A Framework for Linking the Structure of Information Systems with Organizational Requirements for Information Sharing Author(s): Sunro Lee and Richard P. Leifer Reviewed work(s): Source: Journal of Management Information Systems, Vol. 8, No. 4 (Spring, 1992), pp. 27-44 Published by: M.E. Sharpe, Inc. Stable URL: http://www.jstor.org/stable/40397996 . Accessed: 18/03/2013 20:56
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp

.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

.

M.E. Sharpe, Inc. is collaborating with JSTOR to digitize, preserve and extend access to Journal of Management Information Systems.

http://www.jstor.org

This content downloaded on Mon, 18 Mar 2013 20:56:30 PM All use subject to JSTOR Terms and Conditions

A Framework Linking Structure for the of with Information Systems Organizational for Requirements Information Sharing
LEE P. SUNRO ANDRICHARD LEDFER in candidate Management Information at Sunro Lee is a doctoral Systems Rensselaer His research interests include Institute. current issuesin methodological Polytechnic andtesting, decision information group support systems, development expert systems His and infrastructure, end-user computing. papershave appearedor will systems An with Journal, OMEGA,and Systems Applications: International appearinExpert member ACM, TMIS, andDSL of HICSS. He is a student in SchoolofManagement Rensselaer Richard Leifer is anAssociate Professor the at in New York.His undergraduate Institute Troy, education was at the Polytechnic and workwas at theUniversity of of Berkeley, his doctoral University California, concerns organization-information the Madison.His research Wisconsin, systems from He in interface, processing perspective. haspublished primarily an information Administrative Science Quarterly, Journal, of Academy Management Academy of IEEE Transactions Engineering on and Review, Management Management, MIS Quarterly. relations between Abstract: Thispaperdescribes structures IS and organizational for structures based on requirements information Information is sharing. sharing between and defined a linking as Task concept organizational IS structures. characwork and networked structures of teristics, technological interdependency, teams, information the determine of sharing requirements. Adopting paradigm organizations that IS fît this to the postulates for structures be effective organizational school, paper for information should sharing possesscapacities matching requirements. they information inforKey words and phrases: information structure, sharing, system structure. fit, mation-technology organizational

1. Introduction forms information of The purpose of this paper is to investigate how various members. for can meetrequirements information sharing organizational by systems in that more increasing has and critical a postindustrial becomes Thisproblem society and turbulence . and more increasing and [27] complexity, more increasing knowledge, and stemfrom, These characteristics amongotherfactors, global competition a for for fastcycle pressure new products. strategies coping with Organizational
Journal 1992,Vol. 8, No. 4, pp. 27-44 Systems Information o/Management /Spring © Inc., 1992 Copyright M.E. Sharpe,

This content downloaded on Mon, 18 Mar 2013 20:56:30 PM All use subject to JSTOR Terms and Conditions

28

LEE AND LEIFER

suchradically different environments beenby experimenting new forms have with of organizational structure new information infraand/or implementing processing structures. Suchinfrastructures, from derived advanced telecommunication technologies, providefirms opportunities with torestructure dramatically and increase their information flow and information The recentemergence systems of processing capabilities. recentralization dataand/or of and outsourcing support, IS integration, processing, reflect efforts toward information and balancing processing requirements capabilities. Forexample, cross-functional business activities formore call horizontal communication functional andsuchincreased units needfor lateral has among cooperation been met integrating IS data [45] by dispersed including andnetworks orbyreengineering business [25,52], processes In many structures emerging and however, organizations, emerging organizational IS infrastructures the and relationships between twohaveevolved ad-hoc in the ways. In recognition suchconfusion, paperattempts developa framework of this to for the between structures emerging IS and understanding proper alignment management structures baseduponthelinking of concept information sharing. The organization thispaperis as follows. of structural First, general relationships between andIS arereviewed. ThenIS structures defined a model are and organizations fordeterminants IS structure presented. of is based on theconstruct that, Following ofinformation and for a of sharing contingencies information sharing, series propositionsare developedlinking structures emerging with IS emerging organizational structures. Based on therationale traditional structures are developed, organizational revisited. and are Finally, managerial implications conclusions discussed.

2. Structural Associations between and Organizations IS
In general, as organizations grow and mature,they tend movetoward to more mechanistic and adaptability. when systems, thereby However, losingflexibility environments becomemorecompetitive turbulent, and need to organizations return more structures One wayorganizations usedtobecome have [46]. adaptable, organic more is Eventhough cansupport IS organic bydownsizing. organizational downsizing elimination the of needfor middle routine through [1 management 1,2 1] orautomating tasks[55], generally communication tendto remain and patterns unchanged only control becomesmore accountable. Fromthis IS managerial perspective, havebeen viewed management a monitoring as rather as an opportunity than for by mechanism, change. research generated has on strucEmpirical contradictory findings IS-organizational ture Foster Flynn and claimthat [3,29, 37]. Forexample, increased relationships [23] utilization IS creates of task-focused interactions personalized and which contacts, lead an organization toward moreorganic On et hand,Crowston systems. theother al.'s [12] computer case a toward with centralization, conferencing suggests tendency more staffs decisions and madehigher in thehierarchy. specialized up

This content downloaded on Mon, 18 Mar 2013 20:56:30 PM All use subject to JSTOR Terms and Conditions

