
organization. For large-scale change programs involving numerous organi;t;rtional
levels and departments. executives and other managers are extensions of the change
agent, thus, serving in a role of loccrl change agents. Finally, nonmanagerial
organizational members can serve as change agents (i.e., Itori.ymaf change agents).
Ihe.se horizontal change agents can bc those persons who interact socially (on the
job, as well as, off the job) with collcagucs as opinion leaders and can reinforce the
 ‘avorablc interpretation of the message. Naturally, organi;t;ltional members’ pcr-
:cptions of the change agents’ attributes will influence the persuasiveness of an>
:hange message and ultimately commitment. hence, institutionalization of orpan-
zational change.
The single most important attribute that a change agent should possess is
:redibility. Kouzes and Posncr (1993) identified the primary components of credi-
)ility IO be honesty, competence, vision, and inspiration. The importance of
:redibility in changing cognitions and behaviors has been researched for several
Iccadcs and continues to be of interest in current research investigations. According
o research cited by Slater and Kouner (1992). changes in cognitions of organiza-
ional members have been linked empirically to the credibility of the change agent.
rlystrom (1990) found that the quality of the relationship between a change agent
md the organizational members was a significant factor in determining commit-
nent. Similarly, Eisenberger, Fasolo, and Davis-LaMastro (1990) concluded that
organizational member perceptions of being valued and cared about were instru-
nental in influencing innovativeness. In the Bullcr and McEvoy (1989) invcstiga-
ion, trust in the change agent was significant in institutionalizing a new
xrformancc appraisal system. Communicating a shared vision through spcahes.
ncmos, and newsletters along with cxecutivc visibility were significant in dcvel-
aping organizational commitment (Niehoff, Enz, & Grover, 1990). Larkin and
-arkin (1996) refer IO numerous surveys of employees from large organizations
hat revealed the preferred information source was frontline supervisors. implying.
‘rontline employees had littlc confidence in upper-level executives. Likewise, Cobb
:t al. (I99S) summarize the results of several studies that clearly demonstrate the
mportance of the perception among employees that the change agent can bc trusted
md is viewed as being, or attempting to be, fair and just in the way in which the
:hanpe is being managed.
The obvious conclusion from these findings is that a required, although not
iuflicient. condition for institutionalization is that organizational mcmbcrs should
Jerceivc the change agent and his/her rcprcsentativcs as credible. Change agents
icvelop credibility through their behaviors.

Attributes of the Organizational Membership

The organizational membership is the collection of individuals who must modify

heir cognitions and behavior IO achieve the objectives of the change effort.
Jltimately. it is the commitment of that individuals that determines the institu-
ionaliration of a change. Confronted with the same information and intervention,
ndividuals can still bc expected to react differently. IArming to organizational
nembers, even those who may not be sold on the change, can result in a better
nutual understanding of the concerns (Weisbord, 1988). Then, this understanding
:an bc used to anticipate differential reactions of organizational members and can
x1helpful in orchestrating strategies intended to build their commitment for chanpc
106 ACHILLES ARMENAKIS, STANLEY HARRIS, and HUBCRT FEILD

initiatives. Research on two factors, namely, individual diffcrenccs and organiza-

tional diffcrcnriation, provides some guidance in the process of instituConalization.
In terms of individual differences. two scales arc valuable for undersranding
change dynamics. Kinon’s (1084) Adaprion-Innovation Inventory. a papcr-and-
pencil instrument, has been used w categoric individuals as cirher adaptors or
innovators. Innovators are more likely 10 embrace fundamental change Hhilc
adaptors arc less likely to embrace fundamental change (Kirton, 1984). A second
scale. the Self-IMonitoring Scale (Snyder, 1974) \VU used by Burkhardt (I SK,I ) who
found that high self-monitors’ (Lc., those who were more attentive 10 social
comparison information) altitudes toward change were more influenced by opmion
leaders and other individuals in their work groups. In contrasI, low self-monitors
were more influenced by chose individuals who were performing jobs considcrcd
to be hierarchically similar.
The second factor 10 be considered in change institutionali/;ltion is the orgamza-
Conal differentiation Ihat exists rhroughout the organization. Organizational change
[hat originates outside of a group may be perceived as a threat. Defensive mccha-
nisms arc mobilized to ward off the intervention that will undoubtedly upset the
norms. Plans lo address thr needs of these diverse groups (e.g., umon members,
professional classifications) and enlist the support of some a change agents can
cnhancc the success of institutionaliitAtion.
0rganiir;ltional diffcrcntiation has been rcscarched in the change literature ;LS
subcultures (cf. Van Maanen & Barlcy, 1984) and cultural ecology (Baba, IWS).
Baba’s research explained how I5 work groups reacted 10 an organizational
transformalion aimed a1 commonizing the tools and methods used in the producl
development process of a large manufacturer. Baba’s data revealed that the diffcr-
entiat reaction of the work groups was related to the cultural ecology (i.e., the tyw
of work they do, relationships with other mtemal and cxtemal organizational
groups, and availability of resources) of each group. Thus. the cultural ecology of
the work group detcrmincs how individuals within a work group will react to an
organizational change.
Researchers investigating diffusion of agricultural innovations (Ryan & Gross,
1943) found that individuals who were highly rcspccted influenced the willingness
of others 10 institutionalize change. Therefore. ic is critical that opinion lcadcrs
within the cohcsivc groups support the change and. consequently, influence others
10 embrace it, thus building momentum. Organizational commitment and commit-
ment 10 change initiatives can be gcncratcd throughout the organization by peers
serving as role models and providing social support and becoming horizontal
change agents (e.g.. giving cncouragemcnt and posirivc feedback).
Implications of the findings regarding individual differences and organiz&onat
differentiation are that a change agent needs to be familiar with organi;r;ltional
members and should attempt to enlist support for change. For example, organiza-
tional difiercntiation, including unions, engineers, and technical specialists. should
be considcrcd when identifying opinion leaders. Although it may be impractical to
ISSC’SSorganizational members by administering paper-and-pencil tests IO identify
nnovators/adaptors or high/low self-monitors. these findings are important in
mdcrslandinp change dynamics.

Institutionalizing Stratcgics

Naturally, during organization chanpc programs. change agents initiate actions

haI arc intended to redetinc norms and establish ne\v ways of thinking and
>ehaving. Their actions (or lack of action) can have both real and symbolic
zonscqucnces with regard 10 sending and reinforcing the five core mcssape com-
~)ncnts. In short, everything a change agent says (or does not say) and dots (or
Joes not do) can reinforce (or contradict) the change message. Executing the seven
nfluencc strategies shown in Figure 1 is intcndcd to transmit and reinforce the tivc
:ore message components.

4crivc PJrtic~;p<ltion

Participation strategies can enhance the relationship between a change agent and
organizational members, build the credibility of the change agent. and establish
)wnenhip in and reinforce commitment for organiir;ltional change (cf. NUN. 1986).
Grthcrmorc, active participation’s cffcctivencss as a change strategy is based on
he concept of self-discovery, that is, the learning that results through personal
:xpericnccs. Three active participation tactics are particularly useful in building
:ommitmrnt IO chanpc: (a) cnactive mastcry, (b) vicarious learning, and (c) partici-
native decision making.

~r~~clive rnastcry. Enactive mastery is accomplished by gradually building

:omixlcnce and skills through successively accumulating activity blocks; that is.
:hunks of the total responsibilities. In cnactivc mastery, more complicated behav-
ors are NH introduced until the previous behaviors have been mastered. In this way.
ndividuals expcricncc srttclll N%IS (Weick, 1984) and are not confronted with the
nagnitude of the overall behavioral change requirement all at one time. In some
;Ituations, cnactivc mastery may mean initiating a new practice. such as quality
ncetings IO discuss very simple issues on an infrequent basis. Over time, howcvcr,
he frequency and intensiveness of the meetings can bc increased as people become
norc acsustomcd to them. Through enactive mastery, organizational members can
:xpericncc and realize any advantages of the new procedures and can gradually
fcvelop a sense of ct’fcacy with regard to those prcredurcs. ‘Through prolonged
lractice and exposure. it is alw) possible that cnactive mastery can be a soot-cc of
jcmonstrating the appropriateness of a change.

\‘ic~rious le.~ming. Ucarious experiences arc those during which organiit;l-

Ional mcmbcn observe others. most prcfcrably respectrxl collcagucs, in rhe per-
*ormance of rhc new behaviors. IIcnchmarkmg. the process of identifying and
 ACHILLES ARMENAKIS, STANLEY HARRIS, and HUBERT I-EILO

emulating the excellent practices of other companies, is amother type of vicarious

lcaming experience and results in enhanced efficacy (e.g., “if they can do it. so can
I”) and appropriateness. bccau.seindividuals will bc able to observe any advantages
of the new method. Furthermore. principal support for the intervention is observed
as rc3pccted colleagues initiate and continue the adoption of the organizational
change. Desire to adopt the new behavior results from wanting to be identified with
the adopters. Conscqucntly, the vicarious obscrvcr is encouraged to continue
adoptlon and ultimately institutionalizes the new behavior.

