Analyzing the Past to Prepare for the Future: Writing a Literature Review Author(s): Jane Webster and Richard

T. Watson Source: MIS Quarterly, Vol. 26, No. 2 (Jun., 2002), pp. xiii-xxiii Published by: Management Information Systems Research Center, University of Minnesota Stable URL: http://www.jstor.org/stable/4132319 Accessed: 16/08/2009 11:54
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/action/showPublisher?publisherCode=misrc. Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission. JSTOR is a not-for-profit organization founded in 1995 to build trusted digital archives for scholarship. We work with the scholarly community to preserve their work and the materials they rely upon, and to build a common research platform that promotes the discovery and use of these resources. For more information about JSTOR, please contact support@jstor.org.

Management Information Systems Research Center, University of Minnesota is collaborating with JSTOR to digitize, preserve and extend access to MIS Quarterly.

http://www.jstor.org

Webster& Watson/GuestEditorial

ANALYZINGTHE PAST TO PREPARE FOR THE FUTURE: WRITINGA REVIEW LITERATURE

By: Jane Webster Queen's School of Business Queen's University Kingston, ON K7L3N6 CANADA jwebster@business.queensu.ca Richard T. Watson TerryCollege of Business The University of Georgia Athens, GA 30602-6273 U.S.A. rwatson@terry.uga.edu

A reviewof prior,relevantliterature is an essential feature of any academic project.An effective review creates a firmfoundation for advancingknowledge.Itfacilitatestheorydevelopment,closes areas where a plethoraof research exists, and uncovers areas where research is needed. Inthe information systems (IS)field,we see few publishedreviewarticles.As a result,the progress of our field is impeded. To address this concern, the MISQuarterly launched MISQReview several years ago. The clear intention was to accelerate the accumulation of IS knowledge. A particular goal was to advance the state of theorywithinthe IS field. The stated purpose of MISQReview is to ...promote MISresearch by publishingarticlesthat conceptualize research areas and survey and synthesize priorresearch. These articles will provide importantinput in setting directionsforfutureresearch.1 The lack of theoreticalprogress in the IS fieldmay be surprising. Froman empiricalviewpoint,the IS field resembles other management fields. Specifically,as fields of inquiry develop, their theories are often on a from ad hoc classification which placed hierarchy systems (in categories are used to summarize to taxonomies which the empirical observations), (in relationships between the categories can be which to described),to conceptualframeworks (in propositionssummarizeexplanationsand predictions), theoreticalsystems (in which laws are containedwithinaxiomaticor formaltheories) (Parsons and Shils 1962). In its short history,IS research has developed from classificationsystems to conceptual frameworks.Inthe 1970s, itwas considered pre-paradigmatic. the level of development Today,itis approaching in empiricalresearch of othermanagementfields, likeorganizational behavior(Webster2001). However, unlike other fields that have journals devoted to review articles (e.g., the Academy of Management Review), we see few review articles in IS-and hence the creation of MISQReview as a device for accelerating developmentof the discipline. One reason we see so few theoreticalarticlesin IS relates to the youthof the field. Anotherconcerns the field. That is, constructinga review is a chalcomplexityof assembling a review in an interdisciplinary

