The Motivation for Collaborative Discovery Learning Online and Its

Issues in Accounting EducationVol. 15, No. 2May 2000

The Motivation for CollaborativeDiscovery Learning Online and

Its Application in an InformationSystems Assurance CourseA. Faye Borthick and Donald R. Jones

ABSTRACT: This paper develops the motivation for collaborative discovery learn-ing online and explains its application in a master’s course in information systemsassurance. In discovery learning, participants learn to recognize a problem, char-acterize what a solution would look like, search for relevant information, develop asolution strategy, and execute the chosen strategy. In collaborative discovery learn-ing, participants, immersed in a community of practice, solve problems together.In collaborative discovery learning online, participants seek the knowledge theyneed and solve problems together in a virtual environment. For this purpose, virtualenvironments are characterized by web-based access to resource materials andparticipants’ work and web-based discussions occurring in real time (synchro-nously). This approach to learning prepares students for work environments inwhich new problems are the norm and professionals work collaboratively to solvethem in virtual spaces. The paper makes a case for the course being more effectivethan lecture-based instruction because of its use of collaborative discovery learn-ing online, more accessible because participants may be anywhere they haveInternet access, and more affordable if the development and delivery efforts couldbe leveraged across multiple universities.

INTRODUCTION

This paper develops the motiva-tion for collaborative discoverylearning online and explains its

application in a master’s course in in-formation systems assurance.1 In dis-covery learning, participants learn torecognize a problem, characterize whata solution would look like, search forrelevant information, develop a solu-tion strategy, and execute the chosenstrategy. In collaborative discoverylearning, participants, immersed in acommunity of practice, solve problemstogether. In collaborative discoverylearning online, participants seek the

A. Faye Borthick is a Professor andDonald R. Jones is an Assistant Pro-fessor, both at Georgia State University.

The authors are indebted to Harry Dangel,Walter Herndon, anonymous reviewers,Editor David E. Stout, and Associate Edi-tor Jeffrey Cohen for helpful comments.

knowledge they need and solve prob-lems together in a virtual environment1 In the course, information systems assurance

means providing assurance services for highlycomputerized information systems, where as-surance services are “[I]ndependent profes-sional services that improve the quality of in-formation, or its context, for decision makers”(AICPA 1997).

182 Issues in Accounting Education

that enables web-based access to re-source materials and participants’ workand web-based discussions occurring inreal time (synchronously) and sequen-tially (asynchronously) over theInternet.

Collaborative discovery learningonline was applied in a master’s coursein information systems assurance atGeorgia State University. Studentssolved problems, based on progressivelyricher contexts, such as preparing listsof risks, evaluating internal control, cre-ating assurance plans, and implement-ing assurance plans. For each class ses-sion, students prepared their solutionsto the assigned assurance need, pub-lished them where all participants couldhave web access to them, and developeda group solution during synchronousdiscussion. When they were available,professional auditors joined the synchro-nous discussions.

The article argues that the courseimplemented as collaborative discoverylearning online is more effective thanlecture-based instruction for develop-ing problem-solving skills, more acces-sible because participants may be any-where they have Internet access, andmore affordable to universities and stu-dents if the development and deliveryefforts could be leveraged across mul-tiple universities. This approach tolearning also prepares students forwork environments in which new prob-lems are the norm and professionalswork collaboratively to solve them invirtual spaces.

The article is organized into the fol-lowing sections: a motivation for collabo-rative discovery learning online, an ar-gument for the course as collaborativediscovery learning online being moreeffective than lecture-based instruction,arguments for collaborative discoverylearning online making education moreaccessible and affordable, and ap-

proaches for assessing collaborative dis-covery learning online.

MOTIVATION FORCOLLABORATIVE DISCOVERY

LEARNING ONLINEThis section develops the concept of

collaborative discovery learning onlineby integrating (1) theories in cognitiveand social learning and (2) marketforces that flourish in networked com-puting environments. The cognitivelearning theory is that effective prob-lem solving entails discovery learning.The social learning theory is that col-laboration in a community of practiceprompts learners to develop problem-solving skills. With respect to marketforces, organizations cannot resist us-ing networked computing, symbolizedby the Internet, to achieve the opera-tional flexibility associated with increas-ing their responsiveness to constituents.

Cognitive Aspects of LearningLearning theorists characterize

learning to solve problems as “discov-ery” learning, in which participantslearn to recognize a problem, charac-terize what a solution would look like,search for relevant information, de-velop a solution strategy, and executethe chosen strategy.2 In this character-ization of learning as problem solving,specific facts are irrelevant until theyneed to be brought to bear on a par-ticular problem. In problem solving, theacquisition of facts ceases to be the prin-cipal learning activity. Instead, the fo-cus is on how well one can formulateand evaluate problem representationsand match information to aspects of the

2 Different aspects of this process appear inAnzai and Simon (1979); Kulkarni and Simon(1988); Dunbar (1993); Norman and Spohrer(1996); Brandt (1997); Okada and Simon (1997);Sabelli (1998); Schank (1998).

Borthick and Jones 183

problem during the search for a solu-tion. This puts the premium not onwhat one remembers but on how wellone can characterize relevant informa-tion, find required information, ma-nipulate relationships, shift betweenproblem representations, and make in-ferences. According to this theory, de-veloping one’s ability to solve problemsin a given domain should prepare onefor approaching any problem in thedomain, not just those currently beingsolved in the classroom (Mallach 1996).

Social Aspects of LearningLearning, however, is as much a so-

cial as an individual cognitive phenom-enon (Vygotsky 1978, 1986; Roschelle1992; Scardamalia and Bereiter 1994;Harasim et al. 1995; Wenger 1998). Animplication of learning as a social phe-nomenon is that educational coursesdesigned to transfer knowledge in dis-crete, tidy units in a classroom are notgood preparation for enabling partici-pants to contribute to organizationalgoals that cannot be well specified inadvance, e.g., providing assurance forsystems that deploy advances in infor-mation technology in unanticipatedways. Instead, learning events oughtto prompt participation in the commu-nities of practice3 the learners are en-tering. In a community of practice,members of a group learn from eachother by working together as they de-velop a common sense of purpose, in-cluding a common way of thinkingabout how work gets done and what isnecessary to accomplish a task. In thismode, learners “situate the decomposedtask in the context of the overall socialpractice” (Brown and Duguid 1993, 12).This means that instead of being sim-ply lectures, learning events would bedesigned “so that newcomers can legiti-mately and peripherally participate inauthentic social practice in rich and

productive ways—to, in short, make itpossible for learners to ‘steal’ the knowl-edge they need” (Brown and Duguid1993, 11). This kind of participation isknown as legitimate peripheral partici-pation (Lave 1991; Lave and Wenger1991).4 The peripheral aspect of theparticipation connotes newcomers’ mas-tery of knowledge and skills as theydevelop into fully participating mem-bers of a community.

In many work environments, newproblems are the norm and groups ofprofessionals work together to solvethem (Schrage 1990; Brown andDuguid 1991, 1993, 1998; Gundry1992; Nonaka 1994; Macdonald 1995;Vega and Lacey 1996; Leidner andFuller 1997; Raelin 1997). Combin-ing the cognitive and social aspectsof learning leads to the premise thatimmersing learners in a communityof practice in which they solve prob-lems together (collaborative discoverylearning) is more likely to be effectivein preparing students for the currentwork environment than learningevents characterized by teachersstanding in front of classes dispens-ing knowledge—the “sage on thestage” model. That is, it is more im-portant to help students learn how tofind or create knowledge as they needit and to negotiate its meaning withinthe community of practice rather than

3 Communities of practice signifies “a theory oflearning that starts with this assumption: en-gagement in social practice is the fundamentalprocess by which we learn and so become whowe are. The primary unit of analysis is…theinformal ‘communities of practice’ that peopleform as they pursue shared enterprises overtime” (Wenger 1998, iii).

4 Cognitive scientists agree that human cogni-tion has both cognitive and situational aspects,although they do not always reach consensuson how the two aspects should relate to eachother, as discussed in a special issue of Cogni-tive Science 17(1) (1993) on situated action.


to teach them only what the teacherbelieves they need to know now.5

Market ForcesAs soon as one company creates a

competitive advantage for itself throughadroit use of information and commu-nications technology, other companieshave strong incentives to do likewise.The cascading effect of companies do-ing likewise then creates competitivepressures for other companies to followsuit lest their customers move to trad-ing partners that they deem to be moreresponsive to their needs. Because thesecompetitive advantages have not beensustainable over the long run, even firstmovers are compelled to make even moreeffective use of information technology(Deibert 1997).

In the current business milieu, thepremium is on companies that are ableto use information networks to config-ure production chains to take advantageof opportunities for economies of scaleand scope and of the locations of humantalent, raw materials, suppliers, andmarkets (Castells 1996). For example,once Dell Computer Corporation dem-onstrated how to take orders online andthen orchestrate production for eachcustomer, carmakers tried to figure outhow to do likewise (McWilliams andWhite 1999). General Motors Corpora-tion and Ford Motor Corporation are in-dependently creating online suppliernetworks for all the goods and servicesthey buy (White 1999). In addition tousing the networks for their own pur-chases, the carmakers want their sup-pliers to use the networks to do businesswith each other. For example, thecarmakers intend to lower their costs bybuying steel for resale to their suppli-ers, thereby forcing steel companies toaccept lower prices (Matthews 1999).