Paticipativc decision making. Participative decision making is an effective

method to change cognitions and behaviors (cf. Beer, Eisenstat, & Spector, 1990;
Dalton. 1970; Neumann. 1989; Pasmorc & Pagans. 1992). To the extent that
participative decision making cnhanccs members’ psychological ownership of
change in the orgamzation. indivlduals’dispositions toward promoting such change
will likely be enhanced (cf. Dirks, Cummings. & Pierce, 1996). Individuals can bc
involved in decision making for all phases of the change process. Naturally. there
arc limits IO participation. and there are situations in which participation is simpl)
not possible. Some aspects of a change program may involve participation, such as
strategic planning, but other aspects, such as downsizing. may involve only sclectcd
decision makers. ‘Through their involvement in making change-rclatcd decisions,
organizational members self-discover a clearer understanding of the problems and
issues confronting their organization (discrepancy) and the improvements that can
be realized from implementing change (appropriateness). Likewise, having a say
in the changes that are ultimately introduced is likely to make individuals feel
efficacy for those changes. As shown in the goal-setting literature. people have more
confidence in their ability to meet a goal that they participated in setting (cf. Locke
& Latham, 1990). Through participation, individuals observe the support of other
organizational members. Participation also increases the likelihood that changes
consistent with intrinsic/extrinsic motives arc selected and that the change proce-
dure is vicwcd as being more fair and just because participants are given a \*oice
(Cobb et al., 1995; personal valence).

Persuasive Communication

Persuasive communication strategies arc useful in efficiently communicating

information relevant to all five core message components. Oral transmissions can
be formal. such as speeches, or informal, such as chance faceto-face encounters.
In addition, oral transmissions can be transmitted live as well as through audio- and
videotape tcxhnology. Written messages can take the form of memos, c-mail
messages. annual reports. letters/memos, and newsletters.
Cobb et al. (1995) make it clear that persuasive communication is a particularly
important vehicle by which change agents provide social accounts which shape
perceptions of the change and its fairness and justice and can be used to build
support for 1hc change. Drawing off the work of Bies ( 1987), Cobb et al. (1995)
;lescribe four such social accounts. In causal accounts, change agents articulate the
reasons and rationale behind the change and engage paradigms that provide a more
rrccq~able interpretation of the change. In ideological accounts, change agents
justify the change within the context of superordinate goals or visions and clarify
the values which will guide change decision making and se1 the standard for
definitions of what is just. Referential accounts clarify benchmarks for the change
and clarify how things will be worse if change is not made and how things will bc
better once the change is made. Finally. penitential accounts are used to honestly
address the dlfficultics of the change, make apologies, and demonstrate cmpath)
with the employees.
In addition to the verbal content of the messages sent, the nature by which those
messages is sent can have symbolic consequences as well. For example, the time,
energy, and resources utilized in giving verbal communicarion provide symbolic
cvidencc of the change agent’s support for the change effort. In addition, change
agent a1trmpts to communica1c powerful metaphors to summarize the change
process often helps clarify events and processes that might othewise be confusing
for organizational members.
Because of pcoplc’s inherent stance of “I’ll believe it when 1 see it,” pcrsurLsivr
communication cannot Iw expected to be ;FSpowerful as a strategy based in
self-discovery. like cnactivc mastery or vicarious Laming. However. persuasive
communication is a basic and required strategy of all change agents bccausc the act
relates to one or all of the attributes of credibility. Furthcrmorc, the qmholism
emanating l’rom the pcrsurrsive communication incidents can bols1er the message
transmitted by other strategies. To illustrate the integrated use of various wrsuasivc
communication strategies. examples from WhirlF)ol and General Motors arc
briefly 0utlinLui below.

LVhirlf~ol cwmplc. During 1988, Whirlpt~l Corporation leaders changed

the corpmte strategy and reorganized from a functional to a strategic business unit
(SDU) orgamit;ltlonal structure, followed by a 10 percent reduction m managers. A
significant attempt a1 persuasive communication occurred in June 1988. when the
Whirlpool chairman and CEO, Dave Whitwam. addressed the 250 officers. dircc-
tars. and managers (and spouxs) in an 8@minutc, after-dinner specvh. The purpose
of the speech was to demonstrate symbolically his support for the changes. as well
as to provide a degree of confirmation for them. Regarding discrepancy, the CEO
rciteratcd that the changes were needed because of increased global competition
and the accelerated rate of industry consolidation. The old functional structure
inhibited the company’s ability to capitalize on opportunities that existed in the
market. As for appropriateness, Whitwam pronounced that with the changes, the
company could now more effectively compete. To communicate his support for the
changes that were implemented, he assured them “there is no turning back
Whirlpool of the past will be no more” (Confirnuuion Speech. 1988).
110 ACHILLES ARMEhAKIS, STANLEY HAKRIS, and HUBERT i’EILLI

This persuasive communication incident is significant because it required lime

and effort in planning, reinforced the change, and was symbolic. For example.
making arrangcmcnls for all direclors. officers, managers, and their spouses for a
formal dinner and IO prepare a videotape of Whirlpool’s Corzfinnafiorz Spdz for
broader dissemination required more than a casual effort. Furthermore, the cvcnt
was intended IO do more than summarize Ihc process. By incorporating Ihc spouses
at the dinner, the intent was apparently IO acknowledge the impact of the changes
on Ihc whole family and IO enlist familial support.
A symbolic innovative atrcmpt, referred IO as dramaturgy (Kitti & Silver, lY86).
IO institutionalize Ihc changes at Whirlpool was the preparation (in IYYO) of a
videotape of three top cxccutivcs (the CEO, (he CFO. and the Exccut~vc Vice
President) being interviewed by a French ncwscastcr (Business in f’n~file, IWO).
The setting for [he staged intcrviow was live years into Ihc future, and the three
executives wcrc answering questions related IO the reasons for Whirllx)ol’s success
during Ihc 1as1decade of the Iwentleth century. In introducing the executives, the
newscaster referred IO Whirlpool as the renczissunc-e cornpuny of rhr f YWs. In the
staged discussion, the executives attributed the company’s outstanding performance
(in the future) to the organil;lIional changes which were currently being implc-
mentcd in real life. Thus, this dramaturgy was r&forcing (a) their support for the
continuing adoption and insIiIuIionali/aIion of the new/improved ways of opcrat-
inp. (b) the need for and appropriatrncss of change (i.e., fhc new way proved better
than Ihc old ways), (c) efficacy (i.e., Whirlpx)l was capable of responding more
quickly IO external environmental changes) and (cl) valence (i.e., the cvidcncc:
indicates that the organizational changes were indeed something IO bc proud of and
had been rcbvarded in the marketplace).
It is noteworthy IO point OUI that this Whirlpcx)I example is refcrrcd to as r*i&o
fwd-jonturd (Dowrick. IYYI). Dowrick and associates describe some interesting
research findings (with various groups of people, including alcoholics, children
with reading disabilities, and athletes) that drmonstratc its effectiveness in master-
ing new skills and enhancing one's ability IO achieve higher levels of performance.
‘1%~idea is IO permit subjects IO view thcmsclvcs “not as they pcrformcd in IIIL’ past
(including errors), but as Ihcy will perform (correctly) in the future”(pp. 240-24 I).
Using video feed-forward with a nationally ranked pnvcr-lifting athlete, Franks
and &Mailc( IYYI) rrcorded a 26 percent improvcmcnt in performance in a 2%week
period. A 10 pcrccnl incrcinc in a one-year period is considered rcnrfzrkrzble,
cs~cially when one considers (he elite pcrformancr level of such an achletc. Thus,
video feed-forward shows promise for implementing organia~lional change.

GM cxarnplc. In 1YY2, General Motors Corporation announced numerous

executive successions in Ihc wake of staggering operating losses. Subscqucntly, Ihc
new leaders mitiatcd a series of organizational changes intcndcd IO improve Ihc
corporation’s performance. In its 1992 annual rrport. the chairman, John Smalc.
rcfcrrinp IO its austere apl>enrance.cmphasizcd “that this annual rcpon is a dramatic
departure from past practices. I bclicve it demonstrates the C’orporatlon’s commit-
ment to mcantngful change” (Gmcrrd hlorors Annud Reporr, 1992. p. 3,. In
addition, the CEO, Jack Smith communicated the appropriateness of the changes
b) stating “I firmly ~CIILX your (Corporation is, right now, on the way to achieving
the strength it once had. WC know what we have to do, and we‘re going IO do 11.
Watch and see” (General Moror.sAnruu~l Repm, 1992. p. 3). Therefore, this report
reinforced discrepancy (i.e.. company performance was unacceptable) and efficacy
(i.e.. WCarc capnblc of returning to the strong company we once were). This annual
report was symbolic bccausc of its austere appearance. The statements made b)
both John Smale and Jack Smith communicated their support. Furthermore. thcu
statemcms and the strategy wcrr intended to demonstrate their vision and to insplrc
support from GM personnel.