FOR SHARING 29 REQUIREMENTS INFORMATION

indicate IS tends enhance existing that studies to the match organof Contingency with structure environments ,44]. Forexample, izational [33 organizations implementin centralized less complex IS becomemore environments with and more routine ing in nonroutine tasks.However, tasksand uncertain environments computerization leadstodecentralization. Leifer [32] andBurch[10] suggest matching organizational and structures information systems configurations. Similarly, emphasizing proper and and structure, among strategy, IS, Buchanan Linowes[8, 9] developa alignment IS in framework distributing resources whichusergroups' for responsibility-based andtaskspecificity determine degree decentralization the of capabilities computing claimthat effective of IS resources. information are thosethat the fit They systems structure culture. and organizational fieldstudy reports there little On theother that is hand,a recent [1] relationship between structure IS hardware and distribution. Whilefeworganizaorganization - such as standardization, tionalcharacteristics and formalization, control have beenfound be associated to with structure information the of services [39],a firm's of as [48], distribution decision [1], strategy making and size [22] havebeenfound factors determining structures. for IS Takentogether, findings rather these are One the contradictory. reason maybe that levelofgranularity todifferent leads definitions IS centralization decentralizaof and tion.For example,Olson and Chervany and [39] includesystems development as whileAhituv al. [1] focussolelyon hardware et management well as operations in distribution. Another is includes shortcoming thepast research thatit mainly structural contextual variables suchas organizational variables, omitting technology andenvironments. In sum,there seemstobe a reciprocal between characrelationship organizational and teristics IS. IS influence characteristics at thesame time, and, organizational dimensions affect structures, IS research tends more organizational although previous to emphasize impacts IS on organizations theeffects organizational the of than of In on dimensions IS structures. thispaper,by emphasizing contextual variables of a andbyusing mid-range definition IS structure, investiof we organizational design of variance IS and thestructural of associations gatethedeterminants thestructural and between a the organizations IS, thereby suggestingwaytounderstand reciprocal IS between andorganizational structures. relationships

Variables IS Structures of 3. Description
There have been a numberof variables usedtodescribe structures. way IS One has of IS structures beendescribed beenin terms a computer have network topology distribution Other [1]. dimensions, bus,LAN, etc.) [10] orhardware ring, (e.g.,star, of decision-makinclude from derived research, practice datamanagement, previous marketandapplication [8, 19,22, 34,39]. Morerecently, development ingauthority, as havebeenreported examples strategic of of IS oriented applications coupling IS the IS Therefore, additional dimension, boundary, independent among organizations.

This content downloaded on Mon, 18 Mar 2013 20:56:30 PM All use subject to JSTOR Terms and Conditions

30

LEE AND LEIFER

IS structures. should considered understanding andinterorganizational be for intrato the Sincea single dimensional may be approach not appropriate explain mapping this between multifaceted and a multivariate a IS structure, paper organizational considers dimensions IS structure: five of hardware locus distribution, ofapplication and database decision location, authority, systems boundary. development, planning into modes:centralized, Theseareusedas descriptors classifying structure four for IS and decentralized, (IOS) management. hybrid, interorganizational systems

3-L Centralized Systems or Centralized are around central a (central processor mainframe systems designed datacenter databases) with "dumb"or diskless terminals (X maintaining corporate interactive information activities. Centralized control terminals) allowing processing whereusers have limited technical usuallygovernstransaction-oriented systems In thistype system, information of the maintains expertise. (IS) systems department a tight control over planning, network hardstructure, application development, and data ware/software purchasing, centralized control.

3-2. Decentralized Systems
Decentralized are on systems theleastcentralized, focusing network connectivity, rather than end-user In activities. thistype structure, of user controlling computing control owndata, their in groups decision-making authorityapplications development, and In user establish ownfunctional their maintenance, procurement.addition, groups resource centers support computing that the activities theunit of level(e.g.,division, thecentral group IS tends focus to product development etc.).Consequently, group, onstrategic and of facilities and planning, policy setting, maintenance communication utility systems.

3.3. Hybrid Systems sometimes referred as intermediate to Hybridsystems, [29], arrangements or as aredesigned with samebasicconfigurationcentralized the as "watchdog" policy[19], but own computing systems, withlocal processors (smart terminals) havingtheir and Theselocalprocessors communicate devices, databases. capabilities, storage may with others the hub. local users through central With hybrid systems, maybe allowed todeveloptheir ownapplications their with ownprocessors. the However, typically, central group IS makes most from hardware software and decisions, major purchases toapplications In the to development priorities. addition, IS group attempts standardize operating and for and systems communication protocols facilitating applications datasharing. the of control Despite reduction applications development backlogs, tight overequipment and needsapproval the of purchase application (which development central group) slowthe IS can andbecome sources of programming process potential tensions frustrations and [19].

This content downloaded on Mon, 18 Mar 2013 20:56:30 PM All use subject to JSTOR Terms and Conditions

REQUIREMENTS

FOR INFORMATION SHARING

31

34. Interorganizational Systems three are an The previous within firm. a systems gearedtoward IS infrastructure breaktheboundaries a firm's of network structure and systems Interorganizational IS other firms' in thewayofcollaboration. IS connect their with Barrett defines [5] information as an interorganizational data system "an integrated processing/data utilized twoormore firms." other communication The by system separate participant can ontheparticipants' deployment. variables be configured IS depending descriptor

and 4. Contingency Variables IS Structures
Decisions about the particular form of IS structure implement be to can variables suchas formalization, influenced organizational by design specialization, of and as standardization, centralization, complexity, well as hierarchy authority, contextual variablessuch as environments, size, and goals [13, 24]. technology, relations and are contextual, IS structures depicted among organizational, Suggested in figure Of thecontextual 1. variables listedabove, we focuson technology and in environments thefollowing sections because thesedimensions have undergone in the and IS the rapidtransformation pasttenyears perhaps impact morethan other factors. These two contextual variablespose differential for requirements information It that of links therefore, theconcept information sharing. is ourcontention, sharing andhelpsus understand IS structures, structures, relationships among organizational andcontextual we turn information to variables; hence, next sharing.