The cascading effects of competitiveuse of information networks begetting

even more competition also apply tocompanies’ investments in their intel-lectual capital. The existence of net-works that have become the actual pro-duction chains has spawned thegrowing need for more education foremployees and for more continuing edu-cation for them. The traditional provid-ers of education have been universities.Evidence is accumulating, however,that universities’ quasi monopoly aspurveyors of education and educa-tional credentials is eroding (Vedder1998; Katz and Associates 1999). Signsthat universities are not fulfilling allthe educational needs of employers aremanifest in articles and books chroni-cling companies’ attempts to transformthemselves into team-based learningorganizations.6 Another sign of unmeteducational needs in companies is thegrowth of corporate universities (Davisand Botkin 1994; Moore 1997; Meister1998). Given the growing importanceof continuous learning as an organiza-tional strategy (Nonaka 1994;Macdonald 1995), corporate demandsfor learning are likely to continue.

Companies may have recognizedthat they had unmet educational needs,but as long as learning experiences re-quired the physical presence of ahuman instructor, providing them washeld to be cost prohibitive. That

5 Similar themes, e.g., that learning is a socialactivity in which “teaching is enabling, knowl-edge is understanding, and learning is the ac-tive construction of subject matter”(Christensen et al. 1991, xii), have also arisenin the context of discussion teaching(Christensen et al. 1991) and in the educationliterature (Peters 1966; Garrison and Shale1990; Jonassen et al. 1995; Bonk and King1998).

6 See, for example, Hayes et al. (1988); Schrage(1990); Senge (1990); Womack et al. (1990);Drucker (1992, 1993); Katzenbach and Smith(1993a, 1993b); Davis and Botkin (1994);Laubacher et al. (1997); Downes and Mui(1998); McDermott et al. (1998).


perception is changing, however, asinformation technology becomes morepervasive and less costly (Sangster andLymer 1998; Katz and Associates 1999).Now, managers imagine a world inwhich education/training can be deliv-ered through technology any time andany place a learner needs it. If techni-cal, qualitative, and organizational limi-tations were overcome so that educationbecame available ubiquitously, the limi-tation on learning would be learners’attention and their capacity to learn in-stead of inadequacies of the educationaldelivery system. In such a world, uni-versities would likely have competitors,striving to make education more effec-tive, accessible, and affordable(Stecklow 1994). Furthermore, some ofthese competitors, unlike universities,might be able to avoid maintaining ex-tensive physical facilities and coaxingtenured faculty into learning new in-structional approaches (Blustain et al.1999; Farrington 1999). To the extentthey could avoid these fixed costs andspread development costs across largenumbers of learners, competitors offer-ing media-based models of learningmight experience higher profit margins(Davis and Botkin 1994). Thus, competi-tors could potentially provide more ac-cessible learning experiences at less costthan universities (Vedder 1998). Theinitiatives of competitors thereforeprompt a sense of urgency for universi-ties to make learning experiences moreeffective, accessible, and affordable.

MAKING EDUCATION MOREEFFECTIVE THROUGH

COLLABORATIVE DISCOVERYLEARNING ONLINE

This section explains how collabo-rative discovery learning online wasapplied in a master’s course in infor-mation systems assurance7 at Georgia

State University, an urban researchuniversity. The implementation is ex-plained in terms of course learning ob-jectives, the resources available to stu-dents, what students do before andduring online class sessions, the re-sources participants develop during thecollaborative learning that occursonline, the facilitator’s role, the configu-ration of the computing infrastructurefor the course, results for the onlinecourse, and limitations of these onlineclasses and of collaborative discoverylearning online.

Course Learning ObjectivesThe learning objectives for the

course are for students to learn to iden-tify risks, evaluate internal control, de-velop assurance plans, and implementassurance plans for highly automatedinformation systems (Vygotsky 1978,1986).

From a technique perspective, thecourse develops students’ compe-tence with discrete and continuousapplication audits and information sys-tem development audits. Applicationaudit approaches included computer-assisted audit techniques, digital analy-sis, analytical review, data querying, andcontinuous monitoring. The subject mat-ter on system development audits in-cluded audit and control of waterfalland sync-and-stabilize approaches to

7 The syllabus for the course is available at<http://www.gsu.edu/~accafb/ac863.htm>.Graduate students from other universities canenroll in the course (Acct 8630) as “transientstudents” in order to transfer the credit to theirhome universities. For the procedure for be-ing admitted as a transient student, see <http://www.cba.gsu.edu/graduate/oaa/bulletin/>(Item 6: Master’s Admissions, Transient Stu-dents). The course is also available throughthe Southern Regional Educational Board’selectronic campus <http://www.srec.sreb.org/index.asp>.


development8 and development inobject-oriented and enterprise resource-planning (ERP) environments.

The prerequisite for the course isone of the following: (1) an undergradu-ate course in accounting systems andan undergraduate course in auditing,or (2) a master’s course in applicationprototyping with objects. The co-requi-site for the course is a junior-level pro-gramming course in C++ or Visual Ba-sic. The two different prerequisitepaths, one through accounting and onethrough information systems, result inthe course having a student group withcollectively many of the underlyingskills required to solve problems in in-formation systems assurance.

Resources Available by ClassSession

All the resources for the course,organized by class session, are avail-able from web sites (no textbook isused). These resources include an ex-planation of the activity (what stu-dents are to do) for each class session.Consistent with the learning objec-tives, the activities are one of the fol-lowing: preparing a list of risks andcontrol weaknesses for a specific con-text, preparing an evaluation of in-ternal control; creating an audit plan;developing an implementation of anaudit plan; or writing a manuscripton an information systems assurancetopic to submit to a practitionerjournal.

The directions to students for eachday’s activity incorporate resource ma-terials with study questions. Thesematerials provide explanations of con-cepts, techniques, and applications, andinclude professional guidance pertain-ing to the assurance service. The ini-tial complexity of the activities wasminimized by starting with simpler con-texts and including more explicit expla-

nations in the resource materials, anapproach known as “scaffolding”(Vygotsky 1978, 1986).

An Example Activity in SystemDevelopment Auditing

The first activity associated with au-diting system development (occurringabout midway through the course) il-lustrates the kinds of resource materi-als that were provided to students. Thisactivity required students to develop anaudit plan for a system developmentaudit. In system development audits,the auditor examines the process of cre-ating the system rather than the sys-tem results. The directions to studentswere for them to:

Prepare an audit plan that, if ex-ecuted timely, would have de-tected the system developmentand execution flaws soon enoughfor the publisher to have avoidedthe dysfunctional behavior of themagazine fulfillment system. Tominimize the likelihood that youfail to identify all the unsuccess-ful system development practices,start by making a list of them.Then develop an audit plan thataddresses each one. For the con-text, see Rorer (1997).

Resource materials and study ques-tions for this assignment appear inTable 1. The resources are either pre-scriptive in explaining how to auditsystems development, descriptive ofsystem development failures (where, inhindsight, remedies become apparent),

8 In the waterfall approach to system develop-ment, which was common for large projects inthe 1970s and 1980s, frozen product specifica-tions are the basis for design and constructionof components, which are merged in one largeintegration and testing phase at the end of theproject. In the sync-and-stabilize approach,development is iterative in that componentsare synchronized and stabilized incrementally,e.g., daily or weekly.


TABLE 1Resource Materials for Auditing System Development

Prescriptive in Explaining How to Audit System Development

Doughty, K. 1996. Auditing project management of information systems development.EDPACS 23(7), 1–14.

1. What project management practices are associated with failed projects?2. What should be the objectives for audits of project management?3. How would an auditor achieve each objective?

Dykes, W. C., Jr. 1995. Use application program change control to reduce your risks.IS Audit & Control Journal 6, 9–11.

1. Why is change control so important?2. What limits the kind of change control that can be implemented?

Stanford, J. 1995. The project from hell. Computerworld (9/4), 81–84.1. Why is it so hard to obtain an honest assessment of a project’s status?2. What is the danger in keeping bad news from customers?3. What limits the kind of change control that can be implemented?4. What can be done when developers seem to be too busy to follow their own

procedures, e.g., for change control or testing?5. Why might developers be so proud of circumventing established procedures?6. How should development schedules be set? What happens when they are un-

realistic?7. What finally focuses attention on projects in which development is not on schedule?8. How is it that developers can delude themselves into thinking that subverting

change control is productive?

Wessel, D. 1995. A man who governs credit is denied a Toys ‘R’ Us card. Wall StreetJournal (12/14), B1.

1. In spite of the best-intentioned designs, some computer-made decisions areapt to be misguided or just wrong due to inadequate data being considered.When such instances are called to system users’ attention, how should systemusers behave?

Descriptive of System Development Practices

M. A., and Selby, R. W. 1997. How Microsoft builds software. Communications of theACM 40(6): 53–61.

1. How should development audits of applications developed iteratively, e.g., sync-and-stabilize, differ from those developed with sequential (waterfall)methodology?

2. How does the existence of frequent integrations change what an auditor woulddo in an application-development audit?

3. Which is a riskier approach to systems development—sequential or iterative?Why? Explain the risks that matter.

4. What aspects of parallel development with frequent stabilization reduce therisk of projects falling behind?

(Continued on next page)


or descriptive of system developmentpractices. On the premise that practices“can never be fully captured by insti-tutionalized processes,” the prescriptivematerial about how to audit systemsdevelopment is intentionally minimizedto avoid discouraging “the very inven-tiveness that makes practices effective”(Wenger 1998, 10). Instead, the largershare of the materials creates a diver-sity of contexts in system developmentthrough sagas of failed developmentprojects and illustrations of the rangeof current development practices. Thisapproach to providing materials is con-sistent with resource-based learning(Taylor and Laurillard 1995) andscaffolded instruction (Vygotsky 1978,1986), in which support or assistanceprovided by a guide enables learnersto complete tasks they otherwise wouldbe unable to finish.