Information from internal and external sources is a powerful lever for reinforcing
the message necdcd to institutionalize change. Examples of internal data are
employee attitudes. productivity, costs, and other performance indicators (e.g..
scrap) usually considered to he minimally influcncsd by external cnvironmcntal
fluctuations. Examples of external data include direct contact with customers via
tclephonc surveys conducted by cmployres or articles m the business press. Clearly,
much of the content of the information uill bc incorporated into messages conveyrd
using the various fomis of pcrsu:Lsivccommunication. To the extent that individuals
affect& by the change can actively participate in gathering thr: information.
believability is enhanced. Some information allows for the tracking of change
success. so important to celebrating small wins (Weick, 1984) .and keeping indi-
viduals focused on the desired changes.
Collecting survey data has a long history .asa tool for identifying the ncod for
change within the action-research community (cf. .Kadlcr, 1977,~ After a change is
initiatcci, organizational members can be reinforced to continue adoption by track-
ing the progress of change with data collcctcd from internal and external stakehold-
ers.
I.everaging and managing information from external media sources, such as the
popular business press. can be disscminatcd to further enhance adoption and
institutionalization (cf. Mncdonald. 1995). For example. compiling and sharing
press releases regarding the activities of competitors can be used IOjustify change
efforts. Other tactics, like contracting with consultants and high visibility speakers
can be very effective in reinforcing the ideas among organizational members that
the firm is employing leading cdgc technology and is considrrcd among the elite
in their business arca.
112 ACHILLES ARMENAKIS, STANLEY HARRIS, and HUBERT FEILD

Human &source Management Pra3iccs

Human rc’source management (HRM) practices can be used tocomplement other

strategies in the institutionali;r;ition process. HRM practices include selection,
performance appraisal, compensation, and training and development. While each
practice can contribute to reinforcing all five core message components. human
resource practices arc generally a primary source of extrinsic reinforcement for
desired behavior and symbolic cvidencc of organirAtiona1 support for the change.
It is important to realize that much of the reinforcement of change is aocomplishcd
through HRM practices (cf. Iichy, 19X2). As evidence of the power of HR.M
practices, Yeung, Brockbank, and Ulrich (I 99 1) report thr results of a study of I .2(X)
businesses showing that HRM practices supprt change initiatives. However, care
must be taken to consider employee attitudes rclatcd to the perceived faimcss o!
change when using these HRM practices to reinforce change implementation.
Kilboume. O’Leary-Kelly, and Williams (1996) studied employee reactions to a
shift from seniority-based to performance-based layoffs in a large electronics firm.
They theorized that in addition to procedural and distributive justice concerns,
transformational justice issurs impacted employee attitudes toward the perceived
fairness of change. IMorcovcr. they concluded that justice concerns are present
during change implementation, and the interrelatedness among HRhl practices was
critical to employee justice perceptions of change.

Selection. Selection decisions include firing. hiring, transferring, and promot-

ing or demoting employee%. At the most basic level, selection systems can be used
to reinforce change adoption and discourage resistance. Sclestion activiticx can bc
used to institutionalize change by hiring and promoting those individuals whose
values match those represented in the changed state of the organi;rAon and
removing those individuals whose values do not (Pascale, 1085).
During Friods of retrenchment, when organizations arc downsizing. the corn-
mitmcnt of the nt~i~ors may bc influenced by the treatment of the selcctcd \i~-~ims.
Brockncr, Grover. Reed, Dewitt. and O’.Malley (1987) found that publicized
benefits (e.g.. severance pay, job placcmcnt, and continuance of health insurance)
provided to layoff victims significantly affected the commitment of the survivors.
‘Ihercforc. the quality of the relationship bctwcen the agent and orpanicAonal
mrmbcrs prior IO any changes and the actions taken regarding those adversely
affected during crisis will influcncc the survivors’ response.
In addition IO the usual knowledge. skills, and abilities (KSAs). sclcction spcci-
tications may include the likclihocxl of being committed to changes intrtxiuccd.
‘Iherefore, an understanding of salient job requirements, plus the commitment of
the time nrccssary to identify appropriate personnel. will contribute to the institu-
tionalizuion of change programs. Furthermore, this knowledge can bc applied in
promoting and transfcming from within (thus capitalizing on symbolism) thosr
individuals who will initiate and sustain the social dynamics necessary to accelerate
 he change process. Parkas and Wetlaufcr (1996) de-scribed how Goldman-Sachs
rncouraged managers to volunteer for foreign assignments. A managing partner of
the company explained “It was just not valued as an attractive career opportunity
by most of our US people, and their spouses didn’t ncwcssarily want to go, and their
dogs couldn’t possibly endure living in Tokyo...So WC took an exceptionally
talcntcd young banker and promoted him to partner two years ahead of his class
because he went to Asia at great personal sacrifice” (Farkas & Wetlaufer, 1996.
p.121).
Selection can be a powerful practice to reinforce discrepancy, efficacy. support,
and valence. The survivors of an effectively executed downsizing should sense the
change was needed and that the organization feels they are capable of performing
the new behaviors. Job candidates who satisfy the technical and interpersonal job
criteria, after extensive screening, may appreciate the thoroughness of the selection
procedures and be more likely to contribute to a synergistic relationship with peers.
subordinates, and superiors. Furthermore, rewarding individuals for volunteering
for unpopular assignments and performing admirably can transmit new meaning to
unpopular duty. The rewarding of such volunteering will demonstrate the organi-
zation’s support and likely create support among OrganizAtional members.

Pcrform;rrrce apprdisal. The importance of feedback in changing one’s behav-

ior has been documcntcd in field and laboratory studies (Brctz. Milkovich, & Read,
1992). Ilgcn and Moore (1987). for example. found that feedback about quantity
led to higher quantity, feedback about quality led to higher quality. and feedback
about both led to higher quantity and quality. Including change criteria in appraisal
systems serves to constantly reinforce the desired behaviors. Performance appmisal
systems can contribute to institutionalization of change by provtding appropriate
fettdback in a timely manner to organizational members. Pcrformancc appraisal
systems at Control Data (Gomez-Mejia, Page, & Tomow, 1985). Ford Motor
Company (Scherkenbach. 1985), Merck (Wag4 1987). and Xerox (Deets & Tyler,
1986) have been designed to promote behaviors nwessary to accomplish changes
in corporate strategy.
Of late, a particular form of appraisal system. the 360-degree or multisource
feedback (MSF) program, has become popular. MSF is feedback to leaders and
managers from superiors, peers, subordinates, vendors, and customers aimed at
improving organizational effectiveness (cf. London & Beatty, 1993). Its focus on
change tenders it appropriate for the adoption and institutionalization process
because (a) new visions of the types of leaders an organization will begin to reward
and advance can be incorporated and(b) knowledgcable individuals can be involved
in the design of the MSF program (London & Beatty. 1993).
Lepsinger and Lucia (1997) describe an MSF process introduced at a bank. The
leaders responded to the numerous external environmental changes affecting the
banking industry by developing a business strategy that addressed service excel-
lence and customer focus. In order to translate this to the branch-based employees.
114 ACHILLES AKMENAKIS, STANLEY HAKRIS, and HUBERT FElLI:,

an MSF program was designed IO ensure that individual employees’ skills and
orientation were realigned IO support the new strategy. An MSF instrument was
completed by self, superior, direct reports, and peers. Personalizcxl feedback reports
were presented IOeach employee. Skill gaps for each individual were idcntitied and
during one-on-one development meetings, plans were formulated that speciticd
how skill gaps were IO be eliminated withm an 1% IO 24-month time period.
Pcrformancc appraisal and MSF provide an opportunity for an cmployce IO
receive feexlback regarding the adoption and insIiIuIionalizaIion of a11 organini-
Iional change. A difficulty in adoption is related IO the delay of the new behavior
resulting in tangible personal benefits. However, periodically giving feedback about
employee performance in rcxently changed jobs can influence valence and efficacy
sentiments. This feedback can serve IO extend the adoption (trial) period, thus
enhancing the possibility of insIiIuIionaliir;Ition.
A great deal of research has been conducted on performance appraisal focusing
on understanding and improving ability, through changing formats and training.
The motivation fur appraising the performance of others has not bcrn well rc-
searched (cf. Harris, 1994). However, rater motivation in performance appraisal is
an important variable that mu51 be recognired. If performance appraisal is IO scrvc
as a change lever, change agents must recognize the importance of this 1001. II
appears that the popularity of MSF will stimulate research, hopefully on improving
change agent ability and motivation IO appraise performance.

Cornpnsa~ion. The acceptable performance of a job is rewarded through some

Iype of compensation program. Com~nsatton typically consists of fixed interval
plans (e.g., hourly, bi-weekly. monthly.) and variable plans (Le.. incentives). plus
employee benefits (e.g.. pensions, medical allowances, annual/sick leave. and
severance pay). These Iangiblc rewards can be very effective in reinforcing adoption
and ultimately insIiIuIionalizaIion of organizational change.
Bandura (1986) argued that extrinsic incentives (e.g., financial compensation,
special priviledges, social recognition. advancement in rank, and other starus-con-
ferring rewards) should be provided to sustain adoptive behavior until the intrinsic
value becomes apparent. When an organization establishes a new compensation
plan to encourage organizational mcmbcrs IO adopt an organiit;ltional change,
change agent support is apparsnt. Furthermore, when a manager ccremoniallg
recognizes a team or an individual. it is obvious that management supports the
accomplishment. The likelihood of receiving compensation. in general, can be a
significant factor in innuencing the valence for an organizational change. The x~ual
receipt of compensation will reinforce efficacy. When this compensation is tied IO
the goals of the organization change effort. organizational members infer that the
change is appropriate and was indeed needed.
Traditionally. organi&ons have devclopd vertical hierarchical structures such
that advancement within Ihc organization has been pcrccived ;LFupward movement.
‘Typically. pay structures have matched hierarchical levels. However. the recent
rend toward flattening organic~tional structures is incongruent wth 3 rigid multi-
grade pay scale be~~usc the pay grades no longer match the reduced number of
organizational levels. One way to resolve this incongruency is to introduce broad-
wdmp. the act of collapsing pay grades into a few v.idc bands, thus suppwtt’np
oday’s flatter orpaniatwn (Abosch & Hand. 1994). By having fewer grades.
‘estructurmg, with the concomitant changes in roles and responsibilities. i$ less
lift&It. Among the benefits to broadbanding arc that it: (a) provides greater
qwtunity for cross-functional moves; (b) fosters team building and functioning
Krause team members will likely be from the same band: and (c) emphrwzes
;k~ll&ompetcncies (Abosch. l995).
Broadbanding is consistent with the concept of teamwork and with varwus p:~>
Jclivcry strategies. Skill-hwed pay proprams (i.e., rewarding individuals for acqui-
;ition and demonstratwn of work-rclatcd skills and competcncics). carecr-dcvclop-
ncnt pay programs (i.e., rewarding lateral job involvement to encourage breadth
~1’experience. knowledge, skills, and abilities), and merit casl~~ain-shariri~ pro-
Irams (i.c., rc-carn:lble lump-sum awards for indi\ idual/tcam performance) arc all
:onsistent with the broadbanding concept.
Research on these pay delivery strategies has shown them to hc effective III
Gnforcing organizational change (cf. Bullock Kr ‘I’uhbs, 1990; Murray & Gcrharl.
19%). Each provides various degrees of extrinsic rewards (e.g.. periodic distribu-
ions of monetary payments. minimally inllucnccd by external cnvironmcnral
1isturbanccs) and intrinsic rewards (e.g.. autonomy and skill developmrnt).
Some factors that arc possible with these pay delivery stratcgics mny not bc as
,ractical with managerial bonus programs (c.g.. climinntlng the influence 0:
:xtcrnal cnvironmcntal fluctuations). Still, howwcr. the logic of rclniorclng I:W

:ognitions and behaviors \vith extrinsic rewards applies to managcrlnl personnel.