5. Information of Sharing, Contingencies IS andOrganizational and Forms Structures, Emerging Organizational
Informationprocessing involves three AcnvniES: information retrieval (proinformation and use information duction), sharing, information [28].Computer-based and is communication based on a common sharing a multilateral reciprocal activity frame reference of such as corporate communication databases, supported IS, by andapplication as information-sharing IS, development platforms. protocols, support the of tools demonstrating usefulness theinformation have been sharing concept, businessprocesses.Hammer utilizedforreengineering arguesthat"conventional structures fragmented piecemeal,and theylack the integration are and process to business necessary maintain qualityand service"[25, p. 108]. Reengineering needsa cross-functional and/or cross* [25] [17] organizational perspective processes which be achieved information can from by sharing among representatives organizaof tionalsubunitParticularly, amongtheprinciples reengineering suggested by enablesorganizations reducethe to information Hammer, sharing computer-based and consolidate activities of resources, makemore parallel dispersed steps processes, cross-functional [e.g.,17,45, 52]. systems byintegrated In thecontext processreengineering, information is of for sharing a keyconcept

This content downloaded on Mon, 18 Mar 2013 20:56:30 PM All use subject to JSTOR Terms and Conditions

32

LEE AND LEIFER

Context Dimensions -Environments -Technology -Size

▼

Jorg.nlzit.cn., I '. »""««" I
I i

-fi08li

*

I Systems (IS) *IT"^«!^ I Struct«™ ' ti '

} I

Structural Dimensions -Formolization -Specialization -Standardization of -Hierarchy authority -Centralization -CompleHity -Professionalism -Personnelconfiguration

Figure1

Determinants IS Structures of A Conceptual Model for

and can be clustered structural varianceof IS because IS resources explaining in controlled newways,depending information andboundon sharing requirements aries.Therationale IS structuring oninformation of based is to of sharing similar that theinformation fit IS and school[4, 14,51]. Thatis,effective deployment processing and henceincreased can as control, effectiveness, be expressed a organizational function thefit of between organization's an information and sharing requirements its information Two determinants the level of information for sharing capabilities. as aboveandas discussed belowareorganizational and sharing suggested technology environments. organizational Three forms haveemerged having as enhanced capabilitiesfor an needfor information from meeting increased sharing stemming increased environments increasingly and nonroutine increascompetitive technology requiring of flattened forms with self-diingamounts subunit interdependence: organizational work teams (SDWT), intraorganizational networkedsystems and recting networked IS to interorganizational systems. structures designed fittheseemerging forms suggested are next. organizational

5.1. Technology Information and Sharing one of is in Technology, determinant thelevel of information sharing, defined the literature boththedegree subunit as of as wellas management [49], interdependence task and organizational inputtransformational characteristics (task predictability toThompson thework banktellers of [40]. [49], analyzability) According (mediating, orpooledtechnologies) wouldrequire lesserdegree interdependence work a of than

This content downloaded on Mon, 18 Mar 2013 20:56:30 PM All use subject to JSTOR Terms and Conditions

REQUIREMENTS

FOR INFORMATION SHARING

33

or lines(longlinked, sequential Intensive also called on assembly technologies). - technologies, as theworkof a hospital such the team, surgical reciprocal require Further ofpooled interdependence, examples interdependence. greatest requiring includeMcDonald's restaurants, interunit because an outlet little interdependence withan outletin Urbana; thisis also trueof bank in Chicago need notinteract Further and sales offices. branches reexamplesof reciprocal interdependence, include university, a deal ofinterunit a great where wide a interdependence quiring in and support servicesare provided a coordinated of way to variety disciplines of teamthat must coordinate meetmilestone to or and students, subunits a project [13]. requirements budgetary a mustallow forfrequent In the case of reciprocal interdependence,structure for accomplishment. is notthe task This and case communications mutual adjustments wherelittle coordination required is for withpooled interdependence, day-to-day also the of task successful completion [13]. Taskroutineity impacts degree informaand are Routine tasks characterized few tion , by exceptions high analyzability sharing. and for well understood procedures completing them are can be that tasksthat is, taskshave manyexceptions are so On and formalized. theother hand,nonroutine and specialknowledge often that are or unstructured experience (equivocal) vague them[15, 40, 50]. These arguments that information to complete suggest required of and are sharing requirements a function theleveloftaskcharacteristics technologin offigure 2. as ical andtaskinterdependence,shown theupper part and from level derives the information between Therelationship sharing technology in as depicted figure The 3. and characteristics technological oftask interdependence, when show is needed levelofinformation organizational technologies sharing highest while level arenonroutine, thelowest and tasks of a high degree interdependence unit and showlittle is ofinformation interdependence sharing neededwhentechnologies leads andinformation between This are tasks routine. relationship sharing technology us tothefollowing proposition. levelsofinformation will 1: sharing greater Proposition Organizations exhibit and nonroutine with interdependence more technological requirements increased taskcharacteristics. can IS clearly be effective rules, mechanisms, alongwith hierarchy, coordinating with anorganization increased andteams ,andas such, [24] capacity provide planning, value which to IS Forexample, contributesintegrating chains, information for sharing. of the andproviding integrationfunctional life the reduce product-developmentcycle, as subunits ones,decision tasks, wellas routine [17]. In cases ofnonroutine analytic have and for models forecasting planning beenutilized (i.e.,decision support systems) mailforfacilitating [18] and electronic systems alongwithgroupdecisionsupport and coordination information sharing. group include dataintegrainformation for IS Other options increased capacities sharing cross-functional model data common models, tion linking through integration through data and of [26, data,standardization communication protocols, centralized centers as to forIS options be cost-effective, suggested theinformation by 45]. However,

This content downloaded on Mon, 18 Mar 2013 20:56:30 PM All use subject to JSTOR Terms and Conditions

34

LEE AND LEITER

IS I Efftctltft I I Inffttnicturt | Task Characteristic« -R.athta i , IS Stricter«

-No-ftln«

| '* '|

_^__

Sfearlee

,„„..„„

__^

I

T *"

^________

I

T.caa.l.flc.1 I.B.atf.ac.