1http://www.misq.org/misreview/announce.html

MISQuarterly 2002 Vol.26 No. 2, pp. xiii-xxiii/June

xiii

Editorial Webster& Watson/Guest

we have lenging process because we often need to drawon theories froma varietyof fields. Moreover, so few theories of our own. Nevertheless, the literature reviewrepresents the foundationfor research in IS. As such, reviewarticlesare criticalto strengtheningIS as a field of study. Anotherchallenge relates to methods of structuring and presenting these reviews. As the initialsenior editorsfor MISQReview, we quicklylearnedthat many IS scholars are not familiar withthe structureand formatof reviews,whichprovided forthis article.Unfortunately, thisfindingis notuncommon the motivation in other areas, as has been noted: Authorsof literature reviewsare at riskforproducing lists of citationsand mind-numbing that a resemble lots of but not much case, numbers, findings phone book-impressive a coherent review from coherent a plot. [Incontrast] emerges only conceptual strucof the itself. For most this a reviews, turing topic requires guiding theory, a set of a of view the or about point competing models, phenomenon under discussion [Bem 1995, p. 172]. LikeBem in psychology,we seek to encourage more conceptualstructuring of reviews in IS. Drawing on our editorialexperiences, as well as other editors'experiences in related areas (e.g., Daft 1985; Sutton and Staw 1995; Whetten 1989), we believe that we can help prospective MISQReview authorsand IS scholars in general by setting forthsome guidelines and thoughtson how to writea reviewarticle.2 Inthis paper,we firstconsiderwho should writefor MISQReview and identify the types of articlesthatare advice to would-be authorsbased on what we Next, we spend most of the paper providing appropriate. have learned from our experiences. We then discuss the reviewingprocess. Finally,we conclude by summarizingour expectationsfor a reviewarticle.

Prospective Authors and Topics

There are two points in a scholar's life that lend themselves naturally to writinga literature review. First, those who have completed or made substantialprogress on a stream of research are well positionedto tell their colleagues what they have learned and where the field can most fruitfully direct its attention. Second, scholars who have completed a literaturereview priorto embarkingon a project and have developed some theoreticalmodels derivedfromthis revieware also potentialauthors. From another angle, two types of reviews exist. First,authors could deal with a maturetopic where an accumulatedbody of research exists that needs analysis and synthesis. Inthis case, they wouldconduct a thoroughliterature reviewand then propose a conceptualmodel that synthesizes and extends existing research. Second, authorscould tackle an emerging issue that would benefitfromexposure to potential theoreticalfoundations.Here, the reviewof currentliterature on the emerging topic would, of necessity, be shorter. The author's contribution would arise from the fresh theoreticalfoundations proposed in developing a conceptual model. Because literaturereviews are more time-consumingand have fewer outlets than research articles, prospective authorsshould contact the currentsenior editorof MISQReview priorto commencing. An

2Whilethis articleis centered aroundMISQReview, we believe much of what we say has general value for most IS literature reviews.

xiv

MISQuarterly Vol.26 No. 2/June 2002

Webster& Watson/GuestEditorial

on outlineof the reviewwillenable the senior editorto advise whetheranotherauthoris currently working the topic and to give broadguidance on the directionof the work.

Writing a Review Article

of a reviewpaperand provideseveral To assist you in crafting yourpaper,we indicatethe broadstructure issues (e.g., whatshouldbe included suggestions on executingyourreview.We reflecton some pragmatic in the introduction to your paper?) and some more ambiguous issues (e.g., how can you justify a proposition?).Throughout,we provide examples from past articles in a variety of fields to give you exemplarsof how others have addressed these issues.