:or example. before Whirlpool’s SBU rcorganialtion, the bonus system rcwardecl
nanagcrs for the pwformance of the entire corporation. With the SBU rcorganG.a-
Ion. bonuses were based on the performance of the SBU. Thus. the receipt of the
XXIUS was contingent cm the p&*ormance of ;I manager’s SBU and ail manapcrs
wzre treated equitnbly. based on the performance of their SBU.

Training anti dtwlopmcnf. To assist in the qulsition of new kno\vlcdge.

ikills, and abilities (KSAs) for recently changed jobs. an organization WI imtiate
ix~~scd traming programs. Parsons, I .idcn, O’Connor, and Nagao ( 199 1) found that
raining in the use of personal computers. when perceived as imrnrrrwr~ral. contrib-
.~ted to institutionalizing the new behavior. El Shcrit‘ ( 1990) rcportsd that training
NU ;i dynamic component in accclcratmy the institutionalization of dwision
*upport systems. Furthermore. Butler and McEvoy (1089) found that training
:ontribut& tcj the institutionalization of a new performance appraisal system. Also,
:;ocximan and Dean (1982) found that training was a significant factor in institu-
ionalizmg changes brought about through Quality of Work Life programs.
116 ACHILLES AKMENAKIS, STANLEY HARKIS, and HUBEKT fElLI

h’ot only is training considered to be significant in institutionalizing change but

the rrtriners can have an added influence on the outcome. Bandura (1986) proposed
the use of respected peers as trainers or facilitators bcwause of the symbolic so&l
dynamics. A recent example illustrates how Coopers & Lybrand trained a cadre of
Allied-Signal employlzs to conduct total quality management (TQM) workshops
for all other employees (Stewart. 1992). This tactic is symbolically different from
bringing in external trainers; this is more akin to respect& peers spreading their
influence through the organization, encouraging peers to adopt and to institution-
alize the TQM philosophy.
Training and development, if linked to an organizational change, can reintbrce
all message components. First, the superiority of the new way should bc obvious.
Thus, need for change and the appropriateness of the change can be sensed. Second,
the trainee should experience efficacy in performing the new job. because the KSAs
will bc rolatcd to thr: tasks. Organizational support should be apparsnt through the
cxpcnditure of’ funds and by permitting the training to be dons on company timr.
Peer support can bc transmitted by using colleagues of the participants as trainers.
Finally, valence can bc crcatcd by associating tangible and intangible rewards to
the successful performance of the new job.
While skills training is focused more on the individual KSAs, team building
capitalizes on the social dynamics ofgroups. Through team building, organil&)nal
member commitment to the change effort is influenced. Respected peers influence
team mernbcrs to increase their commitment by identifying with the team and
internalizing the values of the change effort.

Dit’iusion Pr,wficc.s

Diffusion is spreading change adoption within one organizational group as well

as spreading the adoption IO other organizational groups. It is a common practice
IO establish a pilot program for experimental purposes to test an innovation. If test
results arc positive, then the innovation may bc shared with potential adopters.
Although some customization of the innovation may be necessary. the originating
unit serves as a valuable resource in demonstrating the benefits, forewarning of
adoption barriers. and advising about mcthodologics to facilitate implementation.
Thus. the diffusion practice is an attempt IO establish some kind of organizational
dialogue whersby organizational rncmbcrs lrrrrn IO r&k. buik the rrllk. and .s~~.src~in
rhe Itdk end nulk (Ashkenas & Jick. 1992).
Research by Johnston and Leendcrs (1990) on the diffusion of minor technical
improvements in q>crations revealed that diffusion was accomplished through an
interpersonal exchange between the potential user and the originating unit. Blakcly,
Emshoff, and Roitman (I 983) identified site visits and diffusion facilitators as
sigmficant factors in institutionalization. Stewart (1991) reported that GE’s suc-
ccssful Exchange of Best Practices program consists of prcsontations made by
respcctcd peers.
 Other means of organizational sponsorship, such as providing research funding
for the diffusion of an innovation, have also been shoivn IO br: cffectivc. Gannis
(,1087) reported that financial support was a major factor for the success of !Merck
8.1Company, the pharmaceutical firm. in changing its research methodology for the
invention of new drugs. ‘lhe awarding of such funds, especially if handled in a
ceremonial manner. may enhance the desire to achieve the high status of the
awardees. Thus. the new practice can be diffused throughout the organization.
Bushe and Shani (199 1) have described the use of a parallel learning structure
(PLS; also referred IO as a transition (cam, collateral orpani/;ltion, and parallel
organization), as a vehicle for diffusing change. A PLS can connect and balance all
the pieces in a change effort (cf. Duck, 1993). In addition to being a diffusion
strategy, a PLS is a form of active participation which develops commitment among
organizational members andcnhances institutionaliTA(ion. A PLS typically consists
of a steering team (for overall direction) and smaller teams representing the various
diffcrcntiated groups resembling the overall organizational membership. The oper-
atmg procedures of a PLS are intended to promote change by representing the
various interests and by assisting in customi&q the change to that organizAtional
group. Duck (1993) emphasized that someone on the PLS should pay particular
attention to the emotional and behavioral issues that can and should surface in
conjunction with the orgamzational change. Consequently. an organiitational
change can be tailored to be more compatible with the organizational networks,
thus facilitating diffusion.
Scss and Cucuzza (l(W5) provided an example of how Chrysler used a PLS in
diftuGnp activity-based costing (ABC) throughout its organization. A PLS w;ts
asscmblcd consisting of 20 employees from the finance, manufacturing. enginecr-
iny. and information systems. Three critical steps were executed in the diffusion
process. First. the PLS persuaded critical opinion leaders to give ABC a fair shake
and ultimately embrace the program. Second, the PLS coordinated training pro-
grams at all levels dealing with principles and mechanics of ABC. Finally, the PLS
set up a pilot program at one plant and then rolled out the program throughout the
company making sure local managers were involved and that there were visible
successes(Ness clcCucum~a, 1995).
USC of diffusion practices is valuable because it effectively transmits the five
message components. The benefits revealed from pilot programs or vicarious
experiences from site visits can produce sufficient evidence that a change is needed
(i.e., there is a better way). lhe actions by the change agent to sponsor the site visits
or coordinate an exchange of best practices program symbolizes support for the
orpanir..ational change. Customizing the innovation, practice. or organizational
change to the uniqucncsscs in the adopting organization addresses the appropriate-
ness issue. Knowing that the organizAtiona change is being used by others and that
It may be customized for ItKal use can create efficacy. Finally. vdtmx is created
through some extrinsic reward (e.g., monetary compensation or social recognition)
for adopting the organization change: furthermore, the realization that others are
II8 ACHILLES ARMENAKIS, STANLEY HARRIS, and HURERT FEILD

adopting the change may extend the adoption .and ultimately rrsult in institution-
alization.