Online access to all source and as-signment materials permits all informa-tion to be more available than it wouldbe on paper, which increases the likeli-hood that relevant information can bebrought to bear on problem solving.With online access, performance is lim-ited not by what one remembers but

by how well one can understand theproblem representation and match in-formation to aspects of the problem.Having machine-readable (and thusmachine-searchable) access to the workof others—students and professionals—facilitates the creation of a collabora-tive learning environment and a col-lective intelligence.9

What Students DoBefore each class session, students

prepare their materials for the class,convert them to HTML files, and loadthem on a presentation server that allclass participants can access.

To join a class session, studentsstart a web browser, enter the URL forthe platform software (WebCT), give

TABLE 1 (continued)

5. Compared to a sync-and-stabilize approach, what are the biggest risks of se-quential development?

6. Why doesn’t following a sync-and-stabilize process guarantee on-time and bug-free products?

7. What aspects of sequential development increase the risk of developer turnover?What aspects of any development project tend to promote developer turnover?

Kit, E. 1995. Configuration management. Software Testing in the Real World: Improv-ing the Process. Reading, MA: Addison-Wesley, 42–44; Validation testing, 77–82.

1. What is the likely outcome of ignoring configuration management?2. Why is testing able to show the presence of errors but not their absence?3. In a system development audit, what evidence would give an auditor assurance

that each of unit, integration, usability, function, system, acceptance, and re-gression testing were completed adequately?

9 Lévy (1997, 9–10) views the computerizationand networking of society as having the po-tential to “promote the construction of intelli-gent communities in which our social and cog-nitive potential can be mutually developed andenhanced.” He envisions new informationtechnologies that help “us navigate knowl-edge, and enable us to think collectivelyrather than simply haul masses of informa-tion around with us” (Lévy 1997, 10). His namefor the new architecture for thought enabledby computer-based technologies is collectiveintelligence.


their names and passwords, and opena presentation window and a discus-sion window. The presentation windowenables students to view (and copy) thematerials that any student, guest, orthe facilitator has prepared for thatclass. The discussion window enablesparticipants to converse through theirkeyboards and screens. The discussionscrolls continuously as participants en-ter their comments. The presentationwindow and the discussion windowappear on screen together.

During a class session, studentshave the opportunity to ask questionsabout the resource materials. Once suchquestions have been answered, stu-dents present their approaches to theday’s assurance need and discuss therelative merits of various aspects of dif-ferent approaches, coming to an under-standing of an effective approach bynegotiating among themselves themeanings of concepts, objectives, andapproaches to satisfying objectives(Keegan 1993; Wenger 1998).10

In synchronous discussion, partici-pants must make comments in order tobe perceived as part of the group(Harasim et al. 1995). Even thoughsoftware for synchronous discussiontypically permits participants to view alist of all logged-on participants, com-ments are required for other partici-pants to have assurance that one ispresent and participating. To the ex-tent participants want to belong to thegroup, this setting promotes activerather than passive learning becauseof the active nature of the act of mak-ing comments.

The written focus of synchronousdiscussion has other advantages forlearning. The “exactness of expressionpossible with written language” (Henri1992, 119) permits a precision and per-manence not possible with speech. Inaddition, the absence of social cues for

individual participants tends to reducethe effects of social differences amongparticipants (Dubrovsky et al. 1991).The social cues that are present in face-to-face discussion but absent in syn-chronous discussion comprise the thingsabout a person that are impossible toignore when they are conveyed by sightand sound, e.g., the color, shape, andage of one’s skin and one’s accent.One’s name in the logged-on list mayconvey one’s sex and ethnic heritage,but these are not nearly as salient inthe absence of the visual cues. Becausethe lack of face-to-face presence promptsless attention to social cues, there ismore opportunity for greater focus onthe task (Kiesler 1992), which “chal-lenges participants to become compe-tent” (Davie and Wells 1991, 20). Elimi-nating the traditional cue of the teacherstanding in front of the class also hasthe potential of prompting greater fo-cus on the group’s task (Kiesler et al.1984). Synchronous discussion may beone of the few opportunities people haveto “be judged solely on the basis ofachievement” (Davie and Wells 1991,20).

Because it does not impose the one-speaker-at-a-time sequence required inface-to-face discussions, synchronousdiscussion has the potential to eliminatethe interference of the current speaker’sutterances with other participants’thoughts. There is no interference,known as production blocking (Steiner1972), because participants can compose

10 An analysis of 17.5 minutes of class dialogueanalyzed from a “community of practice” per-spective (Lave and Wenger 1991; Wenger 1998)is available in Borthick (2000). The dialogue isfrom a class session discussing a system devel-opment audit plan in which meaning is negoti-ated for the concept of “fall-through programlogic,” which is the use of computer program-ming statements that permit an erroneousbranch to be made because not all possible dataconditions were anticipated.


their thoughts without interruptionsfrom a speaker. Furthermore, face-to-face discussion limits the amount of par-ticipation that is possible. In the worst-case scenario, the only person engagedin the discussion at any one moment isthe person speaking.

In the synchronous discussion, ev-ery participant gets to contribute with-out being inhibited or distracted bywhat others say. Because everyone iscomposing/editing thoughts at the sametime and then releasing them, the limi-tations of one-speaker-at-a-time dis-course disappear—everyone who wantsto join the discussion may do so, andone’s participation is not hindered bythe utterance of the first person to speak,which, in face-to-face discussion, maychange the nature of the dialog. Thefact that the participation of severalpersons can occur simultaneously hasthe potential to allow persons to par-ticipate whenever they wish and, thus,be more actively engaged in the discus-sion. The more active one’s participa-tion, the more one is presumed to learn.According to an NTL Institute for Ap-plied Behavioral Sciences study, aver-age retention rates for different modesof engagement are: 5 percent, lecture;10 percent, reading; 20 percent, audio-visual; 30 percent, demonstration; 50percent, discussion group; 75 percent,practice by doing; 80 percent, teachother/immediate use (Meister 1998).

Synchronous discussions also over-come the perceived loss of immediacyand interactivity associated with asyn-chronous discussions, in which partici-pants are not logged on at the same time(Hiltz 1994). In synchronous discus-sions, participants get the double ad-vantage of immediate responses fromothers to their questions and contribu-tions and the opportunity to reflect onand edit their comments before releas-ing them to the conversation.

Resources that ParticipantsDevelop in Class

During or after a class session, agroup solution to that day’s assuranceneed can be created and loaded on theserver for future reference. All class dis-cussions are logged, and the log is madeavailable on the server. The existenceof the logs makes note taking duringclass unnecessary; instead, students,knowing that details are being recorded,can devote all their energy to participat-ing in class. Having a log of class dis-cussions permits students unable to joina class session to catch up.

The platform software also sup-ports email and a bulletin board forparticipants to communicate witheach other. Examinations are admin-istered online, and students can ac-cess their scores as well as questionsolutions online.

Table 2 summarizes the implica-tions, as discussed above, of using syn-chronous discussion in courses.

The Facilitator’s RoleIn courses conducted in synchro-

nous discussion, the teacher becomesa facilitator—a guide rather than adispenser of facts. Rather than pre-senting knowledge, the facilitator en-sures, as unobtrusively as possibleduring discussion, that relevantknowledge is brought to bear on thegroup task, that ideas are integrated,that misinformation is attended torapidly, and that the discussion endswith a summary of the group’sprogress on the task (Harasim 1987;Harasim et al. 1995). The facilitator’srole evolves over the duration of thecourse. At the beginning, the facili-tator has the most prominent role. Asparticipants gain experience with syn-chronous discussion, they participatemore, which permits the facilitator toassume a truly facilitative role.


TABLE 2Increasing the Effectiveness, Accessibility, and Affordability

of Education with an Online Course

Category and OnlineCourse Practice Implications

Effectiveness

Synchronous discussion for class 1. Enables collaborative discovery learning inin a community of practice

2. Promotes students’ engagement in learn-ing by shifting to active participationmodes (discussing and doing rather thanreceiving lectures)

3. Permits every student to participate at once,avoiding the production blocking associatedwith face-to-face discussions

4. Encourages students to make reflective,thoughtful comments rather than just say-ing the first thing that occurs to them

5. Focuses discussion on the task and awayfrom nonverbal cues associated with face-to-face classes

6. Promotes students taking responsibility fortheir own learning

7. Helps students learn to identify and constructknowledge as they need it to solve problems

8. Prepares students for work environments inwhich new problems are the norm andgroups of professionals work together tosolve them

9. Prepares students for work environments inwhich virtual work groups are common

10. Enables a wider range of professional gueststo participate in the course, thereby enrichingstudents’ experience

Web-accessible course materials 1. Makes more information from more sourcesmore readily available for problem solving

2. Permits course materials (including theproblems to be solved) to be more currentand relevant compared to paper-based ma-terials fixed as of the beginning of the course

3. Promotes the use of published materials assources of information for problem solvingrather than just as descriptions of knowl-edge to be absorbed



TABLE 2 (Continued)

Web-accessible discussion logs 1. Makes note taking in class sessionsunnecessary

2. Enables the text of class discussions to beused as a source of information, for the cur-rent class and for future classes

3. Permits absent students to catch up withclass sessions they could not join

4. Gives non-native English speakers a recordof class sessions for subsequent study

5. Enables evaluation of the effectiveness ofdifferent learning strategies and facilitators’implementations of them

Student-published web documents 1. Prepares students for working in web-enabledenvironments

2. Permits students to realize the diversity ofapproaches to solving different problems

3. Makes students’ work accessible to otherstudents, who can build on it to solve otherproblems

4. Permits students to study other students’work before class so that they can makethoughtful comments about such workrather than just hearing or seeing it presentedfor the first time in class

Web-delivered,-evaluated, and 1.Supports problem-solving contexts for eval--returned examinations uating performance rather than just fact

recitation2. Allows quicker return of performance evalu-

ation to students and avoids using class timefor handling exam papers

3. Facilitates assessment of learning outcomesby capturing student performance in machine-readable form


Effectiveness



Participating in synchronous dis-cussion requires students to have suf-ficient cognitive maturity to be able toanalyze and apply alternative theoriesor techniques and to develop criteria forjudging which responses are relativelybetter or worse (McCreary and VanDuren 1987; Perry 1970; Hiltz 1994).Upper-level undergraduate and gradu-ate students are generally thought tohave attained sufficient cognitive ma-turity and writing skill to permit their“active and highly readable engage-ment with ideas and new skills” (Hiltz1994, 107).