Beginning YourArticle
untilwell intothe article.Moreover,the In some papers we have received, the topic does not "emerge" to your paper needs to contributions are not clear. Incontrast,to hook your readerearly,the introduction the paper's of yourkey variable(s),and clearlyarticulate motivateyourtopic, providea workingdefinition that includeproviding a new theoreticalunderstanding contributions. contributions Ways of demonstrating helps to explainpreviouslyconfusingresults, notingthat littleresearch has addressed this topic, providing streams calls fromwell-respectedacademics to examinethistopic, bringing togetherpreviously-disparate for practice. of workto help shed lighton a phenomenon, and suggesting important implications of yourkey variablesand set the The next section of yourpapershould providemoreelaboratedefinitions boundarieson your work. Boundaries include issues like level(s) of analysis, temporaland contextual the scope of your review, and your implicitvalues (Bacharach 1989; Whetten 1989). For limitations, example, you should clearlystate the unitor unitsof analysis undertakenin the review;however,be wary of tryingto go beyond a single unitunless you can providea strong rationalefor a multilevel perspective. Further,if your theory applies only to certain contexts (e.g., types of occupations, organizations, or forthe reader. You also need to supportthe countries)or to certaintime periods,this should be identified of what review: state literature and fields scope your you will draw upon and why these define an the values boundingyour for the chosen and level of appropriate boundary topic analysis. Finally,identify interests are whose served is, (such as top managetheory-that your implicit assumptions concerning ment, IS professionals, users, or other stakeholders:Bacharach 1989; livariet al. 1998). To show how these suggestions mightbe implemented,consider Griffith's (1999) paper on "technology features." She motivatesher topic by providing of use of the examples practicerangingfromaboriginals' steel axe to users' concerns with Pentiumchip errors(p. 472). She then articulatesthe contributions by its gaps, (2) suggesting that she will address these short(1) outliningpast research and highlighting comings by proposing new theory, (3) listing academics who have called for this research, and thatthis research has important for practice(pp.473-474). Hersubsequent sec(4) indicating implications tions providedefinitionsof her key concepts and delineate the boundarieson her research (pp. 474-478).

Identifying the Relevant Literature

A high-quality reviewis complete and focuses on concepts. A complete reviewcovers relevantliterature on the topic and is not confined to one research methodology, one set of journals,or one geographic

MISQuarterly Vol.26 No. 2/June 2002

xv

Webster &Watson/Guest Editorial

fromreviewersis thatMISQ Reviewsubmissions focus region.However,a commentwe receivefrequently on a small set of North American or As one reviewer "top" publications. solely tellinglynoted: Studies of the IS literaturehave consistently been limitedby drawingfrom a small sample of journals. Even thoughthe [ones] investigatedhere may have reputationsas our top journals, that does not excuse an author from investigating "all" published articles in a field..../ just can't see the justificationfor searching by journal instead of searching by topic across all relevantjournals. We recommenda structuredapproachto determinethe source materialfor the review: are likelyto be in the leadingjournals.Itmakes sense, therefore,to startwith (1) The majorcontributions of relevantarticles, them. Whilejournaldatabases likeABI/Inform (ProQuest)accelerate identification others not caught by your keyword scanning a journal'stable of contents is a useful way to pinpoint sieve. You should also examine selected conference proceedings, especially those witha reputation for quality. fieldstraddling Because IS is an interdisciplinary otherdisciplines,you often must look not onlywithin the IS discipline when reviewingand developing theory but also outside the field. Malone and in relatedareas like computer Crowston(1994) providean excellent example of reviewingliterature science, economics, operations research, organizationtheory, and biology. Robey et al. (2000) present another admirableexample of reviewingtwo major streams of research to informtheir research topic. (2) Go backwardby reviewingthe citationsfor the articles identifiedin step 1 to determinepriorarticles you should consider. (3) Go forward by using the Web of Science3 (the electronicversionof the Social Sciences CitationIndex) to identifyarticles citingthe key articles identifiedin the previous steps. Determinewhich of these articlesshould be includedin the review. A systematic search shouldensure thatyou accumulatea relatively completecensus of relevantliterature. You can gauge that your review is nearing completionwhen you are not findingnew concepts in your articleset. Ofcourse, you willmiss some articles.Ifthese are critical to the review,however,they are likely to be identifiedby colleagues who read yourpaper either priorto or afteryoursubmission.

Structuring the Review

A literature reviewis concept-centric.Thus, concepts determinethe organizingframework of a review. In contrast,some authorstake an author-centric approachand essentially presenta summaryof the relevant articles. This method fails to synthesize the literature. The two approaches are easily recognized, as in Table 1. illustrated

3http://www.webofscience .com/

xvi

Vol.26 No.2/June 2002 MISQuarterly

Webster& Watson/GuestEditorial

Concept-centric A, authorB, ...] Concept X ... [author

Concept Y ... [author A, author C, ...]