Rites and Ceremonies

As noted above, symbolism is an important outcome of workplace activities and

is the eq>res.~i\~meaning associated with organizational structures, hchaviors, and
processes (for a thorough treatment see I’rice & Hcycr, 1993). Rites and ceremonies
are considered symbolic practices evident in all organizations. Rites arc’ public
performances that signify the conscquenccs of actions and the extent to which these
actions are consistent with an organicltion’s valurs; ceremonies connect two or
more rites mto a sinplr: occasion (‘I‘ricc & Bcyer. 1993).
Rites and ceremonies are powerful shapers of underlying cultural values (Harris
& Sutton, 1986; ‘I’rice & Beyer, 1993) and powerful sources of the key change
message. Several change researchers have emphasized the role of fcVlirrg go r$lhr
pusr in helping organi/;ltional mcmbcrs adopt and institutionalize change (cf.
I3ridgcs. 1091; Confintwrion Spwt-h, 1988). The transition to a new state is
facilitated by accepting the fact that the old is gone forever. Retirement ceremonies
arc intended to provide an opportunity to summarize one’s accomplishments and
contributions to the organization. as a way of expressing gratitude to the rrtirec and
officially ending one’s formal association with the organization. However. these
ceremonies also focus on the life thereafter, thr new challenges, the application 01
accumulated experience and skills to the many new cndcavors that continue to make
lift fulfilling. Consistent with the notion of letting go of the past, Harris and Sutkm
(1986) dcscribcd parting ceremonies in public- and private-sector organi/;ltions that
were riving. ‘Ihcsc were formally planned functions where employees mourned the
passink of the old; a kind of cathartic experience which was intended to confirm an
end to a rewarding worthwhile cxpcricncc. During thcsc ccrcmonics, participants
shared in a final feast and exchanged stories of the good times and perhaps
cxpressinp sadness and anger at the organization’s death. However, the one indis-
putablc fact was that it was over and it was time to move on. Thus. parting
ccrcmonics can bc useful in adopting and institutionalizing the ~qv organiAonal
change.
Rites of cnhanccmcnt arc intended to reinforce individuals and groups for
adopting new ideas, values, and behaviors. ‘ILnstall (1985) dcscribcd AT&T’s
efforts (during the divestiture transition) to rccognizc and reward individuals for
exemplary performance by establishing the Eagle award for IIL'W marketing ideas
and the <hldsn Boy award for outstanding customer service.
Another rite of cnhancemcnt WLS Whirlpool’s announcement of the joint vcnturc
with N.V. Philips (of the Ncthcrlands). In August 1988, Whirlpool traders quickly
scheduled ;I ~neeting of the offlicers, directors, and managers to formally ~IIII~~IKX
the Joint vcnturc. This joint venture made Whirlpool the world’s largest home
;~ppliancc manufacturer. Hadpes \vith “We’re #l” were distrihutcd to all aHendcc\.
‘I’hc atmosphcrc of the meeting was one of cxcitemcnt and celebration.
This meeting was effecrivc in transmitting the tivc message componcn% thcsc
incxpcnsivc badges scr\ cd as a symbol rcprcscnting the ou~~mc' of the negotiations
with Philips and in reinforcing the appropriateness of organiwtionA changes
recently cxpcricncctd by these c~~q-~loyecs. ‘I‘hc audicncc H’S thrn briefed (for
approxim;ltcly 60 minutes) on the strategic logic employed in seeking the joint
~cnturc with Phtlips. This accomplishmrnt reinforced that the recent strategic and
structural changes were necessary. The pride of being cmploycd by the world’s
largest appliance: manufacturer was an intrinsically valcnt sentiment that also
contributed to the employees’ cfiicxy. Support from the lcadcrs was apparent from
scheduling the meeting and the strategic briefing that followed. The jubilation and
excitement of the officers, directors, and managers during and after the mcctinp
were evidence of the support.
While symbols like the Whirlpool badges may seem insignificant and by thcm-
.sclvcsmay not have the effect of more costly awards, they demonstrate the extent
to which SOIW organi?Aons will go to influenceemotions regarding organiwtionA
change. Furthermore. kvhile the cost of these badges was nominal. over time,
numerous low-cost symbols can have a cumulative effect.
Because symbolism is in virtually all actions (as well as in nonncfions), scvrral
btratcgies/tactics explained above qualify ;1.srites. Kotablc arc, for example. Merck
and Cornpay’s awards for adopting a new pharmaceutical research mcthtxiologj
(i.e.. rite of creation), the dramaturgy video prepared by Whiripoool’s lcadcrs IO
cslcbratc the positive consqucnces of the rcorpanization (i.e., rite of transi[ion).
and Whirlpool’s CEO delivering his Cor~jirm~r~on S/NY~/I that included the phrase
“there is no turning back” (i.e., rite of parting).

In practically all organizational change efforts. there are concomitant chrmpcs in

formal activities ~wcessq to demonstrate emphatic support for the changes being
implcmcntcd which may play a major role in the overall success of the change.
Generally, such formal changes in structure and procedures arc’ substnnti;ll and
difficult. They are unlikely to be undertaken unless the orgmi~tion wholeheartedly
supports the organizational change. For example, in 1988, Whirlpool initiated a
fundamental change in business strategy by changing from a low-cost le;tder to a
product differentiation strategy. Concomitat with this change in stratcgp was ;i
change from a functional to adivisional (i.e., strategic business unit) organizational
structure. To support the new structure, rcviscd job descriptions, policies. procc-
durcs and other formal job requirements were necessary.
Other initiatives like IS0 9000 (Uzumeri, 1997) and recnginecring (Hammer &
Champy, 199.7; Vansina & Taillieu, 1996) arc implcmcntcd concomitantly tochangc
the formal activities. and thus change the way an organization conducts its business.
1 20 ACHILLES ARMEYAKIS, STANLEY HARRIS, and HUUERT FEILD

These formal activities naturally require new behaviors and ;1reintended to support
the organizAonal change. By applying the tactics described within the other
strategies, change agents can facilitate the adoption and institutionalirAon of thcsr
new behaviors and send ;1 powerful message regarding management’s support of
the change.

USC of the Model for Planning Organizational Change

‘Ihe proccessof planning and implementing change obviously requires a great

deal of time. effort, money. rind patience. Summarized in Table I are the major
questions that should be answered regarding the institutionalization of ;1 change
effort. We realize it is common practice for an organization to have multiple changes

Table 1. Checklist for Institutionalizing Change

--
. WhJl b.,cI of CummiImcnI 15 appropridlc fur Ihi org,3ntrdIic~nAl c hJngc! I5 lhi5 Crlmmli-
men1 Ic~v~I con51jIcw1 w11h Ihc lime dlrnenslon nc~‘ewry for lhls c.hJnge!
. 110 WL‘ have rc~hwr ~115wr5 to lhc five* c~tws~~~ms UKUI IO icxu5 lhc. rhqc: rnc~~gct!
. Arc lhc rhangc Jgen15 Jnd ItK’tr rcpreentalivc~ c~rrul~blc!
. Cdn wc jy3IcmJIlc’dtly cJlq+c 0rgJnizaIronat merntwtrs into groups! I)0 wc linow
lhc oplnic>n Ic;ldcrs wilhln each group! How will rhcy JS~CS the change effort! (1.m \se
JnIicipJIc their reaction!
. Dcm lhi5 orgJnr7JlronJt c-hangc lend rtseli IO J p;lrtiClpaIirr Jpprc~Jc-h? Who do WC’ In-
volvc! Cari lhcs ov~rJIl change ciforl tr dlvidd into componc~nls, thus pcrmllling se-
qucnc,ing rhc componcnrs, Jnd CJpilJlizing on lhc conccpl oi (*naClivc mJslcry! t-low
Can we inCorlmr.rtc VI~J~IOUS expcricncc% IO iJCil&a? Jnd cxrend lhc ,ldopIion ui the
rhangc cifm!
. How should we Incorporate lhc vJrlou5 Ixfirs ie.g., II\T preenlalions, \vrilIcn cc~nlmu-
nic.alion5, &~~lroniC IrJnsmissions) of persuasive c~ummuniCJIion? t-low frqucanlty
should lhc lJtliC5 tkz u~d! Arc some of lhe IJctlrs more .lpproprl.llc ior um JUCJI-
CNd
. WhJl mIr*nJl mitxrrurlon cJn tx used lo Ir.msmiI 1hc rrx”jsJgc! WhJl exlcrnal iniorrnd-
Ilun cJn tx us& lo IrJnsmlI Ihc ma5Jge!
. Rqxding HRM pr.rcIltcr:
l How Cdn wc ixe~uIc lhc 5CtcYllon I~clic. IO communir.aIc Itlc IIK’sWgC!
l Is our pcvformanc~c? JppfdisJt pqrJn1 in synch with Ihc OrgJnizarionJt change!
. Arca ~(8 assc?,s~rrg Ihc JpprJisee in Icmis oi their commilmcnl lo lhc change pro-
gr;1m?
. DCKT our compcntir~on prq,rJm supporr lhc orgJnrr,rrtonat c.h.wy,r!
l WhJI dlCJIk~lJl progrJm5 Jre needed Jnd who CJn ronducl Ihcm nx)sI diet.-
Ilvc*ly!
. C‘.an WC bcneii1 irum c-AAt5hing J paratlct IcJrning s~ruc.turc! LVho should Ihc rc*prc-
senI.lIrve5 t*c!
WhJI 5ymtxAir acrlons/c~vcnls CJll wc cxc’culc’ IO IrJn5miI 1he rnc%,lglgc9
What form.lt JClivillL7 JW rWCe55Uy IO sup(w)rt the o~Jrll.ZJllOnJt ChJngd
tiow are we goinK 10 ~5x35 ItW tcvet of cc~mmiln~c~nI mwng wr enlptoyWs!
Does this org:anirationJI Change reemblc J profi’.‘r”-oi-rhc-mc,nrll!
 wing implemented simultaneously. By planning each change, using the model as
1 guide, a change agent can apprcciatc the magnitude of the effort required for
Implementation. Consequently. it should be apparent that c-hnnS~c overlotd can be
txperienccd quite quickly. Thus. a change agent can more realistically anticipate
the challenges faced in implementing changes successfully. In situations requiring
implementation of multiple changes at least two alternatives are appmxit. One
alternative may be to coordinate the various projects and USCthe strategies and
tactics common to more than one project to facilitate institutionalization. Another
alternative may be to prioritiise the change projects and initiate a plan to implement
those that arc most critical early on and those that are less critical subsequently. In
either case, by conscientiously answering these questions, change agents can
appreciate the magnitude of implementing organizational change. Furthermore, the
symbolic effect of using multiple strategies should be that organizational members
will realize that the change program is not simply another program-of-the-month.