Even with cognitively mature par-ticipants, achieving discussion objec-tives requires the facilitator to make par-ticipation expectations clear withrespect to, for example, regularity ofparticipation, relevancy of contribu-

tions, responsiveness to other partici-pants, analysis and evaluation of thediscussion, and timeliness of posting ofwork (Harasim et al. 1995; Hiltz 1994).Increasing the difficulty of achievingdiscussion objectives is the fact that untilstudents have practice in synchronousdiscussion, they are apt to behave muchas they do in traditional classrooms—listening but not actively participatingfor a substantial portion of time.

For example, in traditional classes,students often exhibit the “politenesssyndrome,” in which they say only po-lite, nice things about other students’work (Hiltz 1994). This phenomenonis not surprising—it is hard for anyoneto make compelling comments aboutwork that is being presented to themfor the first time, which is typical of thesetting in which groups of students

TABLE 2 (continued)


Virtual class sessions 1. Avoids the need for physical presence to par-ticipate in class sessions, which makes tele-commuting feasible

2. Makes the course accessible to students notable to attend physically, e.g., due to work-related travel requirements or mobility im-pairments

3.Makes the course accessible to visually orhearing-impaired students (with suitableaids)

4.Facilitates participation by professionalguests

Affordability

Leverage across students from 1.Makes the course available to studentsmultiple universities whose home universities do not offer it

2. Enables reciprocal outsourcing of courses toimprove utilization of faculty and universityresources

3. Makes competency-based assessment oflearning outcomes more feasible

Accessibility


present their original work to a classorally, with or without visual aids. Inthe context of an online course, the ex-pectations can be quite different. Ifparticipants are expected to load theirprojects/papers on a web server a fewdays before the discussion, then otherparticipants can be expected to haveexamined the work and be ready toanalyze it. Achieving this objective inthe discussion, however, requires thatthe facilitator make the expectationsclear and that participants learn newbehaviors.

Configuration of the ComputingInfrastructure

The online course as implementedrequires the following web-supportedcapabilities: chat (synchronous discus-sion); chat logging (recording ofdiscussion so that session logs can bepublished to a web site); exam admin-istration and scoring; secure web sitesfor participant-created materials; email;and web sites for facilitator-providedcourse materials at different levels ofsecurity. WebCT was used for chat, chatlogging, exam functions, and secure websites for materials deemed private to theclass, e.g., participant-developed mate-rials for each day’s class and solutionsto completed exams, and for materialsprivate to individual students, e.g., stu-dents’ marked exams.11 Students chosetheir own email system, from the uni-versity or a proprietary Internet Ser-vice Provider (ISP).

Unix servers were used for web sitesfor course materials at different levelsof security. The site for the syllabus andpages explaining each session’s activi-ties was accessible to anyone. The sitecontaining copyrighted materials waspassword-protected in accordance withthe University System of Georgia Boardof Regents’ policy that, subject to fair useprovisions, a password-protected copy is

equivalent to a paper copy in thelibrary.12

University personnel at the centralcomputing facility installed, maintained,and upgraded the Unix servers and theWebCT software platform. They alsocreated server accounts and a WebCTaccount for the course. The instructorreceived one hour of technical consult-ing for configuring WebCT pages froma university support group forWebCT.13 From a technical standpoint,the instructor, who prepared all web-resident materials, was reasonably pro-ficient (but not expert) in web publish-ing in HTML. The instructor receivedone course release for course develop-ment. PC and Internet access was madeavailable for class sessions for studentsthat had on-campus courses immediatelybefore the online course. Most studentsjoined most of the class sessions from off-campus locations.

Results for the Online CourseThis section presents the online par-

ticipants’ reactions to class sessions, acharacterization of the pervasivenessof online participation, and a compari-son of student performance in the onlineclasses with that in the last face-to-faceclass.

Three online sections, one per term,were conducted over an 18-month pe-riod. In all terms, the class met insynchronous discussion on Thursdayevenings in 2.25-hour sessions. Sincethe first online section, the course has

11 Other proprietary platforms, e.g., Blackboard,Convene, and eCollege, provide similar capa-bilities.

12 The University System of Georgia Board ofRegents’ Guide to Understanding Copyrightand Educational Fair Use is available at <http://www.peachnet.edu/admin/legal/copyright/>

13 The one hour was principally devoted to un-derstanding the differences between WebCTand a different platform used in the first twosections.


been taught only online. The same in-structor taught all three online sectionsand the earlier face-to-face section.

Online Participants’ DailyReactions

Students responded at the end ofeach class to questions about their re-actions to that session. A different setof questions was presented for the threekinds of online sessions: daily class dis-cussions, group project/paper discus-sions, and exam completion. To elicitthe full range of student reactions, thesurveys included Likert-type questionson a 1 to 7 scale and open-ended ques-tions. Anchors for the Likert-type ques-tions, grouped by the kind of onlinesession, are shown in Table 3 with themeans.

The surveys were administeredelectronically, as emailed attachmentsin the first two online terms and as webforms in the third online term. Theywere administered for two purposes: toelicit student reactions to each class ses-sion and to acclimatize students to theexam format. For the latter purpose,responses were attributable toindividuals.

For daily class discussions, studentssaid they learned more than they ex-pected (all means on item 6 “Learnedless/learned more than expected” were5.0 or greater). They also indicated thatthey thought the discussion was veryproductive (all means on item 4 “Un-productive/productive interaction” itemwere 5.1 or greater).

The effect of students’ exhibitingsubstantive rather than just polite be-havior is apparent in the shift in par-ticipants’ responses for the group project/paper discussions between the first/sec-ond and third online terms. In the thirdonline term, the facilitator made clearthat the purpose of the discussion of theproject/paper was to test its readiness for

execution (the project) or for submission(the paper) to the target publication. Thisfocus put the emphasis on improving theproject/paper and away from merelysaying polite, nice things about it. Thedifference is evident in the participants’responses. For Item 3 “Other groups’work what/not what I expected,” par-ticipants in Term 3 were more willing torecognize that other groups’ work sur-prised them. Likewise, Term 3 partici-pants indicated greater presentationanxiety in the electronic environment(Item 7), which could be attributed tothe perceived need to make substantiverather than simply perfunctory com-ments about other groups’ work. Simi-larly, Term 3 participants were less ea-ger to comment (Item 8). But given theprospect that others might make com-ments that would help them improvetheir work, participants were much moreeager to receive comments (Item 9).These responses are consistent with theneed for the facilitator to be explicitabout expectations for participation andthe belief that participants can learnnew behaviors that enhance theirlearning.

The means for the exam completionsessions indicated that students in allonline terms felt greater anxiety forelectronic exams than for pencil-paperexams, although the anxiety decreasedover time (Item 7 “Anxiety greater forpencil-paper/electronic exam”). Thisreaction pattern is consistent with stu-dents, over a two-year period, becom-ing more comfortable in virtualenvironments.

In responding to the open-endedquestion “What most surprised youabout today’s discussion?” students in-dicated that they really liked the choicethey had of where to be during classsessions and the existence of thediscussion logs that freed them from tak-ing notes. Non-native English speakers


TABLE 3Attributable Reactions

Means of StudentResponses to Online Class Sessions

Online Terms1 2 3 All

n = 6 n = 14 n = 12 n = 32

Response Anchor Means Response Anchor

Daily Online Class Discussions

1. 1 = Unprepared for class 5.1 5.0 5.0 5.0 7 = Prepared for class2. 1 = Mundane class 5.4 5.7 5.4 5.5 7 = Enlightening class3. 1 = Uncomfortable with soft- 6.0 6.2 5.9 6.0 7 = Comfortable with

ware software4. 1 = Unproductive interaction 5.1 5.3 5.5 5.3 7 = Productive interaction5. 1 = Reluctant to participate 4.9 5.5 4.8 5.1 7 = Eager to participate6. 1 = Learned less than 5.1 5.3 5.0 5.1 7 = Learned more than

expected expected7. 1 = Hostile discussion 5.5 5.8 6.1 5.8 7 = Friendly discussion8. 1 = Pace too fast 4.1 3.9 3.8 3.9 7 = Pace too slow9. 1 = Good ideas at beginning 4.8 5.0 5.1 5.0 7 = Good ideas throughout