,y*-^Y /
1 | 1 Varii Taaait

'

'

ljr.ry.ll.« Seerlee

I

'

^

''¿S*" - iiuone

I üsíZZ ''
IS Olai.a.l.at - Mrtwara lUtrtaatlaa - ippllcatlaa o.«.i.»m.at

1

I Eavlraaaiaatt

1

- ♦

H9tmwtt stmctarat

I

|

.f

irj..»ù.tl^»

|

...t.M..l.«.«.a

- Plaaalaf Rttlaarltg - Sa»l.m l.wad.ni

Figure2

IS Model for Structures An Information Sharing

Nonroutlna

^^^^

Routinenett

^^r

-

Routine

^^r^

Pooled

Sa^uanllal

Reciprocol

Interdependence

Figure3

The Level ofInformation Requirements Sharing

that fit should for IS infrastructurecanachieve opt the processing school, organizations a balancebetween information the of and sharing capacities theIS options thesharing needsamong Based upontheabovearguments, infrastructure inforIS and subunits. mation lead tothefollowing sharing requirements propositions. la: with will Proposition Organizations higher technological interdependence have a higher IS integration, morecentralized with control over degreeof databaseand communication facilities. with need lb: Proposition Organizations little fortechnological interdependence to willdecentralize and databases, hardware, applications development localized units. decision making

This content downloaded on Mon, 18 Mar 2013 20:56:30 PM All use subject to JSTOR Terms and Conditions

REQUIREMENTS

FOR INFORMATION SHARING

35

and 5.2. Environments Information Sharing
In addition technology, are to environments a sourceof uncertainty 20] and [13, as [14]. DaftandLengel[14] define equivocality uncertainty "theabsenceof inforas the mation" and equivocality "ambiguity, existence multiple conflicting of and aboutan organizational situation." environments As becomemore interpretations the of and competitive, amount uncertainty equivocality and conhostile, complex, within them tend increase 41]. Thus, to tained information [14, organizational sharing of increase a result theextra as andequivocality uncertainty requirements stemming and environmental from uncertainty equivocality. information from Due to greatly and requirements technology increasing sharing and uncertainties equivocalities, areexperimentenvironmental many organizations of mechanisms organizational and newforms coordination structures [51], ingwith in workteamsand networked such as self-directed structures, orderto increase For increased information information sharing capacities. IS tosupport organizational IS structures tobe approofthese neworganizational need forms, sharing capacities with neworganizational matched structured these and forms. Theserelationpriately below. aredescribed ships WorkTeamsandIS Structures 5.2.1. Self-directed for and have led many and Environmental market pressures flexibility innovation into of to hierartraditional delayered organizations reform a combination flattened, work work chiesalong withself-directed teams(SDWT) (also called autonomous orself-regulating teams). work teams, self-managing groups, and work redraw information Flattened organizations self-directing teams sharing on of teams. Basedonteam Sundstrom boundaries, depending thetypes work patterns of intofour et al. [47] classify Advice/Involveapplications workteams categories: andAction/Negotiation. catThese ment, Production/Service, Project/Development, are associatedwithdiffering levels of integration differentiation. and The egories are with A high levelsofintegration associated level technological interdependence. a levelofcoordination synchronization managand with ofintegration requires high suchthat teamthat a a ers,peers,and customers requires highdegreeof external needsa high levelofinformation with and work units. integration linkage sharing other A teamcan be differentiated otherworkunitsthrough from members' special access tophysical facilities. example, For teamlifespan,or exclusive we expertise, suchas project to differentiated wouldexpecthighly teams, teams, be moreloosely as with teams[53] due totheir other counterpartscompared production coupledwith with and which makeinformation units sharing other knowledge facilities, specialized and difficult cumbersome. for information can that Thisreasoning sharing be suggests requirements interunit of differentiation integration, shown and as determined a combination work-team by IS and infigure As information 4. increase, integration centralrequirements sharing In the will be morevisiblein theIS infrastructure.orderto support ized control

This content downloaded on Mon, 18 Mar 2013 20:56:30 PM All use subject to JSTOR Terms and Conditions

36

LEE AND LEIFER

with sharing the should varieties SDWT, theIS infrastructure be designed of requirein us of teams mind, ments these propositions: leading tothefollowing in that 2a: project/develProposition TheIS structure organizations havemore will and of opment types workteams(highdifferentiation low integration) be teams. with than more decentralized organizations other of types work 'serin that 2b: production! Proposition TheIS structure organizations havemore will and vicetypes work teams (lowdifferentiation high integration) be more of teams. than with centralized organizations other types work of and 5.2.2. Network Organizations IS Structures not and environments Turbulent competitive require organizations onlyto adaptto environmental but process.Whenthe requirements, also to speedup theadaptation life become oftechnological is rapid, shorten markets and change product cycles pace tend more that of specialized, meaning economies vertical integration tolose flexibiland ity efficiencies traditionally gained[43]. in or to Vertical each suborganization subunit a network disaggregation requires A network a certain of thebusiness structure on function. perform portion depends for rather markets a mechanism coordinating flowof materials services, as the and A market-based on than managerial decisions from managerial the coordihierarchy. nationmechanism undera network structure provideeach participant with can and the cost thanthat the of flexibility adaptability, although governance is higher hierarchical coordination [54]. Network in structures redefine meaning organizational the of boundary thesense that boundary the"organization" defined theplayers thenetwork. the of in In is by this a network is interorganizational than rather intraorganizational, sense, boundary in network belong onelegalentity toparts several the to or of although players the may as [36] through legal entities. "Downsizing" disaggregation, opposedto delayering, is a majordifference traditional from downsized structures characterized vertical by Networked downsized firms built are around specialized a of integration. competence skillsand knowledge, with seekingarrangements other specializedfirms through contract or research, jointventure, equitypartnership, cross-marketing, licensing, collaborations. multiorganizational In an interorganizational network from vertical the of resulting disaggregation businessfunctions, each organization opportunities building particular has for its distinctive sources know-how of locatedoutside [38],andfor competence accessing ofthe . Inthis ofarrangement, for are [43] organization type requirements support open to competition insideoroutside focalorganization. from the rather Complementing, than the of enhances spread inforthe of competing, nature thenetwork participants in mation order integrate to sources technological of multiple knowledge. A variation theinterorganizational on network notion collaboration is firms. among As collaboration firms information increases, among sharing through intraorganizationalIS expands theinterfirm with increase itsstrategic to level an in value.Konsynski