Commitment dnd Asscssmcnt

The assessment of any organizational change effort is intcndrti to determine the

extent to which the objectives have been accomplished. Understandably. a change:
implemented in response to unacceptable levels of productivity or protitabilit>
would be ultimately evaluated in those terms. However, early researchers (cf.
Georgopoulos & Tanncnhaum, 1957; Likcrt, 1967) established the link bet\vcen
managerial actions and behavior (i.e., causal variables); pcrccptions, attitudes, and
satisfaction (i.e., intervening variables); and productivity. scrap. and employee
absenteeism and turnover (i.e., end-result variables; Likcrt. 1967). Relying solely
on assessment criteria. such as productivity and profitability, may not accurately
portray the existing commitment level. The asscssmcnt of commitment can be used
to predict future trends in end-result criteria. In other words, commitment may be
a leading organizational indicator of productivity. That IS,intcrnaii~cdcommitme:nt
is more informative about institutionalization than simply reporting prtductivit)
has increased. An increase in productivity could be associated with a higher Icvcl
of compliance commitment. Thus, assessing the effectiveness with which change
stratcgics are executed and received should bc based on level of commitment and
selected end-result variables that are relevant to the specific organi?Aon change
effort.
The research on commitment that we used in developing the model provides
useful detail is ‘assessingprogress toward institutionalization. This conceptualiza-
tion of commitment is relevant to the theory, rcscarch. and practice of organizAona1
change because it permits a change agent to establish objectives rclatcd to each
level. In addition, it serves as a barometer of the extent to which institutionalization
has been achieved. For example. a change agent may elect to establish time horizons
for achieving each level of commitment. In some situations, particularly those
involving business turnarounds. a change agent may attempt to accomplish only
122 ACHILLES ARaMENAKIS, SrAhLEY HARRIS, and HUUER’T fEll.1)

compliance-b;lsed commitment. Howcvcr. for the long-term the goal might bc

internalization commitment. hloreovcr, some of the strategies and tactics arc uscfuI
in short-tcmm situations while others may bc expected to produce changes over the
longer term. For example, formalization activities may influence one’s behavior in
the short term because org:miAonal members will bc required to comply with
new job duties. However, over time, other tactics, such as vicarious Icarnmg. may
rquire the longer term perspective.
The research provided by Becker (1992) has demonstrated that WCcan refine the
assessment of commitment into overall compliance commitment and identification
and internalization commitment as each applies to the organization, an orgamza-
tional memba’s supervisor, and work group. Furthermore, the other research (cf.
Gcorgs, IY90; Kcichcrs, IYM) implies that we could extend the measurement of
commitment to the change agent and the change initiative. Thus, it may bc possible
to ;ISSC~S one’s commitment (i.c.. idcntilic;ltion and internalization) to each entity.
namely. the organization. the change agent. the change initmtivc, the work group,
and the supervisor. A comparison of thr various Icvcls of commitment to each emit!
(e.g., the change initiative with the supervisor) can indicatr whether organizational
members perccive their supervisor as supporting the change initiative. This added
refincmcnt in mer~surement can assist in pinpointing where additional rcsourccs
might be needed in institutionAlzing change.
In addition to assessingcommitment. the extent to which organizational members
have heurd and heliere the tivc core message components should be assessed.If an
organizational change effort is not having the intended effects on commitment. then
it might be valuable to know whether one or more message components has been
ignored or lacks believability. Through this part of assessment, a diagnosis can be
initiated to identify why the message lacks believability and what actions need to
be taken to remedy the situation. Each clement in the mode1 must bc analyzed. For
example. (a) is there something unacceptable about the message componsnts. (b)
do some groups have different beliefs about the message. (c) dv the change agents
lack crsdiblhty, and (d) are the strategies being effectively cxccuted’!
The methodological issues related to assessment are beyond the scope of this
paper. IIowcvcr, one must recognize that assessment of change offers some intcr-
csting methodological challenges (cf. Armrnakis. 19X8: Svyantck, O’Connell, &
Baumgardncr, 1002; Woodman, 1989).

CONCLUSIONS AND IMPLICATIONS

Organizational change efforts comprise corrtrt~r and pmcess issues. Content issues
are related to \$41rrris to bc changed. Changes in business stratrgy and organizational
restructuring and other formalicltion activities arc example.. of content is$ucs.
Thcsc may bc the required changes that must be embraced and effectively imple-
mented by organizational members in order for the organization to improve its
 xrformancc. On the other hand, process issues deal with /IC)H.the content issues
IX implemented. We proposed and cxplaincd a comprehensive process model
lscful in institutionalizing organizational changes. Although the model is primarily
<;cuscd on process issues, we do not want to undcremphasizc the required changes.
3ur m:cnt in developing the model was to empha?;ize that a change agent must be
<nowlcdgeablc of horh the content and process of change.
Each construct in the model (see Figure I) should become a conscious and
:xplicit step in planning, implementing, and assessing organizational change. We
4cctcd commitment because we feel il is a natural outcome measure for institu-
tionalization. Most people seek to bc committed, in terms of psychological attach-
ment. to some entity in order to satisfy a set of needs. Some interesting findings
regarding loyalty and commitment have rcvcalcd that bcwause of the organizational
responses (e.3.. mcrgcrs and acquisitions, rccnginecring, downsizing) to external
environmental changes, cmployecs arc finding less reason to be committed to their
employer than to some professional or trade organization (cf. Rousseau, 1997: and
this trend is worldwide; set Kanter. 1991). WC think this IS unfortunate and
unncvcssary and thus made commitment a ccntcrpicce in the model.
In realizing this trend in commitment, change agents should build and peri-
odically assesscommitment. Not only should change agents be concerned about
losing good employees who have significant investment value and who can con-
tribute to an 0rganiAon’s succclrs. they should also be concerned about the
commitment lcvcl of all employees.
The recent research on commitment has initiated some interesting debates.
Scvcral researchers (cf. Becker, 1992: Becker. Billings, Evclcth. & Gilbert, 1996;
Harris, Hirschfcld, Fcild, & hlocsholdcr, 1993; Meyer & Allen, 1997; O’Rcilly,
Chatman, & Caldwcll, 1991) have refined and extended Kelman’s (1958) original
conceptualization. While differences in the typologics exist, they are similar and
compatible (Meyer & Allen, 1997). Although we used Becker’s extension of
Kelman. other conceptualizations can certainly be used with the model.
Change agent attributes were included in the mcxidclto emphasize not only the
importance of the driving force behind the change but the importance of those
rcprcscntativcs (e.g., managers and rcqcyted pers) who may act as change agent
extensions in diffusing change throughout the organiation. It was argued that
change agents must bc pcrcsivcd as cr&ible if institutionaliwtion is to occur. The
belIcvability of the message is related to the credibility of the change agent.
The value in including attributes of organiz%ional members in the model IS to
create awareness that it is unlikely that change will he institutionali& homogcnc-
ously throughout the organization. Understanding the role of individual dii??rcncrs
and organizAtiona cculogy is irnlx)rtant in appreciating the challenges change
agents face in institutionalizing change. Change agents should be familiar enough
with organir;rtional members to systematically catcgorirz individuals into one or
more groups. Furthermore. it is useful to bc able to identify the opinion leaders and
to anticipate their assessment of and reaction to the change. RtTognizing thcsc
124 ACHILLES ARMENAKIS, STANLEY HARRIS, and HUBERT fElLI

attributes and planning the implcmcntation accordingly will enhance the prob-
ability of success. In order IO establish positive momentum for the change and
minimiire the likelihood of a change initiative being cynically labeled a program-
of-the-month, organizational members nec4 IO be converted IO change agents.
The social distance that exists between the change agent’s actions and organiza-
tional members is a major issue that must be dealt with in any organizational change.
A CEO can initiate a change program, but there should be an explicit endorsement
at all levels of the organiir;ltion; otherwise, the change &fort will not succeed. Those
in leadership positions should be considered credible. should endorse the change.
and should mimic the behavior of the change agents. AI the local level these leaders
should be expected IO use the strategies in the m&l and apply them in ways that
are appropriate for the level. Thus. all strategies in the model should be used at the
macro- and micro-levels.
All strategies and tactics should be explicitly linked to the organi;t;itional change
effort and should bc considered to be complementary. In terms of linking tactics to
the change effort, the human resource management practices should dcmonstratc
support. For example, criteria rclatcd IO the change effort may be included in the
existing performance appraisal program; a multisource feedback program may bc
initiated and aimed at supporting the change effort. Similarly, the compensation
package must reinforce the change effort.
The factors in the model can be expected IO have interacrivc relationships. Thus,
a change agent who was not considered IO bc credible could negate the potential
positive effects of the strategies and tactics. WC cannot identify which factors in the
model are the most significant in the institutionalization process. Intuitively, WC
believe that those factors based on the concept of self-discovery have more impact
than those that are directly from the change agent. However, active participation
cannot be cxpcctcd IO be the sole ingrdient in the change process, although it is
significant. The other strategies should be considered as supportive and used as
appropriate IO institutionalize change. Until rcscarch can identify which are most
effective, we argue that all should be coordinated and applied in all organi&onal
change efforts. We prefer IO think about the model in terms of an athletic metaphor
and argue that omitting any element in the model is tantamount IO fielding a team
without all the playcn. Thus. we advocate USCof the contplele model in the adoption
and institutionalization of change.