Author-centric AuthorA ... concept X, concept Y, .

Author B ... concept X, concept W,.

Articles A 1 2 K
I I

Concepts B K C K D K

Articles A Unitof analysis 1 2 X O G I O B G X X X I

Concepts C O G I X X 0 D G I O G X

G (group),I (individual) Legend: O (organizational),

To make the transition fromauthor-to concept-centric,we recommendthat you compile a concept matrix as you read each article(Table 2), an idea we have adapted from Salipante et al. (1982). When your readingis complete, synthesize the literature by discussing each identifiedconcept. Before commencing this step, take some time to develop a logicalapproachto groupingand presentingthe key concepts you have uncovered. You mightneed to add a further to handlethe unitof analysis (Table3). dimensionto the concept matrix Forexample, Te'eni(2002) foundthatthe concept "communication had different meaningswhen strategy" and cognitive utterancelevels. Isolatingconcepts considered fromthe organizational, group, individual, by unitof analysis should resultin a crisperreview because it is easier to detect when you let a concept stray outside the scope of its domain.

Vol.26 No. 2/June 2002 MISQuarterly

xvii

Webster& Watson/GuestEditorial

Tables and figurescan be an effectivemeans of communicating major findingsand insights. Nonetheless, tables cannot be merely lists of articles. They need to add value by categorizing articles based on a scheme that helps to define the topic area, such as types of variablesexamined, level of analysis, gaps in the literature, or other important theoreticalissues. on IS success the literature Forinstance, DeLoneand McLean(1992) includea set of tables summarizing by level of analysis, type of study, and success measures. As anotherexample, Bem (1995) describes a review in which past research is categorized by whether the studies support one of three competing models-by doing so, the authors are able to discover a recognizable patternsupportingone of the models. Alaviand Leidner's(2001) table of knowledgetaxonomiesmakes iteasy forthe readerto quickly determinethe meaningof a particular knowledgetype, which is especially useful in a long article.Finally, the six figures and 13 tables in Te'eni's (2001) review articlehelp to communicatehis message more clearly. A reviewsucceeds when it helps otherscholars to make sense of the accumulatedknowledgeon a topic. aroundthe topic's central We believe thatsense-making is enhanced when a reviewis logicallystructured ideas and makes good use of tables and figuresto convey economicallythe keyfindingsand relationships.

Tone
A successful literature reviewconstructively informsthe readeraboutwhat has been learned. Incontrast to specific and criticalreviews of individual papers, tell the readerwhat patternsyou are seeing in the literature.Do not fall intothe trapof being overlycritical,as Daft(1985, p. 198) argued when describing why he rejected some journalsubmissions: ...another indicatorof amateurismwas an overly negative approachto the previous is easy, and of littlevalue; it literature.... Previousworkis always vulnerable.Criticizing to explainhow research buildsupon previousfindingsratherthan to is more important claim previousresearch is inadequateand incompetent. Respect the workof those who laboredto create the foundationforyourcurrentworkby keeping in mind that all research is flawed (McGrath 1982). Of course, you cannot cite others' workblindly-sometimes research is poorlydesigned and conducted, and you willneed to make harddecisions about whetherto include this work in your review or to downplay its significance. Further,if a research stream has a infutureresearch, you willneed to pointthis out in orderto move the common"error" that must be rectified field forward. Ingeneral, though, be faulttolerant. Recognize that knowledgeis accumulatedslowlyin a a reviewarticle. piecemeal fashion and thatwe all make compromises in our research, even when writing

Tense
about priorresearch you should mainlyuse the present or past Opinionis variedon whetherwhen writing tense. When eithertense can communicateequallyeffectively,we opt forthe presentforseveral reasons. First,itgives the readera greatersense of immediacy.Second, when discussing concepts, and in linewith our concept-centricapproachto literature reviews, it is logicalto use the present tense because concepts are always here and now.Third, the presenttense is terserand thus fasterforthe readerto process. There a An author'sopinionscan change withtime. When attributing is an exception to this recommendation. Weber may no longer be saying what statement or idea to a person, therefore,use the past tense: "Max he once said"(Starbuck1999).