This content downloaded on Mon, 18 Mar 2013 20:56:30 PM All use subject to JSTOR Terms and Conditions

REQUIREMENTS FOR INFORMATION SHARING

37

High

lí!prrt ,áme!T?tr* iOiSSSift L°cl !"* FftlîT*?.!ÏA11*.111!.'*0" Llttlasynchronlzotlon

Mtmbtrt CMpert Specleltee* facilities Enlamad taam Ufaspan Clasa synchronlzetlon.

íorctf $yr9tn< cockpit ttam$t crew$ MSãsssituSitTMit
Laui lavai of Information Sharing Noeti

Work-Team

Differentiation

Common facilillas Shorttaam Ufaspan llttlasynchranlzatlan

Represontotuie momborthlp

Sdoro* facilillas Uartablateam Ufaspan Clasa synchronization

Uerleftle mambarshlp requirements

centralgroup Committees, Quality

teems. Manufacturing Assembly et Highlevel ef Inform Ion Shoring

Lou»

High

Integration

Figure4 Typesof WorkTeams and TheirNeeds forInformation Sharing (Adoptedand Modified from [47])

and McFarlan[31] call such electronic collaboration "information partnerships." new channels distribution to of in These contribute building resulting operational for efficiencies creating Miles and by opportunities scale and cross-marketing. Snow [38] suggestthatinterorganizational be fullydisclosed to network IS and tools participants' contributions their to participants be usedas verification for In or the vein,Maloneetal. [35] predict emergence agreements contracts. a similar thatconnectsmanydifferent of an "electronic brokerage" buyersand suppliers a and communication, through centraldatabase. Such electronicintegration, to contribute developingsuperior substitute currently products(serbrokerage and of vices) andalterthedefinition domain businessoperations. Well-publicized IS Airlines'SABRE resexamplesfortheinterorganizational includeAmerican ASAP order-entry American ervation HospitalSupplyCompany's system, sysEconomost,and Citicorp'sATM. These recent tems,McKesson Corporation's and of viewed as partnership" "electronic brokerage," prototypes "information "full-disclosure information toward [38], tend systems" systems first-generation in theinterorganizational driven information to increase setting sharing capacities information increased requirements. sharing by their IS configuration own and not maintain only Ina network structure, participants Whether networks. communication but control, also overseeinterfirm establishing in firms a network value-added or own networks usinga third-party their network, and electronic their intrafirm musttransition structure integration communication structures the for and electronic to structures interfirm integration communication a specialtype data Electronic exchange tobe successful. network (EDI), arrangement of on communication thebasisoffour the ofIOS, facilitates network types standard standard andinternational format: [26], cross-industry, industry-specific, proprietary, a on reliance IS requires high structures Withnetwork supported EDI, increased by

This content downloaded on Mon, 18 Mar 2013 20:56:30 PM All use subject to JSTOR Terms and Conditions

38

LEE AND LEIFER

levelof security control communications databases and on and acrosstheorganizational network. ensure security control, To this and andsystems applications programIS rather thanby individual user mingshouldbe developedby central personnel, In network mostof hardware/software and groups. addition, planning, purchasing, datacontrol should exercised thecentralized group. be IS by In sum, and environments market create uncertain, competitive, turbulent pressures, which lead somefirms form to collaborative network structures. interorganizational Theseneworganizational structures create for inforopportunities interorganizational mation increased levelsofcontrol standards, more and and sharing, requiring competent personnel. IS Suchchanges lead tothefollowing proposition. 3: collaborate with other with Proposition Firmsthat firms interorganizational morecentralized structures. IS (IOS) willbe associatedwith systems most connected thenetwork to leader'ssystem. Currently, IOS havebeencentrally The leader'sinternal are notrestricted a certain IS to of but to type network, rather control overnetwork databasesecurity. and willbe more disaster Further, recovery critical than under intrafirm in networks posited proposition Astley Brahm as 3. and on information as [2] comment Miles and Snow's "full-disclosure systems" being unrealistic. Such systems will that firms abandontraditional sourcesof mayimply such as proprietary know-how Hypothetically, the in [2]. competitive advantage, full-blown network can structure, CAD/CAMdesigners senddesigns organizational and specifications to suppliers producers. and Producers disclosetheir directly production schedule suppliers can nowcontribute reducing to who to invenproducers' tories wellas tocontrolling own.Therefore, as their network structures influence may IS structures follows: as interorganizational 4: structures the will mature, levelofdisclosure increase Proposition As network and network control become will more distributed decentralized. and

6. Current Structures IS Structures: and Organizational A Reinterpretation
As discussed above, previous research indicates there little no relation that is or between traditional structures IS structures. and the organizational However, question, remains unanswered. research not," tomaporgani"why Furthermore, past attempts zationsontoIS structures rather idiosyncratically, thanusingfiner-grained building suchas work teams divisions. sensitive and Not blocks, to levelofanalysis enough this studies IS problem, subunits many investigate atthe level, organizational while mixing We feel thatsubunits shouldbe treated building as blocks with (e.g., divisions). structures treated consisting these as of blocks. From organizational multiple building this we and the research more of perspective, canrevisit reinterpret structure-systems current forms. organizational