Group Project/Paper Discussions

1. 1 = More comfortable 4.1 4.2 4.2 4.2 7 = More comfortablecommenting orally commenting

electronically2. 1 = Unprepared for discussion 4.7 5.0 5.0 4.9 7 = Prepared for discussion3. 1 = Other groups’ work not 4.5 4.5 3.6 4.2 7 = Other groups’ work what

what I expected I expected4. 1 = Oral discussion easier 4.0 3.9 3.7 3.9 7 = Electronic discussion

easier5. 1 = My participation lower 3.7 4.0 3.5 3.7 7 = My participation higher

than I expected than I expected6. 1 = Better feedback orally 4.4 4.3 3.8 4.2 7 = Better feedback

electronically7. 1 = Presentation anxiety 2.7 2.8 4.0 3.2 7 = Presentation anxiety

greater orally greater electronically8. 1 = Reluctant to comment 4.0 4.5 3.7 4.1 7 = Eager to comment9. 1 = Reluctant to receive 5.1 5.0 5.9 5.3 7 = Eager to receive

comments comments

Exam Completion

1. 1 = Prefer pencil-paper exam 3.4 3.4 3.9 3.6 7 = Prefer electronic exam2. 1 = Unprepared for exam 5.1 5.6 5.5 5.4 7 = Prepared for exam3. 1 = Exam not what I expected 3.4 5.0 4.5 4.3 7 = Exam what I expected4. 1 = Pencil-paper exam easier 3.7 3.6 4.3 3.9 7 = Electronic exam easier5. 1 = Score lower than expected 3.0 3.1 3.2 3.1 7 = Score higher than

expected6. 1 = Better feedback for 4.3 3.9 3.6 3.9 7 = Better feedback for

pencil-paper exam electronic exam7. 1 = Anxiety greater for 4.9 4.8 4.2 4.7 7 = Anxiety greater for

pencil-paper exam electronic exam8. 1 = Greater worry about losing 4.4 5.1 4.1 4.5 7 = Greater worry about

pencil-paper exam losing electronic exam9. 1 = Greater worry about confi- 4.2 4.4 3.9 4.2 7 = Greater worry about con-

dentiality of pencil-paper fidentiality of electronicexam exam


commented that they appreciated hav-ing access to discussion logs. The com-ments of one student after reading thelog of a class session that she missed il-lustrate the utility of the logs:

I wish more classes were offeredonline. It’s amazing how manydifferent ideas are offered duringthe discussion. In reading the log,I would try to come up to an an-swer to your question beforereading the responses. It wasamazing how many times therewere answers that were very dif-ferent (both from my answer andone another’s) and yet all valid andhelpful in analyzing the situation.It was nice being able to objec-tively see the comments made bythe other students.

Pervasiveness of ParticipationIn the dialogue from one class ses-

sion analyzed in Borthick (2000), half(four of eight) logged-on participantsentered the discussion. This proportionis more than the few “talkers” com-monly answering most of the questionsin traditional classrooms (Hiltz 1986)and more than the one-fourth to one-third active participants that was theinstructor’s informal measure in earlierface-to-face sections of the course. It isalso consistent with others’ finding ofmore equality of participation in com-puter-mediated than in face-to-facecommunication (Hiltz and Wellman1997). Increased student participationis important because it is associatedwith the development of personal power(Jones 1968). Several self-proclaimedshy students participating in each termof the online course stated they believedtheir participation in the synchronousdiscussion was greater than it wouldhave been in face-to-face discussion,primarily because they could edit theircomments before submitting them.

The facilitator was responsible for

43 percent of the comments. This pro-portion is less than the 60 percent to80 percent generally ascribed to in-structors in face-to-face classes(Duncan and Biddle 1974) and lessthan the proportion occurring in thefacilitator’s face-to-face courses at thesame level. Increased participant talkis important because increases in learn-ing are associated with participation(Flanders 1970).

Across the three online terms, twostudents said they disliked class sessionsconducted as synchronous discussionbecause they missed the nonverbal cuesassociated with face-to-face communi-cations. A strategy for helping thesestudents participate is to encouragethem to prepare beforehand the ques-tions they might ask during the discus-sion, e.g., a question about the mean-ing of a specific concept or its applicationin a specific situation. It is importantfor facilitators of online courses to helpstudents develop participation strate-gies because of the evidence that onceimpediments arise with electronic com-munication channels, it may be diffi-cult for participants to overcome them(Jarvenpaa et al. 1998; Hallowell 1999).Not only is it important for facilitatinglearning in online courses, but beingskillful with electronic communicationis becoming a more important organi-zational skill as more organizations be-come increasingly dependent on the pro-ductive functioning of virtual workgroups (Barua et al. 1995; Apgar 1998;Jarvenpaa et al. 1998; Malone andLaubacher 1998).

Comparison with a TraditionalCourse

An analysis of variance, adjusted forstudent GPA, was performed on finalexam scores between students in the lastface-to-face term and in the three onlineterms. (The last face-to-face term occurred


before the first online term.) The finalexams were a combination of equivalentobjective questions and questions that re-quired the development of audit plans forspecific information systems. There wasno statistically significant difference(F(1,48) = 0.248; p = 0.621) between per-formance of students in the last face-to-face term and in the three online terms.This outcome is consistent with the onlinestudents having mastered as many“facts” and their application as face-to-face students.

Table 4 shows students’ anonymousmean responses to college-administeredstudent evaluations of instructor perfor-mance in four sections of the course, onesection per term: in the last face-to-faceterm and in all three online terms. Thisevaluation instrument is administeredto all students in all courses in the col-lege every term. The numbers at the leftmargin indicate the order of the itemsas they appear on the instrument. Theitems appear exactly as they are wordedon the instrument. The category anditem list in the left column show howthe college groups the items, a configu-ration that was arrived at through fac-tor analysis. The columns with studentresponse means appear in chronologi-cal order: means for the last face-to-faceterm followed by means for three con-secutive online terms. The school meanis the average for all graduate sectionsduring the same term as the secondonline term.

The means in the first online termwere generally lower than they were inthe face-to-face term that preceded it, butthe online course means increased in sub-sequent terms. It is unlikely that this re-sponse pattern across terms is associatedwith changes in students’ performancebecause, as explained above, student per-formance on examinations did not dif-fer in the face-to-face and online terms.It is more likely that the increases over

online terms are a function of improve-ments in the course over time.

Consistent with Alavi’s (1994) re-sults for computer-mediated collabora-tive learning, the course appears to beas effective as the traditional offeringwith respect to traditional learning ob-jectives. In addition, the course is de-signed to impart skills that are not em-braced in traditional educationalsettings, such as working in collabora-tive groups to identify and solve prob-lems, considering more information andmore alternatives for solving problems,cultivating the continuous renewal ofskills, functioning in virtual teams thatform and disband as work dictates, andcreating “communities of practice”(Wenger 1998) that enable organiza-tions to develop their own collective in-telligence (Lévy 1997).

The online course helps studentslearn to function in a workplace in whichprofessionals collaborate with electronictools to achieve team objectives but rarelysee each other due to time or distance con-straints (Gundry 1992; Jarvenpaa andIves 1996; Dennis et al. 1998; Fritz et al.1998). These students will be ready toparticipate in the “temporary, self-man-aged gathering[s] of diverse individualsengaged in a common task” (Malone andLaubacher 1998, 146) that has beenproposed as the model for knowledgework in the future (Laubacher et al.1997).

Impediments to SynchronousDiscussion

The two most significant impedi-ments to synchronous discussion werestudents’ initial unease at followingmulti-threaded discussion, i.e., discus-sion with more than one conversationoccurring at a time, and occasionallapses in connectivity that interrupteddiscussion momentum. A few studentswished they had better typing skills.


TABLE 4Anonymous Reactions

Means of Responses to StudentEvaluations of Instructor Performance

Face-to- SchoolFace Mean at

Category and Item Terma 1 2 3 Term2c

Number of Respondents 16 of 18 4 of 6 10 of 14 8 of 12Presentation Ability 4.6d 4.3 4.7 4.8 4.419. Cares about quality teaching 4.7 4.0 4.5 4.9 4.420. Genuine interest in students 4.5 4.0 4.8 4.9 4.422. Dynamic and energetic person 4.5 4.0 4.6 4.9 4.223. Interesting presentation style 4.3 3.7 4.6 4.9 4.224. Enjoys teaching 4.7 4.3 4.7 5.0 4.425. Enthusiastic about subject 4.9 5.0 5.0 4.8 4.526. Self confident 4.6 5.0 4.7 4.6 4.627. Varies speed and tone of voice 4.4 4.7 4.6 4.5 4.3

Organization/Clarity 4.6 3.7 4.5 4.5 4.25. Well prepared 4.9 4.3 4.7 4.6 4.46. Easy to understand 4.6 4.0 4.3 4.5 4.3

11. Explains clearly 4.7 3.7 4.3 4.3 4.212. Lectures easy to outline 4.5 3.0 4.5 4.3 4.213. Answers questions carefully 4.6 3.7 4.5 4.8 4.214. Summarizes major points 4.5 3.3 4.6 4.4 4.315. States objectives of each class 4.5 3.7 4.8 4.6 4.318. Knows if class understands 4.5 4.0 4.0 4.3 4.0

Grading/Assignments 4.5 4.5 4.5 4.6 4.31. Follows plan in syllabus 4.8 5.0 4.6 4.8 4.52. Assignments related to goals 4.5 4.7 4.6 4.8 4.43. Explains grading clearly 4.3 4.3 4.6 4.4 4.44. Accessible outside class 4.8 4.0 4.7 4.8 4.3

31. Returns exams/papers quickly 4.8 5.0 4.7 4.8 4.432. Reasonable assignments/exams 4.2 4.3 4.0 4.1 4.133. Fair and impartial grader 4.4 4.7 4.6 4.5 4.3

Intellectual/Scholarly 4.7 4.3 4.5 4.6 4.37. Discusses other viewpoints 4.7 3.7 4.5 4.6 4.28. Contrasts theories 4.6 4.3 4.4 4.5 4.39. Discusses recent developments 4.9 5.0 4.7 4.8 4.4

Online Termsb



10. Presents origins of ideas 4.8 4.3 4.3 4.4 4.4

Student Interaction 4.6 4.0 4.6 4.7 4.316. Encourages class discussion 4.8 5.0 5.0 4.8 4.517. Invites criticism of own ideas 4.6 3.0 4.4 4.6 4.221. Treats students as individuals 4.5 4.0 4.3 4.8 4.2

Student Motivation 4.5 4.6 4.5 4.3 4.228. Made me work harder 4.6 4.0 4.7 4.3 4.329. Motivates me to do best work 4.5 4.7 4.3 4.4 4.130. Creative thinking on exams 4.5 5.0 4.5 4.4 4.2

Overall34. Effectiveness of instructor 4.6 4.0 4.0 4.5 4.135. Relative worth of course 4.7 4.7 4.4 4.1 4.1

a The face-to-face term preceded the first online term.b The online terms appear in chronological order.c The school mean is the average for all accounting sections (n = 45) occurring during the same term

as the second online term.d Students responded to each item on a 1–5 scale with higher numbers associated with better perfor-

mance. Category means appear in boldface.