xviii

MISQuarterly Vol.26 No. 2/June 2002

Webster& Watson/GuestEditorial

Theoretical Development in YourArticle

A reviewshould identify criticalknowledgegaps and thus motivateresearchers to close this breach. That not of past research, but means makinga chartforfuture a review is, writing only requiresan examination research. Forexample, the MISQReview articlesby Alaviand Leidner(2001) and Te'eni (2001) pinpoint questions for futureinquiry. the discrepancybetween what we knowand what we need to knowalerts other scholars to Highlighting for a key contribution. Usually,this roadmapis accomplished by developing a conceptual opportunities model withsupportingpropositions.Inthis paper,we focus on this traditional approach.However,there For contribution are othermeans of makinga significant instance, (Whetten1989). showinghowcompeting can be theories or philosophicalassumptions explain an important (e.g., phenomenon very influential Allison's[1969] analysis of the Cuban missile crisis). theories or developing new theories willcreate directionsforfutureresearch. However, Extendingcurrent or task and is often the weakest partof a review.Nonetheless, theories is a difficult developing extending it is the most important partof a reviewand generallyneeds the most elaboration.Here,we providesome for researchers who wish to develop a model and justifyits propositions. recommendations and Robey Conceptualmodels are generallyderivedfromvariance (factor)or process theories (Markus independent 1988; Mohr1982). Variancetheories incorporate independentvariablesthat cause variation variables.Incontrast,process theories use events and states to help explaindynamicphenomena.Thus, models may look very differentin the two approaches (see Figure 1 of Langley[1999] for examples of these twotypes of models). Of course, reviewarticlesmay drawfrombothvarianceand process research to develop conceptualmodels to guide futureresearch. Infact, DiMaggio (1995, p. 392) arguedthat"many of these approaches. Moreover, Sabherwal of the best theories are hybrids,combiningthe best qualities" and Robey (1995) demonstratehowthe two approachescan be reconciledin one study.Thus, do nottreat the results of variance and process research as independentelements of a review. Rather,make every effortto show how these two approaches reveal a deeper understandingof the topic. Forexample, the variable,as explanatorypower of a process model mightbe contingenton the strength of a particular Newman and Sabherwal(1991) illustrate. Models and propositionscapture relationshipsbetween variables, but do not, on their own, represent (1999) proposition la (p. 480) states that, theory (Sutton and Staw 1995). For example, Griffith's concrete features are more likelyto be experienced as novel than new/adaptedabstract "New/adapted for these relationshipsrepresents the crucialpartof the features." Rather,the reasoning or justification process. theory-development The reasoningforpropositions maycome fromthree mainsources: theoreticalexplanationsfor"why," past and empirical findings, practiceor experience. The whyor logicalreasoning is the most important compoItrepresents "thetheoreticalglue that nent of the explanation. Itmust always be partof any justification. welds the model together"(Whetten 1989, p. 491). Past empiricalresearch also should be included if it exists. If it does not exist in the specific area of interest, however, empiricalresearch in related areas should be presented as (weaker)support(Gayand Diehl1992). Experience,if available,can also help to itmayarise fromthe author'sown experiences ininteracting withorganizations orfrom justifya proposition; the practiceliterature. Nonetheless, while past findingsand experience can help to supporta proposition, keep in mindthat they are not a substitutefor logical reasoning (Suttonand Staw 1995). As justifyingpropositionsoften represents one of the most challengingaspects of a reviewpaper, let us research concerningtechnologyfeatures, we see look at several examples. First,ifwe examine Griffith's