This content downloaded on Mon, 18 Mar 2013 20:56:30 PM All use subject to JSTOR Terms and Conditions

REQUIREMENTS

FOR INFORMATION SHARING

39

Structures Centralized/Hybrid and 6.L Functional Systems
Traditional wisdom wouldsuggest a centralized that decision and hierarchy vertically a IS for controlled/centralized improvproduction process requires tightly integrated the structure often the"systems-strucTherefore, functional ing efficiency. adopts tural" [16] perspectiveof organizational communications which emphasizes and rather information information From acquisition distribution, than interpretation. IS a thissystems-structural structure utilized are to approach, supportingfunctional data processing, such as transaction as well as to reducethe support processing, information of and of overload, standardization reports facilitation bottom-up by driven as of communication a response top-down directives. report In thefunctional vertical is structure, integrationused to developin-depth organizationalskillsand memories. suchas corporate databases knowledge-based and IS, in maintain history product a of residing a central processor, systems, designand Suchcentralized databases basescanserve and/or development. corporate knowledge of as a frame reference, and institutionalizing organizational experience maintaining aboutaction-outcome relations. centralized develHowever, knowledge application tends the of and between opment maintenance tohinder process coupling value-adding DP and often functions IS, sincecentral groups locateatsomedistance business from of and functions lacktheunderstandingstrategic business busirelationships among and nessfunctions IS. the Thesetypes arrangements of from preclude organization keeping with up rapidly and with markets, competitors. Therefore, technologies, organizations funcchanging tional often structures setupspecialunits work or (taskforces, self-managing groups, as active sensors interpreters and ofenvironmental [42]) "intrapreneurships" changes and innovators withsuch unitsbeginto have an interpretive [38]. Organizations in of communications of IS perspective organizational resulting a migration their to that certain of characteristicsdecentralized systems incorporate hybrid systems. 5: with structures develophybrid will IS Proposition Organizations functional the on structures, exact form which of depends thetechnological interdependence and thenature work teams of they employ.

and 6.2. DivisionalStructures Hybrid/Decentralized Systems to the from structuresdivisional structures triggers often Making transition functional from centralized to distributed/decentralized For systems restructuring systems. exHanover werecompletely theIS ofManufacturers cen(MHO) Corporation ample, the and as tralized operated a batch the system during 1960s.Then, processing during on-line to and 1970s,theIS werechanged distributed systems IS groups beganto In its with operations operations. 1985,MHO restructuredbanking integrate business MHO's IS have been decentralized mostof the and intofive sectors. Recently, was to IS resources handed thesector group that overcomputing control operational in sectors couldbuilda hybrid [6]. system their

This content downloaded on Mon, 18 Mar 2013 20:56:30 PM All use subject to JSTOR Terms and Conditions

40

LEE AND LEIFER

A major of is division that canquickly it advantage a downsized adapttoa changing and environment control product its ofMHO's development cycle.Therestructuring IS demonstrates strategic between the fit structures IS. MHO's IS and organizational aremorecloselylinked divisional to who of managers, areexposedto a wider range and handle them with increased managerial problems whocan then autonomy. Theprimary requirement divisional in IS structuretosupport cyclemanageis fast life customers' newwants ment, reducing product-development cyclesandmeeting the fromimproving internal to [7]. Consequently, focus of IS shifts efficiency a multifaceted mechanism directly for or to providing delivering products services clients. resources tobe more IS tend in usersites where users can closely deployed the their with integrate business expertise information technologies. often lead to duplication effort well as of as However,decentralized systems in difficultiesaccessing shared data.Recognizing these a more recent trend problems, has beento recentralize decentralized by integrating IS vendors'software, multiple and data a DP becomes Hence, centralized function purchasing, consolidating centers. a central entity creating IS for standards driving and cross-functional while systems, For allowinglocalized applications development. example,Benetton integrates achievesdesignflexibility compresses CAD/CAM withtheproduction and center, activities.As connectivity the value-adding technologies improve, pressureto recentralize someIS functions increases a greater lesser to or extent. Again, applying theinformation we framework, propose: sharing 6: withdivisionalstructures decentralize will IS Proposition Organizations hardware in divisions thatshowless (applications development, distribution) on but IS divisions, willcentralize (databases, technological dependence other decision on divisions are moreinterdependent that with planning making) those other divisions.

6.3. Matrix Structures Hybrid and Systems
One requirement successofa matrix for structure theneedfor is economies scale of in theuse ofinternal resources. Thiscondition should lead matrix to organizations a shared and flexible of IS resources. use a IS Thus,we wouldexpectto find central associated with matrix structures to itstight due control overcommunication group utilities equipment and as However, matrix purchasing. organizations implement orproduct teams with central administrative such project as support, teams, mentioned to above,tend be highly a of overtheir differentiated, needing certain degree control owncomputing resources. IS groups be located can belowthe business unit Dispersed whereas professionals work IS can level, with eachfunctional controlling closely unit, and team. can readily We applications datagenerated eachproject by applyproposition2a/2b matrix to andfurther as organizations postulate follows: 7: will decentralized Proposition IS ofmatrix organizations be more (application thanthoseoffunctional but development, databases) structures, IS planning

This content downloaded on Mon, 18 Mar 2013 20:56:30 PM All use subject to JSTOR Terms and Conditions