TABLE 4 (Continued)

Face-to- SchoolFace Mean at

Category and Item Terma 1 2 3 Term2c

Online Termsb

Some students found it difficult atfirst to follow multi-threaded discussion,in which there might be two or moresimultaneous conversations. Multipleconversations are unavoidable in syn-chronous discussion because a responseto a specific comment may not appearuntil after other comments on differ-ent topics have been interspersed in thedialogue. When students commented onthe apparent disjointed nature of thediscussion, the facilitator suggestedsome practices to make the conversa-tions easier to follow. The practices in-cluded making an explicit reference toan antecedent comment, especially if along time had elapsed since the ante-cedent, and ending a comment with a“continued next post” (cnp) tag whenthe author was intending to continue

the comment. After participants beganobserving these practices during discus-sion, comments about disjointed discus-sions ceased.

In the authors’ opinion, multi-threaded discussion seems much lessdisjointed when the facilitator helps theparticipants implement a strategy forachieving a session’s objectives. For ex-ample, if the objective for a session is todevelop an assurance plan in a particu-lar context, then one strategy might beto verify that participants understandnew concepts that are germane to theassurance need and the context, promptparticipants to characterize individualassurance objectives, point participantsin the direction of insufficiently treatedaspects (and ensure that they are ad-dressed), ensure that assurance


procedures are appropriate for thecorresponding assurance objectives,and help participants reach closure inthe development of the assurance plan.

In every term, there was an occa-sion on which all participants lost con-nectivity for more than 15 minutes. Inthese instances, it seemed difficult forparticipants to regain the momentumin discussion they had achieved beforethey lost connectivity. After shorteroutages of the whole class or after shortoutages of just a few individuals, dis-cussion momentum seemed unaffected.Regardless of the cause of an outage(likely unknown when it occurs), par-ticipants seemed appreciative of learn-ing, after the fact, what the cause wasand what steps were taken to avoid arecurrence.

LimitationsOmitted Learning Objectives

No discussion of learning objectivesin assurance would be complete with-out treatment of interviewing and oralpresentation skills. In the short run,these objectives were included in otherrequired master’s courses in auditing,financial accounting, and managementaccounting. In the long run, students’interviewing and oral presentationskills could be incorporated into anonline course through videotaping ofstudent performance, individually or ingroups. Students could have the oppor-tunity to re-tape segments that they,their peers, or their coaches deem defi-cient. Streamed video of these perfor-mances could then be made availableto class participants, who could evalu-ate the effectiveness of the interview-ing behavior or oral presentation. If theevaluation were conducted in a syn-chronous class session, an objective ofthe session could be to identify waysthe interviewers/presenters could im-prove their effectiveness for the in-

tended purpose. Although streamingvideo capabilities are available now,their widespread use will likely dependon the general availability of highbandwidth communications links.

Limitations Associated withStudying These Online Classes

The small numbers of students inthe online terms (6, 14, and 12) increasethe difficulty of generalizing the resultsto much larger class sizes. Seven same-college M.B.A. courses that have beentaught online with significant synchro-nous discussion components have beensuccessful. The largest section had 35students, and there was no indicationthat the presence of that many studentswas an impediment to the discussionor to learning. In fact, the instructorsof the M.B.A. courses reported livelyand stimulating discussions.

Generalizability to other instructorsand courses is hindered by the fact thatall terms of the course were taught bythe same instructor. In particular, in-structors viewing themselves as per-formers are probably not good candi-dates for implementing pedagogy inwhich instructors become coaches andfacilitators. Overcoming decades of so-cialization in the “sage on the stage”model in favor of a “guide at the side”model could require more personal in-vestment than some instructors wouldbe willing to make.

Limitations Associated withCollaborative Discovery LearningOnline

To instructors considering imple-menting their courses as collaborativediscovery learning, a significant im-pediment is likely to be the start-uptime required on their part to trans-form courses, to develop proficiencywith the software platform, and to in-tegrate use of software tools into the


courses. Because transformed coursesare so different from their predeces-sors, it is likely that significant coursechanges will continue into multipleterms. There are two reasons for on-going attention to course design. First,it is unlikely that instructors wouldbe able to create the perfect designwith no experience with the approach.Second, even if instructors got thedesign right immediately, businesscontexts change rapidly, which im-plies the need to update activities andproblem contexts in the course.

MAKING EDUCATION MOREACCESSIBLE WITH

INFORMATION TECHNOLOGYConducting class sessions with syn-

chronous discussion makes the onlinecourse more accessible to students be-cause the discussion space is accessiblefrom anywhere through the Internet.Because students do not have to be to-gether physically, the course is accessibleto those working or living elsewhere, tem-porarily or permanently (Harasim et al.1995). Professionals can take the courseregardless of their work locations, andstudents can take it even if it is not avail-able from their home institutions. Moststudents joined the class discussions fromoff-campus locations.

The synchronous discussion of theonline course makes it easier for pro-fessional guests to participate in classes.Information systems assurance is acourse for which there is a ready sup-ply of accountants and auditors willingto enrich student learning by sharingtheir experiences and counsel. Havingonline classes makes it easier for themto join a class and make relevant mate-rials about the accounting professionand business and their organizationsavailable to students.

The professionals joining onlineclass sessions were uniformly very en-

thusiastic about the experience and thepotential of synchronous courses to sat-isfy learning needs of students and pro-fessionals. Analogous to the situationof professionals in face-to-face class ses-sions, the professionals’ comments inthe synchronous sessions were accordeda credibility that is hard for instructors’comments to equal (Nickles and Runde1997). Professionals are currently en-gaged in practice, whereas professors,even though they may have practicedin the past, are not perceived to be asimmediately situated in practice.

MAKING EDUCATION MOREAFFORDABLE WITH

INFORMATION TECHNOLOGYFrom a societal perspective, the

online course makes education moreaffordable to the extent the develop-ment effort and delivery expensecould be leveraged across studentsfrom multiple universities. Not all ac-counting programs currently offer amaster’s information systems assur-ance course. The course, a direct de-scendant of EDP auditing and infor-mation systems auditing courses, hastypically been a specialization course,and, analogous to the situation formany specialty courses in accounting,not every accounting program in auniversity has a faculty member ableand willing to conduct it. By default,a specialty course not offered by astudent’s home institution is usuallysimply unaffordable because studentsare unlikely to be able to spend thetime and expense of physically attend-ing multiple universities.

The accessibility afforded by syn-chronous discussion over the Internetwould, however, permit qualified stu-dents at any university to enroll inthe course. If accounting programs of-fered their specialty courses over theInternet, all parties could be better


off: students would have access to agreater variety of specialty coursestaught by faculty with the skills andinterest in doing so; accounting pro-grams could ensure their students’ en-rollment in courses that were not of-fered locally; faculty efforts could beless fragmented across multiple sub-jects, which could permit more effec-tive use of faculty time; and employ-ers could hire more graduates withpreparation in the specialty areas de-manded in their practices. “[F]ace-to-face classroom interaction limits thereach of each instructor” (Blustain etal. 1999, 51), but use of technologyin alternative delivery methods suchas synchronous discussion could openup new possibilities for making edu-cation more affordable.

ASSESSING COLLABORATIVEDISCOVERY LEARNING

ONLINEThis article has argued that collabo-

rative discovery learning online has thepotential to make learning more effec-tive, accessible, and affordable. Theargument has been illustrated with theexample of an information systems as-surance course.

Certainty about the extent of ben-efits associated with collaborative discov-ery learning online awaits research com-paring results for specific learningobjectives. An informal assessment of theeffectiveness of this approach will be thespeed with which organizations such asprofessional service firms adopt collabo-rative discovering learning online (orvariants of it) for continuing educationof their employees and problem solvingin virtual teams.

For low-enrollment courses like in-formation systems assurance, there maynot be sufficient numbers of students towarrant the development effort of com-petency testing that could yield criterion-

based comparisons. Instead, it might bedesirable to calibrate performance onprincipal learning objectives, e.g., developan assurance plan for a highly auto-mated application, by enlisting the aidof professionals to help develop standard-ized scoring keys for specific contexts.Performance instruments could then beadministered online to students in tradi-tional and online information systemsassurance courses worldwide. Subject tobeing kept current with respect to con-text, standardized scoring could be usedwithin an institution to compare the rela-tive performance of its students acrosstime. Administering performance instru-ments to students worldwide seems muchmore feasible for high-enrollment courses.A group of educators and professionalscould establish competencies for whichperformance measures are designed, vali-dated, and administered to students.Adjusted for systematic sources of varia-tion across students, the results could helpfaculty evaluate the success of their ef-forts to enable learning.