MISQuarterly Vol.26 No. 2/June 2002

xix

Webster& Watson/Guest Editorial

that she draws on each of the three types of justification sources at varioustimes. For example, for one and gives an example from proposition,she providesa theoreticalexplanation(concerningverification) practice (Lotus Notes). For anotherproposition,she draws on past empiricalresearch in a related area (concerningrewardsystems) and providesan example fromher own experience (at several automobile assembly plants). As anotherexample, examine Moorman and Miner's (1998) paperon organizational memory.Manyof the to use all of For a three sources. propositions types justification example, justify propositionconcerning proceduralmemoryand speed, the authorsfirstdrawon theoryaboutthe automaticqualityof procedural memory(fromcognition),then reporton empirical supportfroma relatedarea (teachers),and end withan from example practice(improvisation duringwar).

Evaluating Your Theory

how do you know Once you have developed yourtheory(such as model, propositions,and justifications), if it is good? This evaluationis difficult and nebulous. Writersargue thatgood theories should be memorableand provideanswers to why.They should explain, predict, and delight (Weick 1995). Others propose that they should be interesting(Davis 1971) yet from falsifiable,and useful(Suttonand Staw 1995). Some arguethattheoriesshouldbe built parsimonious, Thus, they should exhibitcreativity,relevance, and comprehenmultipleparadigms(metatriangulation). siveness (Lewisand Grimes 1999, p. 685). Reviewers are lookingfor good theories, but there is no cookbook approachto accomplishingthis. One important way to assist you in this evaluationprocess is to have colleagues read and comment on your workbefore submittingit for review (Bem 1995). As Daft(1985, p. 207) observed: Witheach revision,the paper ripens. Expose yourpaper to the fresh airand sunshine of collegialfeedback. Witheach discussion, new ideas emerge. The ripeningprocess is facilitatedwithhardworkand frequentrevisions.

Creating Your Discussion and Conclusions

Some reviews end abruptlywith a short conclusion. However, even though you have completed the of yourreviewpaper at this point,you can stilltell yourcolleagues more. Forinstance, returning majority to Griffith, we see that after she justifies her propositions,she goes on to demonstrate how her work extends past research, to suggest ways that her theory can be empiricallyexamined, and to draw for practiceand futuretheorizing(pp. 484-486). implications

The Reviewing and Revision Process

A reviewpaperembodies the "stateof the field." As such, it represents a benchmark forothers conducting futureresearch in your area. You should reap the benefits of citationsto your articlefor years to come because yourworkshould hew a pathfor others. Because of the value and importanceof a high-quality reviewfor the field, the firstchoices for reviewersare currentMISQuarterly Associate Editors.We also

xx

MISQuarterly Vol.26 No. 2/June 2002

Webster& Watson/GuestEditorial

contact senior experts conductingresearch on the particular topicof the review.Thus, reviews are written accomplished scholars. The benefit is that you willreceive a detailed, developmental by well-qualified, review.The downside is that these reviewerswill recognize many opportunities for you to improvethe qualityof yourwork. As a result,the revisionwilltake more effortthan the revisionfor a regulararticle. Whatconcerns have reviewersgenerallypointedout intheircommentson MISQ Review papers? "What's new?"always seems to be highlightedby reviewers, and earlierwe listed ways of demonstratingthe of yourpaper.However,reviewersare lookingfornotone, buta combination of contributions. contributions that "it done will hasn't been before" on its own not convince reviewers. More Saying generally,and your we have foundthat consistentwithWhetten's(1989) most-frequently of reviewer concerns, occurring types reviewers are looking for contribution("what'snew?"), impact ("so what?"), logic ("whyso?"), and thoroughness("well done?"). Youwillbe well on yourway to a publishablepaperifyou can address these fourmajorconcerns when firstsubmitting your paper. research paperintwo majoraspects. First, Respondingto the reviewers'concerns differsfroma traditional on the revisionprocess is generally longer because it takes time to reread and reinterpret the literature whichyourarticleis based. You also willneed to read and integratearticlessuggested by the reviewers. Second, because a revisionto a long paper can involvemanychanges, it is helpfulto develop a planfor the review and share it with the editor and reviewers.A plan clarifies how you will handle the possibly disparaterecommendationsof reviewers.Ifthe reviewersdisagree on how you willreconciletheiradvice, then considerabletime and anguish is saved ifthis divergenceof views is sorted out beforeyou launchon a majorrevision.MISQReview does not requireauthorsto submita plan prior to revision,butwe certainly to consider this urge you option.