REFERENCES
Armcnakis. A., HUT&. S.. 8: !vlosshol&r. K. (1993). Creating rerdincss for organifalional change
Iluman Relu~icar. 4t5(3). l-23.
Ashkens. R.. & Jick, T. (lYY2). From dialogue IO action in GE workout: Dcvclopmcnlal lcaming in a
change process. In W. Psmorc & R. Woodman (Us.). Rtscnrrh in orguni:uliunul chunge und
drvelopmenr (Vol. 6. pp. 267..287). Greenwich. CT: JAI Press.
Baba. hf. (1995). The cultural ecology of the corporation: Explaming divcrsiry in work group rcqxmss
to organizational transformalion. Ioumuf of Applwd &huvioruf Science. J/(l). 202-233.
Bandwa. A. (lYt46,) So&f foumfolionr of I/IOU.~/II und uclicnr: A so&l c-qniftre fhroq. Fnglc\v~rti
Cliffs. Sl: Prcmicc-Hall.
Becker. T. (IW2). l%ci and baws of commltmunr: Arc [hey dislincnon woflh malimg’? Acx&rny oi
Manugcmcnr Joumul. 35.232~25J.
Rocker. T.. Billings. R.. Evclclh, D.. & Gilben. N. (lYY6). Foci and bases of cmploycc commitmcnr:
Implicarions for job ~xrformancc. Acudcm~ ofMunugcmen~ Jwmul. 3Y. W-M.
Box. M. (lY76). The ~cchnology of organi7ation development. In M. Dunnc~tc (Ed.). H~uI&& ol
irt&srriof und oarmi~ficmcllps?u-hob~~ (pp. 937-993). Chicago, IL: Rand McNally.
Beer, M.. t%cnstat. R.. Rc Spector. H. (1990. November-Drxcmbcr). Why change programs don’1
prrducc change. Hunonf Buriness Rcvinv. 68. I St& 166.
Bits, R.J. (1987). The predicament of injuslxc: The managcnxnt of moral outrage. In L. Cummings CL:
IX Staw (E&.). Xrscurri~ in oqu~rixrfiomrl bchuvior (Vol. Y. pp. 289-310). Grwnwich. CT: JAI
Plcss.
Blnkcly, C.. Emshoff. J., & Roilman. I>. (1984). Implcrncntmg innovative programs in public sector
organizations. In Oskamp. S. (Ed.). Applirdsucialps)uho[~.~~ unnuul: Applic-orions in oquni:lr.
rionrrl scffings (Vol. 5. pp. 87 109). Beverly Hills. CA: Sage.
I3rc1z. R.. Milkovich. G.. & Read. W. (lYY2). Tbc cvmm ztalc of performance appraisal rcscrrch and
practice: Concerns. directions, amJ implications. Ycur!\ Xcvtcw o/&nqqrmcnr. I??(2). 321-352.
Bridges. W. (1991). MuncqinS rmaritionr: ,%fuking fhe mcxr ujchungr. Reading. MA: Add&on-Wcslcy.
l3mckncr~ J., Groxr. S.. Recd. T.. DcWirr, R.. 8: O’Mnllcy, hi. (1987). Survivon’ reactions 10 IayofTs.
We gel by wilh a little help ftx our fricnrk. Adntinisrrutiw Scirnse (&mer/,v. 32(J). 5X-!LI I.
Huller. P., & McEvoy, G. (19119). I)crcrminanls of Ihc imtiturionalizalion of planned organizarlonzd
change. Gmup & Oqvukurion Sludics. /J( 1). 33 -50.
Hullock. R.. % Tubbs. M. (IYYO). A tax mela-analysis of gainphanng plnm ircorpani~~~ion Jevclopmcnr
intcncnrions. Joumnl uf Applied fi’churiorul Science. 26. 3X3-4&8
Burkhardt. hl. ( IYY 1. August). INrifurronulizrrrif~n~~l/n,vlnR u rr~hnolqgrccll rhun.ry. Paper prcrcmcd
a1 dre Fifly-firs1 Annual Academy of Managemen Mating. Miami. 1.1..
Bushc. G., & Shsni. A. (IYY I) Punrlkl leomrng srru(-fun’.s: lncrrrsin~ inntw~Iwn in huwuu~ruc (CF.
Rcdmg. MA: Ad&on-W&y.
Ruriness in PrufiL. (19%)). (Vidcolspc]. Hcnlon Harbor. Ml: Whirlpool Corpor~lion
Church, A.. & Burke. W. (IYYS). Pradtioncr artiUcs aboo dr fxld of organirauon dcvclopmcnt. In
W. Pasmorc % R. Woodman (Ws.). Rrscurch in organia~rionul rhuqe and drwkymenf (Vol 8.
pp. t -46). Greenwich. CT: JAI Pnxs.
Cobb, A.. Wootm. K.. & Folgcr. R. (1995). Justia in ttx making: Toward undcrs~anding Ihe thcxxy and
praaicc of jusricc in orgamz~tional change and bevclopmcnt. In W. Pasmorc & R. Wocdmm
(I&.), Resron+r in orgunizufional chuqr CUK! drvclopmenf (Vol. 8. pp. 243-29s). Gruznwich.
m JAI Press.
Confimu~rion Sptrrh. (1088). [Vidcotarxj. Bcnton Harbor, Ml: Whir@4 Coqx>rahon.
Dalron. G. (1970). lnflucnce and organizational change. In G. Dalton. P. L;rwrcncc. 8r 1.. Grcirlcr (IUS. j.
Oryniznitmul ~hungc und dticlupmtnr (pp. 230-1,(H). Homcwood. IL: Richard D. lnvin and
Tlw Ihscy pnrss.
Damanpour, I:. (1091). Organizalional rmvauon: A fnera-analysis of ctTccls of dclcnmnann and
modcralors. At-r&my of.&fonqrnunr Joumul. 34(3). 55s -590.
 ACHILLES ARMEMKIS, STAQLEY HAKRIS, and HlIHEKT FEILD

lkcls. N.. k Tyler. 11 (lY86). Ilow Xcron impro\cd il pcrform3ncc appraisal<. I’rr.ronnelI~iwnd. 65(JJ.
so- s2.
Dirkz. K.. Cummmps. L.. & Plerr-c. 1. (19%) Psychologd owncnhip m org3ruzalwn\’ (‘ontlitloru
undur which indivdu31s prunhz4s and rwlsl ctimge. In R. Wtxxhnnn & W. f’.wmorc cFJd\.).
Neseun h tn oqwnr~~~~onrd thun~e und do ehpmrrtr (Vol. 9, pp. I -24). Grwnu ich. Cl. JAI I’m\\.
IX.

Dowick. f? ( I99 I ) flrwi(-ul guide fo u.~in$ wd~o in rftc br/urwontl .rcwncc~. New York. John H’ilcy &
Sow.
Duck, 1. (IY93). htan3ging change: ‘17~ art of tduwnp. flur-wnf Huswe.r.c Neb inv. 7/(6). I 0~ I 1%
Ewrihcrgcr. K.. fGol0. I?. Sr Davts-I~hla\lru. V. (1990). Pcrccivcd org3mral~on3l supporl 3rd
cmploycc dillgrxc. commitmcnr. 3rd iruwvalwn. JOV~CJ/ o/Apf~hcd P~yfw/q~. 73 1). 5 I - SY.
lil Shsnf. It. (IYYO). M3n3ging in~lilul~onal~zal~orl of slralcgic &&ion support for IIE iigyph;cn cabmcl.
/nfer$Jc-rs. 20(I). 97-I I4
Fark~s, C., & Wcllaufcr. S. (IYY6). l’hc ways chief cxcculivc officers lc3d fkwrd Hushsr Kn-irv..
743). I10 122.

Ilarnmcr. 51 . & Ch3mpy. J (IYY3). Hecnguwering rhr cwporunon. Srw York: fiarpcr Rurincw
fhms. hl. ( IY94). Ralcr mouv3wn m 11s pcrformancc appraisal conlcxl A lhcorclid fr3mcuwk
Joumul qf ,Hunc~.~rmenr. 2(&J). 737-7.56.
Harris. S . Ilirschfcld. ‘I:. I+dd. 11.. & MossMdcr. K. (1993). I’cycholog~cal nllachnxznl: Rcl;l~wnshqn
wlh JOT char3cIcnsl~cs. 31111udc’s. 3rd prcfcnwcs for ncwcomcr tkvclopmcn~. C‘m;rp r~rd
Oqpttwioft Munrrgemmr. JR. 4SY 48 I.

Kclm3n. I{. (105X). C‘ornpli;ulcc. idenlificalwn. awl inlcnnhzal~tm JoumrrfofCon/hr rHr.wlurion. 3 I ).