REQUIREMENTS

FOR INFORMATION SHARING

41

than will centralized that divisional in decisions matrix of organizations be more structures. is matrix structures that dual authority participant and with A problem associated In information in structure toincrease tend interactions thematrix equivocality. order between functional barrier subunits to reduce and the to overcome communication of will communication drive an perspective organizational equivocality, interpretive IS of thedevelopment enhanced network connectivity. have to network Advanced cognitive technologies beenutilized support processes role and a potential forsupplying rich as wellas dataprocessing havedemonstrated in contacts task-focused and interactions been have As media. anincrease personalized IS communications" in observed theuse ofdecentralized [23],theuse of"electronic members share a sourceof richmediaby helping [35] can provide organizational Theseelectronic such causesofconflicts. and communications, assumptions discover and electronic decision as group mail, computer conferencing, local systems, support and the of to have areanetworks, beenutilized facilitate exchange opinions toprovide information thus [18, capabilities 28]. Therefore, sharing rapidfeedback, increasing as we can postulate follows: will as 8: communications,a means horizontal of linkage, Proposition Electronic or than structures infunctional divisional be more foundin matrix frequently structures.

and 7. Implications Conclusions with structures Informationsharing is a key concept for organizational linking of the is The IS structures. argument that information sharing capabilities IS should between needs. information to be structuredmeet organizaProper alignment sharing as to for is structures critical organizations achieveflexibility, well as and IS tional This can environments. approach be usedto and in efficiency, competitive turbulent ones. as andIS alignments, wellas emerging current understand organizational technoloof dimensions organizational contextual design, organizational Among levelsof information of sharing requiregies and thenature workteamsdetermine ments. High levels of information sharingcapabilities can be achieved by in of IS databases, integration the form consolidated recentralizing and systems netof andin thedevelopment interorganizational unit cross-functional integration, work systems. structure research and for are There twomajor organizational-IS implications future information of the derivedfrom conclusions thispaper.First, matching practice when be should considered and organizadesigning requirements capabilities sharing for these information and tions when Usinginfororganizations. systems developing in inconsistencies future assistresearchers resolving as mation sharing a lensshould the is intheorganizational-IS Second,there a needfor level they relationships study. the is than currently case. Thatis,unlesssubunit, tobe smaller ofanalysis granularity distinctions are levelsofanalysis used,important or regarding department, division

This content downloaded on Mon, 18 Mar 2013 20:56:30 PM All use subject to JSTOR Terms and Conditions

42

LEE AND LEIFER

to will thispaperwill contribute differential structures be lost.It is hopedthat IS and It future research practice. is also and current research toimproving understanding for research a here the hopedthat propositions suggested willprovide direction future for structure-IS structure relationships. understanding organization REFERENCES the M. 1. Ahituv, Neumann, andZviran, Factors N.; S.; affecting policyfordistributing resources. Quarterly, 4 (December MIS 13, 1989),389-401. computing the for R.A. Organizational 2. Astley, G.W.,andBrahm, strategies: designs post-industrial In collaboration. C.C. Snow,ed.,Strategyf roleofinterorganizational Design,and Organization CT: HumanResource Greenwich, JAIPress,1988. Management. we what and 3 . Attewell, andRule,J.Computing organizations: weknowandwhat don't P., know.Communicationsthe 1984),1184-1192. of ACM,27, 12 (December and R.P. Information 4. Balaguer, N.S., andLeifer, capabilities organizational processing Graduate SchoolofBusinessAdministration, paper, design:a modelandfield study. Working Harvard 1990. University, an S. alternatives inter-organizational and 5. Barrett, Strategic system implementations: 5-16. overview. Journal Management 3, 1986/87), of Information Systems, 3 (Winter Sloan Management the 6. Belle,A.L., andNyce,H.E. Whither IT organization? Review, 24, 4 (Summer 1987),75-86. for 7. Bower,J.L.,and Hout,T.M. Fast-cycle power.Harvard capability competitive Business Review, 6 (November/December 88, 1988),110-118. 8. Buchanan, and R.G. Understanding data Harvard distributed processing. J.R., Linowes, Business Review, 4 (July/August 58, 1980),143-154. and 9. Buchanan, R.G. Makingdistributed processing data work. Harvard J.R., Linowes, Business Review, 5 (September/October 143-161. 58, 1980), 10. Burch, J.G.Network the bindinformation Data Managetopologies: tiesthat systems. ment, 12 (December 23, 1985),34-37. 11. Clement, andGotlieb, C.C. Evolution anorganizational of interface: newbusiness the A., at firm. ACM Transactions Office on department a largeinsurance 5, Information Systems, 4 (October 1987),328-339. 12. Crowston, Malone,T.W.; andLin,F. Cognitive scienceandorganizational a K.; design: case study computer of MIT SloanSchoolofManagement. conferencing. Working paper, 1986. 13. Daft, R.L. OrganizationTheory Design, ed. St.Paul,MN: WestPublishing. and 2d 1986. 14. Daft,R.L., andLengel,R.H. Organizational information mediarichness requirements, andstructural Science, 5 (May 1986),554-571. design. Management 32, 15. Daft, N.B. into amount equivocality and R.L.,andMacintosh, A tentative exploration the ofinformation in work Administrative ScienceQuarterly, processing organizational units. 26, 2 (June 1981).207-224. 16. Daft, G. learn: communication a R.L., andHuber, How organizations framework. S. Li Bacharach N. Tomasso,eds.,Research Sociology Organizations, 5. Greenwich, and in vol. of CT: JAIPress,1986. 17. Davenport, J.E. information T.H., andShort, Thenewindustrial engineering: technology andbusiness Sloan Management process redesign. Review, 4 (Summer 31, 1990), 11-25. 18. DeSanctis, andGallupe, R.B. A foundation thestudy group for of G., decision support Science,32, 5 (May 1987),589-609. systems. Management 19. Donovan,J.Beyondchiefinformation officer network to HarvardBusiness manager. Review, 1988, September-October pp. 134-140. 20. Downey,H.K.; Hellriegel, and Slocum,J.W.Environmental the D.; uncertainty: construct itsapplication. and Administrative ScienceQuarterly. 4 (December 20. 1975Ì.613-629. 21. Driscoll,J. How to humanizeofficeautomation. Office, Technology, People, 1, 2 1982),167-176. (September 22. Ein-Dor, and Segev,E. Organizational context MIS structure: P., and some empirical evidence. MIS Quarterly, 3 (September 6, 1982),55-68.