Summary
An ideal article: * * * * * * * motivatesthe research topic and explainsthe review'scontributions describes the key concepts delineates the boundariesof the research reviews relevantpriorliterature in IS and relatedareas a model to future research develops guide justifies propositionsby presenting theoreticalexplanations, past empiricalfindings, and practical examples for researchers and managers. presents concludingimplications

And on top of this, the exemplaryreviewarticleshould be explanatoryand creative! articles, so authors have the MISQReview articles are significantlylonger than regularMIS Quarterly We challenge you space to develop such an ideal article.Nevertheless, this task is not straightforward. to craftsuch contributions for MISQReview to move the field forward. LikeSutton and Staw's (1995, p. 380) conclusion about organizationalresearch, we believe that Withoutconstant pressure for theorybuilding, the fieldwould surely slide to its natural resting place in dust-bowlempiricism.

MISQuarterly Vol.26 No. 2/June 2002

xxi

Webster &Watson/Guest Editorial

Acknowledgments
We thank BrentGallupe,TerriGriffith, DorothyLeidner,Ji-Ye Mao, Shan LingPan, Dan Robey, Mikko of this paper.Theirinsightsand experiences Siponen, and DovTe'enifortheircommentson an earlierdraft added considerable valuable.

References
Alavi, M., and Leidner, D. E. "KnowledgeManagement and Knowledge Management Systems: ConceptualFoundationsand Research Issues," MISQuarterly (25:1), March2001, pp. 107-136. Modelsand the CubanMissileCrisis,"TheAmericanPoliticalScience Review Allison,G. T. "Conceptual (63:3), 1969, pp. 689-718. Theories: Some Criteriafor Evaluation," Bacharach, S. B. "Organizational Academy of Management Review (14:4) 1989, pp 496-515. a ReviewArticle for PsychologicalBulletin," Bem, D. J. "Writing PsychologicalBulletin(118:2), 1995, pp. 172-177. in Be Rejected and WhatYou Can Do AboutIt," Daft,R. L. "WhyI RecommendedThatYourManuscript Homewood, Sciences, L. L. Cummingsand P. J. Frost(eds.), Irwin, Publishingin the Organizational IL,1985, pp. 193-209. Davis, M.S. "That'sInteresting!," Philosophyof Social Science (1), 1971, pp. 309-344. DeLone, W. H., and McLean, E. R. "Information Systems Success: The Quest for the Dependent Information Variable," Systems Research (3:1), 1992, pp. 60-95. on 'WhatTheoryIs Not'," Science Quarterly Administrative DiMaggio,P. J. "Comments (40:3), 1995, pp. 391-397. Gay, L. R., and Diehl, P. L. Research Methods for Business and Management,Macmillan,New York, 1992, pp. xxiv,679. T. L. "TechnologyFeatures as Triggersfor Sensemaking,"Academy of Management Review Griffith, (24:3), 1999, pp. 472-488. Information livari,J., Hirschheim,R. A., and Klein,H. K. "AParadigmatic Systems Analysis Contrasting Information DevelopmentApproachesand Methodologies," Systems Research (9:2), 1998, pp. 164193. forTheorizing fromProcess Data," Langley,A. "Strategies Academyof ManagementReview (24:4), 1999, pp. 691-710. Lewis,M.W., and Grimes,A. J. "Metatriangulations: Building Academy TheoryfromMultiple Paradigms," of ManagementReview (24:4), 1999, pp. 672-690. ACMComputing Malone,T. W., and Crowston,K. "TheInterdisciplinary Studyof Coordination," Surveys (26:1), 1994, pp. 87-119. Markus,M.L., and Robey, D. "Information Change: Causal Structurein Technologyand Organizational and Research,"ManagementScience (34:5), 1988, pp. 583-598. Theory J. E. "Dilemmatics:The Study of Research Choices and Dilemmas,"in Judgment Calls in McGrath, J. Martin, and R. A. Kulka(eds.), Sage, BeverlyHills,CA, 1982, pp. 69-102. Research, J. E. McGrath, Behavior(1st' Mohr,L. B. ExplainingOrganizational ed.), Jossey-Bass, San Francisco, 1982, pp. xv, 260. and Organizational C., and Miner,A. S. "Organizational Moorman, Improvisation Academy of Memory," ManagementReview (23:4), 1998, pp. 698-723. Newman,M.,and Sabherwal,R. "Information Systems Development: FourProcess Scenarios withCase Studies,"Journalof Information Systems (5:1), 1991. a General Theoryof Action,Harper& Row, New York,1962. Parsons, T., and Shils, E. A. Toward