The automatic capture of the dialogueof synchronous class sessions creates anopportunity for the analysis of participantand facilitator behaviors and the learn-ing that emerges. For example, the dia-logue could be examined as a function ofthe design dimensions for a communityof practice: participation and reification,designed teaching and emergent learn-ing, local and global practice, and identi-fication and negotiability (Wenger1998).14

14 Borthick (2000) analyzed one dialogue seg-ment this way with the conclusion that thedialogue achieved the facilitator’s specific ob-jectives for the session (participants develop-ing an understanding of (1) the meaning offall-through logic, (2) how fall-through logiccaused the erroneous printing sequence, and(3) how auditors could detect fall-throughlogic) and a general course objective (partici-pants gaining an understanding of the needfor systems auditors to have competence ininformation systems).


Applying Wenger’s (1998) designdimensions constitutes one approach toevaluating learning from the sociocul-tural perspective (Vygotsky 1978,1986). Other lenses for analyzing dia-logues from the sociocultural andconstructivist viewpoints include:

• Tharp and Gallimore (1988) andTharp’s (1993) seven means of pro-viding learning assistance: model-ing, contingency management,feeding back, instructing, question-ing, cognitive structuring, and taskstructuring, as refined by Melothand Deering (1994) to includeVygotsky’s (1978, 1986) concept ofscaffolding as an explicit facilitatoractivity. In “scaffolding,” an expertor more capable peer provides sup-port or assistance that lets thelearner achieve learning goals thatwould not have been possible with-out the support. Important issuesare the identity of essential ele-ments of the scaffolding and howsoon these elements can be removedwithout threatening learning.

• Selman’s (1980) degree of perspec-tive taking, a developmental theoryof social cognitive skills, especiallyStage 3, in which individuals inter-pret others’ perspectives and canassume a third-person role, andStage 4, in which individuals takeon neutral third-party and multi-dimensional perspectives. The pur-pose of examining students’ per-spectives, as reflected in discussion,would be to devise new ways to helpthem move from their egocentricviews to multidimensional views ofthe world that permit them to col-laborate with others.

• Rogoff’s (1990) categorization ofscaffolded learning as an implemen-tation of zones of proximal devel-opment (ZPD) (Vygotsky 1978,1986). A ZPD is the difference be-

tween what an individual may ac-complish on a task alone and whathe or she may accomplish withguidance or in collaboration withothers. ZPDs are a function of thelearner and the learner’s interac-tion with others and the tools avail-able in common activity. Character-izing ZPDs might permit thedevelopment of strategies for help-ing learners internalize skills andcapacities as independent self-regulatory processes, which wouldenable learners to enlarge theirZPDs.

• Walther’s (1996) framework for ex-amining the development of inter-personal qualities among computer-mediated communication (CMC)users. The framework comprises athree-level matrix of the interper-sonal effects of electronic inter-ac-tivity: impersonality, interper-sonality, and hyperpersonality.Although it has been associatedwith impersonality, CMC has alsobeen associated with greater taskfocus. With respect to interpersonaleffects, use of CMC may be condu-cive to increasing interpersonal ef-fects longitudinally. The hyper-per-sonality effect concerns CMC users’tendency to generalize other par-ticipants into social categoriesrather than attempting to under-stand the attributes of individualparticipants. The interpersonal per-spectives are worth investigatingbecause of the potential for learninghow to enhance group coherencyand effective task performance.

• Zhu’s (1998) synthesis of Hatanoand Inagaki’s (1991) theory ofgroup interaction and Graesser andPerson’s (1994) theory of questionanalysis to focus on horizontal in-teraction among peers and verticalinteraction in which participants


concentrate on more capable partici-pants’ answers. The implication forassessment is that sorting out thecircumstances when horizontal in-teraction and vertical interactionmight be more effective might leadto better strategies for guiding syn-chronous discussion.

• Bloom’s (1956) taxonomy as ameans of characterizing the levelof individuals’ statements andquestions. Using the taxonomy is auseful way or organizing partici-pants’ questions by their cognitiveskill level. Comparing average cog-nitive skill levels across differentdiscussion session might be helpfulin determining whether partici-

pants are attaining higher cogni-tive levels.

Except for Bloom (1956) andSelman (1980), these theories aroseafter electronic communication becamecommonplace. Theories arising afterthe mid-1980s explicitly recognize thepossibilities inherent in electronic com-munication to foster learning in socialcontexts (Bonk and King 1998). Nosingle one of these approaches is likelyto answer all the questions about learn-ing through electronic communication,but, applied together, they and theo-ries-in-the-making may guide coursedesigners in realizing the learning po-tential discussed in this article.

REFERENCESAlavi, M. 1994. Computer-mediated collaborative learning: An empirical evalu-

ation. MIS Quarterly (June): 159–174.American Institute of Certified Public Accountants (AICPA). 1997. Assur-

ance Services. Report of the Special Committee on Assurance Services.New York, NY: AICPA. http://www.aicpa.org.

Anzai, Y., and H. A. Simon. 1979. The theory of learning by doing. Psycho-logical Review 86 (2): 124–140.

Apgar IV, M. 1998. The alternative workplace: Changing where and howpeople work. Harvard Business Review (May/June): 121–136.

Barua, A., C.-H. S. Lee, and A. B. Whinston. 1995. Incentives and computingsystems for team-based organizations. Organization Science 6 (4):487–504.

Bloom, B. 1956. Taxonomy of Educational Objectives: The Classification ofEducational Goals, By a Committee of College and University Examin-ers. New York, NY: Longmans, Green.

Blustain, H., P. Goldstein, and G. Lozier. 1999. Assessing the new competi-tive landscape. In Dancing with the Devil: Information Technology andthe New Competition in Higher Education, edited by R. N. Katz andAssociates. San Francisco, CA: Jossey-Bass.

Bonk, C. J., and K. S. King, eds. 1998. Electronic Collaborators: Learner-Centered Technologies for Literacy, Apprenticeship, and Discourse.Mahwah, NJ: Erlbaum.

Borthick, A. F. 2000. Analysis of design from a community of practice dialog:Negotiating the meaning of auditing information system development.Journal of Information Systems 14.

Brandt, D. S. 1997. Constructivism: Teaching for understanding of theInternet. Communications of the ACM 40 (10): 112–117.

Brown, J., and P. Duguid. 1991. Organizational learning and communities ofpractice: Toward a unified view of working, learning, and innovation.Organizational Science 2(1): 40–57.


———, and ———. 1993. Stolen knowledge. Educational Technology (March):10–15.

———, and ———. 1998. Organizing knowledge. California ManagementReview 40 (3): 90–111.

Castells, M. 1996. The Rise of the Network Society. Boston, MA: Blackwell.Christensen, C. R., D. A. Garvin, and A. Sweet. 1991. Education for Judg-

ment: The Artistry of Discussion Leadership. Boston, MA: Harvard Busi-ness School.

Davie, L. E., and R. Wells. 1991. Empowering the learner through com-puter-mediated communication. American Journal of Distance Educa-tion 5 (1): 15–23.

Davis, S., and J. Botkin. 1994. The Monster Under the Bed. New York, NY:Simon & Schuster.

Deibert, R. J. 1997. Parchment, Printing, and Hypermedia: Communicationin World Order Transformation. New York, NY: Columbia University.

Dennis, A. R., S. K. Pootheri, and V. L. Natarajan. 1998. Lessons from theearly adopters of web groupware. Journal of Management InformationSystems 14 (4): 65–86.

Downes, L., and C. Mui. 1998. Unleashing the Killer App: Digital Strategiesfor Market Dominance. Boston, MA: Harvard Business School Press.

Drucker, P. F. 1992. The new society of organizations. Harvard BusinessReview 70 (5): 95–104.

———. 1993. Post-Capitalist Society. New York, NY: HarperCollins.Dubrovsky, V., S. Kiesler, and B. Sethna. 1991. The equalization phenom-

enon: Status effects in computer-mediated and face-to-face decision-making groups. Human Computer Interaction 6: 119–146.

Dunbar, K. 1993. Concept discovery in a scientific domain. Cognitive Science17: 397–434.

Duncan, M., and B. Biddle. 1974. The Study of Teaching. New York, NY:Holt, Rinehart and Winston.

Farrington, G. C. 1999. The new technologies and the future of residentialundergraduate education. In Dancing with the Devil: Information Tech-nology and the New Competition in Higher Education, edited by R. N.Katz and Associates. San Francisco, CA: Jossey-Bass.

Flanders, N. A. 1970. Analyzing Teacher Behavior. Reading, MA: Addison-Wesley.

Fritz, M. B. W., S. Narasimhan, and H.-S. Rhee. 1998. Communication andcoordination in the virtual office. Journal of Management InformationSystems 14 (4): 7–28.

Garrison, D., and D. Shale. 1990. Education at a Distance: From Issues toPractice. Melbourne, FL: Krieger.

Graesser, A. C., and N. K. Person. 1994. Question asking during tutoring.American Educational Research Journal 31(1): 104–137.

Gundry, J. 1992. Understanding collaborative learning in networked organi-zations. In Collaborative Learning Through Computer Conferencing,edited by A. R. Kaye. 167–178. Berlin, Germany: Springer-Verlag.

Hallowell, E. M. 1999. The human moment at work. Harvard Business Re-view (January-February): 58–66.

Harasim, L. 1987. Teaching and learning online: Issues in computer-medi-ated graduate courses. Canadian Journal of Educational Communica-tion 16 (2): 117–135. Cited in Hiltz, S. R. 1994. The Virtual Classroom:Learning Without Limits via Computer Networks. Norwood, NJ: Ablex.


———, S. R. Hiltz, L. Teles, and M. Turoff. 1995. Learning Networks: AField Guide to Teaching and Learning Online. Cambridge, MA: MITPress.

Hatano, C., and K. Inagaki. 1991. Sharing cognition through collective com-prehension activity. In Perspectives on Socially Shared Cognition, ed-ited by L. Resnick, J. Levine, and S. D. Teasley, 331–349. Washington,D.C.: American Psychology Association.