Sl-60
Kilhxm+c. I... O’Irary-Kcllcy. A.. & W~lhams. S. (IYYf)). Empluyrr pcrccpliorfi of f3irw:rs stun
humw rcwurw sy~rcrnr change ‘lhc c&w of cmplo~cc hyoff~ In R. Hixdm;lr~ R: W l’3.w~orc
(Ed\.). Kc.sc.irrh m orpamr3Ilon3l change nrd dc~cloprncnl (Vd Y. pp 25 -48). Grwriuich. CT:
JAI Ph.\\. Inc.
Kinon. %1. (198‘1). Adapron and mnovators: \+I) I)C’U im1ia1ivcs gc’ blocked. Irmg Htmge I’!~~mn~y.
17(Z). 137 .I-l3.
Kourcs. J.. & Posncr. B. ( 1997). C’redibibry: Ifow budera .ruin curd /we 11. N+I> per@ demcmd if. San
IGciscico. CA: Jorsy. Has.
IA% ‘I’.. k La&n. S. (1996) Reaching and changq fronrhnc cmployccs Ilwwnl Iluzines.r Kevrew.
743). OS- 10-l.
Lcpsmgcr. R.. h Lucia, A. ( 19Y7). The (WI rrml science o/.VW/erd~crt:k. San Frwcisc-o. (.‘A: Pfalfcr.
L.&n. K. ( 1947). Fwl1icrs in group dynamirz. Ilumcm Rcfu~icw. 1. S 4 I.
Likcrr. R. (1967j. 7%e /turncut orq~~~~fron. SW Sork: 5lcCirw-lfill.
Locke. E.. & L3rham. G. (199Oj. A rlw~~ ~f~wl serfin!: and rclck perjormruw linplcwood Cliffs. SJ
I’rcnIicc-Hall
London. 51.. h f)sarry. R. (1993) 360~d-cgrcc fccdhack ;CI z compcrit~\c advantage. ffumnn R~.wurc-~~
.\frmrr.pvnenf, .+Z($ ci: 3). .lS3-372
hlardonald. S. I IYYS). Lcarnmg lo clwngc. An inform;ruon f~rspccl~~c on Icarning in 0s or~an~rar~on
Orpu;rIron St-irnc-e. MS). 557-568.
M.II~ICU. J.. Sr Kohlcr. S. (1990). A cnn\-lcvcl cxanrinabon of group ;~bwncc ~nllucrws WI irvhvitlu;~l
atwncc. /ow71ul ofApplied A,vc-hrdoXy 7.5. Z 17-1-N
XIcycr. J , & Allen. N. (1997). Comnurmenf In f/w wrkpkrr~e: 77wq r~scun~h. orrd crpplicuf~urt.
Thousand Oaks. CA. Sage.
hlorray. 13.. & Gxhart. 13. (199X). An crnplrical analysts of a skill-bawd pa) pn’gram and plant
pcrfomn~ncc ouIccw.13. Acrufem~ oj.Wrmc~p~mrnl /0umd. 4/( I I. 6X-7X.
Nadlcr. r) (1977j. f~irdtark turf oyoni5rJion tfevelr~pmenr: Lknp cfurctJxrsed mrrhdv. Rcadmg. Xl,\
Addlu’n-WC\lC>.
Scss. J.. & C’ucurz~. T (19%) Tapping dr full lwcnunl of AM: ffnnd I~rc.si~~css Rexrew. ?.i(4).
l30..13X.
Neumann. J. (1989). Why pcoplc dun'1 pnicipatc in qani/Awnal change In R. Wowlrna~~ h \V.
Pwnorc (PA 1. Hesecan-Jr in or~cmi:ufiond c-hanxe wxf de’c*hpnrmJ (Vol. 3. pp. I H I .-? I?).
<;rccnwich. CT JAI.
Nichoff, H.. t%z. C.. Kc Grwcr. R. (1990). The impact of lop msnagcmcn1 actions on cmploycc 11111uch
and pxccpiofs. G‘mup & Orqmi~arion Sludirti. /J(3). 337- 351.
NUN. P. (1986). Tnc~ics of implcmcnralion. Acucfrmy of&nu~emmr Iwmd. 20(Z). ?30-201.
Nysznm~. P. (IWO). Vc:crtical cxchangcs and oganwA~nal cornmltm+cnl\ cf hmcncan t)u\incs, m3nap-
crs. Crwp & Oryanizufallrm Sltafics. f 5(3). 206-3 I?..
O’Rc~lly. c’.. C%urman. J.. 8; Caldwcll. I). (1991). I’coplc and orgarur;trionaI cul~urc: h pmfilc
comparison approach IO asccwnp pcnon-organisalwn fir. AI rufmry oj ,M~rncq~mc-nr lovmct!.
5l(3),.$87-516
Panon\. C.. Lldcn. R . O'Coru~or. 1:. & Sagan. II. (1991 ) limploycc rcsponw\ IO ~cctmulo~~;all~
drlvcn change. The m~plcmcnta~ion of ofticc awmauon in a WT\‘ICC wgww~mon. !lunuur
KeL;~ions.44(12~. 1?31-1356.
bcalc. K. (19X5) The p.uadox of “corpor~rc cul1urv”: Rcconcilmg o~uscl~cc IO uxialvarwn. (.@~r.
Cc Hwagemrnr Re,ww. 27(Z). 26 42.
Ynsmorc. W.. & Fagam. hl. (I%?). Particlpallon. indlvldual dcvclopnicn~. and orgnnir31tonal dc\clop-
nsnl. A rcvicw and synlhc~ls.loumcd c~MunU,ipvnen~. IN(?). 375-397.
Rclctrr~. A. (1985). A rcvicu and rcconccpruahrAon of organwAonal commilmcr:b AC-crcfrmy GJ
Mrmugemertf RerirH: 10. 465476.
Rltti. R.. h Silwr. J. (1986). 1%~ proccws of ins1l1u1ionali?ation. The dramarurgy of cxchangc III
inlcrorganiwtional r&ions. Admmisf~fw Sclencc Qwrrrri~. 31. 25 -42.
Rousseau. D. (1947). Organira1ional behavior in Ihc IXW qanizahonal era. ~r:nuaf HCVIOV c!f
I’sjcholo~~. 48. 5 IS ..S-l6.
Ryan. H., & Qos. N (1953) The diffusion of hybrid seed corn 11) NO low commurutw Kurui
Stn~Iolo~y. 8. 1 s- 24
 ACHILLES AKMENAKIS, STANLEY HARRIS, and HUBERT FEILL)

Schcrkcnbach. W. (1985). Pcrformancc apprai.sal and qualiIy: Ford’s new Philosophy. puJirv f’rqre~~,
/‘V(4). 40-46.
Sla~cr. M.. & Kouncr. D. (lYY2). Conlidcncc in bclicfs abom social groups as an ou~comc of mr~sapc
cxposurc and iIs role in bclicf change punisIcncc Cwwwnicu~wn Rrsrarrh. fq5). 597417.
Snyder. M. (1974). The sclfmoniforing of cxprcssivc behavior. Joumul uf Prrscmultry and .Suc-cul
Pqrhology. .TO. 526-537.
SIcwarc. T. ( 1991, August 12). GE keeps ~hosc ideas coming fi~rrunc. /ZS( 15). 4 I -49
Sfcwar~. T. (1992. Novcmbcr 30). Allied-Signal’s Iumaround hliIz tirnmc. /.X(12). 72-76.
SvyanIck. D.. O’Connell. M.. & Raumgardncr. T. (IY92). A Baycsian approach IO the cvaluauon of
organiradonal dcvclopmcnl cffor~s. In W. Pasmom & R. Woodman (I&.). RCSCCJK/J in oquni~J.
rioru~l e/umgc und drvrkopmrnf (Vol. 6. pp. 235 -266) Greenwich. CI’: JAI Press.
Tichy. S.M. (IYg2. Autumn). Managing change suaIcgically: The ~cchnical. pliIical and culnnal keys
Oqunizalicmul I~numies. pp 59- g0.
T~Iwky. I,.. & Klein. K. ( 1982). Innovation characIcrisIic~ and innovaIion a~~i~)n-implcnrn~Ii~)~l~
A mco-analysis of findings. IEEE ‘rrtJJWJCliOt!S En.qinrrrinX Munu~rmcnf. E.%f-2u( 1). B-45
Trirc. H . & Bcycr. J. ( 1993). ‘77ie cu~rvms o/H~o~(. oquniz~mms bnglcwood Cliffs, SJ: Prcnticc-Hall.
TumIall. W. ( IYgS). Disc-onnrc~rtn.q punrrs-munugmg I/IC Hell System brrcrliup: An insidr view Ncu
York: .M&rau-llill.
U7umcri. M. (1997). IS0 9000 and oltrr mclaslandards: Principles for managcimml pracl1cc”~ectiern.v
r~/~bfw~r~~emcnf Ihw4frve. I,‘( I ). 2 I ‘-3b
Van Maarwn. 1.. & Rarlcy. S. (IYW). Occuparional communiucs: Cuhc and comrol in orgaruraIions.
In B Slaw & L. Cummings (Ii& ). RCSCCJ~ in oquni~rionol Ixhuiior (pp. 287-365). Gmcn-
wIch. CT: JAI Press
Vansma. L. & Taillicu. T. (lYY6). Business process rc-cnginccring or .socio-Icctuucal system design in
NW clc~s. In R. Woodman & W. Pasmorc (Ed.%), Ressrurrii in oflunizurionul rhunge curd
devc/upmcnr(Vol. Y. pp. RI-100). Greenwich CT JAI Press.
Wagcl. W. (l9g7). Pcrformancc appraisal wirh a diffcrcncc. Prrxmnrl. (w(2). 4-b.
Wcick. K. (19M). Small wins: Kc&fining 1hc scale of so&l problems. Americ-em Psy-/~olo~isr. 3Y,
40-JY,
Wcisbord. M. (1988). Towards a new pcacrice thcoey of OD: NOW on snapshctiing and movicmaking
In W. Pasmorc & K. Woodman (F*is.). Rcsmrch tn o~unirurionerl c-hungr und drwlopmrn~ (Vol
2. pp 59.%I). Greenwich. CT: JAI Press.
Wdtnm, R. (1989) EvaluaIion research on organizational change: Argumcnrs for a “combined
paradigm” approach. In ft. Wcodman & W. Pasmorc (ti.). RUWCJJT~ in o~wtizu:iunul ~IuJngr
und drvrlupmrnl (Vol. 3. pp. 161- lw)). Greenwich. CT: JAI Press.
Ycung. A.K 0.. Bnx-kbank. J W.. & Ulrich. D.0 (1991). Organizational cul~urc and human rcsourcc
pracriccs: An empirical arscssrncnl. In R. Woodman & W’. Parmorc (Fiis.). Rrsrun-lt in ogunku-
fiunul chungr und drvrlupmrnr (Vol. 5. Pp. 59-81). Greenwich. CI’: JAI Press.