xxii

MISQuarterly 2002 Vol.26 No.2/June

Webster &Watson/Guest Editorial

Robey, D., Boudreau,M.-C.,and Rose, G. M. "Information Technologyand Organizational Learning:A Review and Assessment of Research," Accounting, Management,and Information Technologies (10:2), 2000, pp. 125-155. Variance and Process Strategies for StudyingInformation Sabherwal, R., and Robey, D. "Reconciling Information Systems Development," Systems Research (6:4), 1995, pp. 303-323. Salipante, P., Notz, W., and Bigelow, J. "A Matrix Approach to LiteratureReviews,"in Research in Behavior,B. M.Staw and L. L.Cummings(eds.), JAIPress, Greenwich,CT,1982, pp. Organizational 321-348. Starbuck,W. H. "Fussy Professor Starbuck'sCookbook of Handy-DandyPrescriptionsfor Ambitious Academic Authors," 1999, http://www.stern.nyu.edu/-wstarbuc/Writing/Fussy.htm (accessed March 30, 2002). Administrative Science Quarterly Sutton,R. I.,and Staw, B. M. "What TheoryIs Not," (40), 1995, pp. 371384. Modelof Organizational Communication forDesigningIT," MISQuarterly Te'eni,D. "ACognitive-Affective 251-312. (25:2), 2001, pp. 2002. Te'eni, D. Personal Communication, We Haven'tLearned," MISQuarterly Webster, J. (in Lee, A. S.). "What (25:4), 2001, pp. xii-xiii. of 'Theory'," in BlackwellDictionary of Organizational Weick, K. "Definition Behavior,N. Nicholson(ed.), Blackwell,Oxford,1995. Constitutesa TheoreticalContribution?," Whetten, D. A. "What Academyof ManagementReview (14:4), 1989, pp. 490-495.

About the Authors

Jane Webster is the currentsenior editor for MISQReview. She received her Ph.D. from New York and is a professor in the School of Business at Queen's University in Canada. Her research University interestscenter aroundelectronicmonitoring, and communication, computer-basedselecorganizational tion, training,and testing issues. She has published in a varietyof journals, includingMIS Quarterly, Science. Research, and Organization Academy of ManagementJournal,Communication Richard Watson is the J. Rex FuquaDistinguishedChairfor Internet of the Center Strategyand Director for Information of Georgia. He has Systems Leadershipin the TerryCollege of Business, the University publishedin leadingjournalsin several fields as well as authoredbooks on data managementand electroniccommerce and given invitedseminars in nearly20 countries.He is VP of Communications of AIS. He was a senior editorof MISQuarterly with responsibility for MISQReview. He is a visitingprofessorat Agder University College, Norway,and a consultingeditorto John Wiley& Sons.

MISQuarterly Vol.26 No.2/June2002

xxiii