Hayes, R. H., S. C. Wheelwright, and K. B. Clark. 1988. Dynamic Manufac-turing: Creating the Learning Organization. New York, NY: Free Press.

Henri, F. 1992. Computer conferencing and content analysis. In Collabora-tive Learning Through Computer Conferencing, edited by A. R. Kaye,117–136. Berlin, Germany: Springer-Verlag.

Hiltz, S. R. 1986. The “virtual classroom”: Using computer-mediated com-munication for university teaching. Journal of Communication 36 (2):95–104.

———. 1994. The Virtual Classroom: Learning Without Limits via ComputerNetworks. Norwood, NJ: Ablex.

———, and B. Wellman. 1997. Asynchronous learning networks as a virtualclassroom. Communications of the ACM 40 (9): 44–49.

Jarvenpaa, S. L., and B. Ives. 1996. Introducing transformational informa-tion technologies: The case of the World Wide Web technology. Inter-national Journal of Electronic Commerce 1 (1): 95–126.

———, K. Knoll, and D. R. Leidner. 1998. Is anybody out there? Antecedentsof trust in global virtual teams. Journal of Management InformationSystems 14 (4): 29–64.

Jonassen, D., M. Davidson, M. Collins, J. Campbell, and B. B. Haag. 1995.Constructivism and computer-mediated communication in distance edu-cation. The American Journal of Distance Education 9 (2): 7–26.

Jones, R. M. 1968. Fantasy and Feeling in Education. New York, NY: NewYork University Press.

Katz, R. N., and Associates. 1999. Dancing with the Devil: Information Tech-nology and the New Competition in Higher Education. San Francisco,CA: Jossey-Bass.

Katzenbach, J. R., and D. K. Smith. 1993a. The discipline of teams. HarvardBusiness Review 71 (2): 111–120.

———, and ———. 1993b. The Wisdom of Teams. Boston, MA: Harvard Busi-ness School Press.

Keegan, D. 1993. Reintegration of the teaching acts. In Theoretical Prin-ciples of Distance Education, edited by D. Keegan. London, U.K.:Routledge.

Kiesler, S., J. Siegel, and T. W. McGuire. 1984. Social psychological aspectsof computer-mediated communication. American Psychologist 39 (10):1123–1134.

———. 1992. Talking, teaching, and learning in network groups: Lessonsfrom research. In Collaborative Learning Through ComputerConferencing, edited by A. R. Kaye, 147–165. Berlin, Germany: Springer-Verlag.

Kulkarni, D., and H. A. Simon. 1988. The processes of scientific discovery:The strategy of experimentation. Cognitive Science 12: 139–175.


Laubacher, R. J., and T. W. Malone, and the MIT Scenario Working Group.1997. Two scenarios for 21st century organizations: Shifting networks ofsmall firms or all-encompassing “virtual countries”? MIT Initiative on In-venting the Organizations of the 21st Century Working Paper No. 001,Massachusetts Institute of Technology. http://ccs.mit.edu/21c/21CWP001.html.

Lave, J. 1991. Socially shared cognition. In Perspectives on Socially SharedCognition, edited by L. B. Resnick, J. M. Levine, and S. D. Teasley.Washington, D.C.: American Psychological Association.

———, and E. Wenger. 1991. Situated Learning: Legitimate Peripheral Par-ticipation. Cambridge, MA: Cambridge University Press.

Leidner, D. E., and M. Fuller. 1997. Improving student learning of concep-tual information: GSS supported collaborative learning vs. individualconstructive learning. Decision Support Systems 20: 149–163.

Lévy, P. 1997. Collective Intelligence: Mankind’s Emerging World inCyberspace. Translated from the French under the title L’intelligenceCollective: Pour Une Anthropologie du Cyberspace by R. Bononno. 1995.New York, NY: Plenum.

Macdonald, S. 1995. Learning to change: An information perspective on learn-ing in the organization. Organization Science 6 (5): 557–568.

Mallach, E. G. 1996. Education and training aren’t the same. Computerworld(October 7): 37.

Malone, T. W., and R. J. Laubacher. 1998. The dawn of the e-lance economy.Harvard Business Review 76 (5): 144–152.

Mathews, R. G. 1999. Steelmakers face online threat from auto industry.Wall Street Journal (November 26): B4.

McCreary, E. K., and J. Van Duren. 1987. Educational applications of com-puter conferencing. Canadian Journal of Educational Communication16 (2): 107–115. Cited in Hiltz, S. R. 1994. The Virtual Classroom: Learn-ing Without Limits via Computer Networks. Norwood, NJ: Ablex.

McDermott, L. C., N. Brawley, and W. W. Waite. 1998. World-Class Teams:Working Across Borders. New York, NY: Wiley.

McWilliams, G., and J. B. White. 1999. Dell to Detroit: Get into gear online!The Wall Street Journal (December 1): B1.

Meister, J. C. 1998. Corporate Universities: Lessons in Building a World-Class Work Force. New York, NY: McGraw-Hill.

Meloth, M., and P. Deering. 1994. Task talk and task awareness under dif-ferent cooperative learning conditions. American Educational ResearchJournal 31 (1): 138–165.

Moore, T. E. 1997. The corporate university: Transforming managementeducation. Accounting Horizons 11(1): 77–85.

Nickles, S. H., and C. Runde. 1997. Collaborative legal education. In The Learn-ing Revolution: The Challenge of Information Technology in the Academy,edited by D. G. Oblinger, and S. C. Rush, 129–137. Bolton, MA: Anker.

Nonaka, I. 1994. A dynamic theory of organizational knowledge creation. Or-ganization Science 5 (1): 14–37.

Norman, D. A., and J. C. Spohrer. 1996. Learner-centered education. Com-munications of the ACM 39 (1): 24–27.

Okada, T., and H. A. Simon. 1997. Collaborative discovery in a scientificdomain. Cognitive Science 21 (2): 109–146.

Perry, W. G., Jr. 1970. Forms of Intellectual and Ethical Development in theCollege Years. New York, NY: Holt, Rinehart and Winston.


Peters, R. 1966. Ethics and Education. London, U.K.: Allen & Unwin.Raelin, J. A. 1997. A model of work-based learning. Organization Science 8

(6): 563–578.Rogoff, B. 1990. Apprenticeship in Thinking: Cognitive Development in Social

Context. New York, NY: Oxford University Press.Rorer, S. 1997. A case of catastrophic system failure. EDPACS (January):

11–15.Roschelle, J. 1992. Learning by collaborating: Convergent conceptual change.

The Journal of the Learning Sciences 3 (3): 235–276.Sabelli, N. H. 1998. We are no longer a priesthood. Communications of the

ACM 41 (1): 20–21.Sangster, A., and A. Lymer. 1998. How to survive a new educational era.

Issues in Accounting Education 13 (4): 1095–1109.Scardamalia, M., and C. Bereiter. 1994. Computer support for knowledge-

building communities. The Journal of the Learning Sciences 3 (3):265–283.

Schank, R. C. 1998. Horses for courses. Communications of the ACM 41 (7):23–25.

Schrage, M. 1990. Shared Minds: The New Technologies of Collaboration.New York, NY: Random House.

Selman, R. 1980. Taking another’s perspective: Role-taking development inearly childhood. Child Development 42: 1721–1734.

Senge, P. M. 1990. The Fifth Discipline: The Art and Practice of the LearningOrganization. New York, NY: Doubleday.

Stecklow, S. 1994. Virtual U.: At Phoenix University, class can be anywhere—even in cyberspace. Wall Street Journal (September 12): A1.

Steiner, I. D. 1972. Group Process and Productivity. San Diego, CA: Aca-demic Press.

Taylor, J., and D. Laurillard. 1995. Supporting resource based learning. InInformation Technology and Society, edited by N. Heap, R. Thomas, G.Einon, R. Mason, and H. Mackay, 237–248. London, U.K.: Sage.

Tharp, R., and R. Gallimore. 1988. Rousing Minds to Life: Teaching, Learn-ing, and Schooling in Social Context. New York, NY: Cambridge Uni-versity Press.

———. 1993. Institutional and social context of educational reform: Practiceand reform. In Contexts for Learning: Sociocultural Dynamics inChildren’s Development, edited by E. A. Forman, N. Minnick, and C. A.Stone, 269–282. New York, NY: Oxford University Press.

Vedder, R. 1998. Higher education at lower cost. Wall Street Journal (Au-gust 31): A18.

Vega, G., and J. W. Lacey. 1996. The new social contract: 10 suggestions forchange. Social Policy (Summer): 56–63.

Vygotsky, L. S. 1978. Mind in Society: The Development of Higher Psycho-logical Processes. Edited by M. Cole, V. John-Steiner, S. Scribner, andE. Souberman. Cambridge, MA: Harvard University Press.

———. 1986. Thought and Language. Cambridge, MA: MIT Press.Walther, J. B. 1996. Computer-mediated communication: Impersonal, inter-

personal and hyperpersonal interaction. Communication Researcher 23(1): 3–43.

Wenger, E. 1998. Communities of Practice. Cambridge, MA: Cambridge Uni-versity Press.

White, G. L. 1999. How GM, Ford, think Web can make splash on the factory


floor. Wall Street Journal (December 3): A1.Womack, J. P., D. T. Jones, and D. Roos. 1990. The Machine That Changed

the World. New York, NY: HarperCollins.Zhu, E. 1998. Learning and mentoring: Electronic discussion in a distance-

learning course. In Electronic Collaborators: Learner-Centered Tech-nologies for Literacy, Apprenticeship, and Discourse, edited by C. J.Bonk, and K. S. King, 233–259. Mahwah, NJ: Erlbaum.

Documents

The Motivation for Collaborative Discovery Learning Online and Its