This content downloaded on Mon, 18 Mar 2013 20:56:30 PM All use subject to JSTOR Terms and Conditions

FOR SHARING 43 REQUIREMENTS INFORMATION information D.M. Management its on 23. Foster, L.W., and Flynn, technology: effects MIS form 8, organizational andfunction. Quarterly, 4 (December1984),229-236. MA: Addison-Wesley, 1977. J. 24. Galbraith, Organizational Design.Reading, obliterate. Harvard work: don'tautomate, Business M. 25. Hammer, Reengineering Review, 1990,pp. 104-112. July-August to AMA PublicaK.C. EDI: FromUnderstandingImplementation. Membership 26. Hinge, 1988. American tions Association, Division, Management of and G.P. 27. Huber, Thenature design post-industrial Science, organizations. Management 30, 8 (August 1984),928-951. decision MIS of G.P. Issuesin thedesign group 28. Huber, 8, support systems. Quarterly, 3 (September 1984),195-204. considerations decentralized and versus J.L. 29. King, Centralized organizational computing: 15, 1983),319-350. management options. Computing Surveys, 4 (December in theoretical 30. Kling,R. Social analysesof computing: perspectives recent empirical research. 12, Survey, 1, (March1980),61-110. Computing - shareddata,shared W.F. Information 31. Konsynski, B.R., and McFarlan, partnerships Business scale.Harvard 1990,pp. 114-120. Review, September-October with information structure. R. 32. Leifer, Matching systems organizational computer-based MIS Quarterly, 1 (March1988),63-72. 12, as E.F. 33. Leifer, andMcDonough, Computerizationapredominant R., technology effecting Annual on work structure. unit 1985, Conference Information Systems, Proceedings the of Sixth pp. 238-248. of structures the for management information In 34. Lucas,H.C. Alternative processing. R. vol. 2. New York: and eds.,The Economics Information of Processing, Goldberg H. Lorin, John Wiley,1982,pp. 55-61. R.I. markets electronic and hierar35. Malone,T.W.; Yates,J.;andBenjamin, Electronic chies.CommunicationstheACM,30, 6 (June 1987),484-497. of of G.C. The evolution transaction in 36. Mariotti, andCainarca, S., governance thetextile1 Behavior Organization, (1986),351-374. and Journal Economic of industry. clothing and 37. Markus, M.L., andRobey,D. Information technology organization change:causal and in structure theory research. Science, 5 (May 1988),583-598. 34, Management new for 38. Miles, R.E., and Snow,C.C. Organizations: concepts new forms. California Review, 3 (Spring 28, 1986),62-73. Management N.L. between characteristics 39. Olson, M.H.,andChervany, Therelationship organizational services function. Quarterly, 2 (June MIS of information andthestructure the 4, 1980),57-67. for 40. Perrow, A framework the comparative C. American analysisof organizations. Review, 2 (April1967),194-208. 32, Sociological Controlof Organizations: Resource A 41. Pfeffer, and Salancik,G.R. The External J., New York:Harper Row, 1978. and Perspective. Dependence New and G. 42. Pinchot, Intrapreneuring. York:Harper Row, 1985. or new 43. Powell,W.W. Hybrid organizational arrangements: form transitional development? Review, 1 (Fall 1987),66-87. 30, Management California and structure: empirical some re-examined. 44. Robey, Computers management D. findings Human Relations, 11 (November 30, 1977),963-976. decentralized" organization. IS Harvard Business E.M. The "centrally 45. Simson, Review, July-Aueust 1990,pp. 158-162. struc46. Slevin,D.P., and Covin,J.G.Juggling entrepreneurial and organizational style SloanManagement act ture howto eetyour together. Review, 2 (Winter 31, 1990),43-53. D. teams. American 47. Sundstrom, De Meuse,K.P.; andFutrell, Work 45, E.; Psychologist, 2 (February 1990),120-133. structure organizational with comH. the 48. Tavakolian, Linking information technology MIS a 13, 1989),309-317. strategy:survey. Quarterly, 3 (September petitive New York:McGraw-Hill, in 1967. J.D. 49. Thompson, Organizations Action. and communication structure: contina M.L. Workcharacteristics subunit 50. Tushman, ScienceQuarterly, 1 (March1979),82-98. Administrative 24, analysis. gency as in D.A. Information 51. Tushman, M.L., andNadler, concept processing an integrating Review, 3 (July 3, 1978),613-624. Academy Management of design. organization

This content downloaded on Mon, 18 Mar 2013 20:56:30 PM All use subject to JSTOR Terms and Conditions

44

LEE AND LEIFER

J.W. to Business 52. Verity, Is ittime junkthewayyouuse computers? Week, 22, 1991, July pp. 66-69. elements. In of 53. Weick, K.E. Management organizational change among looselycoupled San P.S. Goodmanand Associates, eds., Changein Organizations. Francisco: Jossey-Bass, 1982,pp. 357^08. New York:FreePress,1975. and O.E. 54. Williamson, Markets Hierarchies. on LaborProcess.NewYork:Monthly 1979. 55. ZimbaUst, ed.Case Studies the Review, A.,

This content downloaded on Mon, 18 Mar 2013 20:56:30 PM All use subject to JSTOR Terms and Conditions