compensatory adaptation to media obstaclescits.tamiu.edu › kock › pubs › journals ›...

Information Resources Management Journal, 18(2), 41-67, April-June 2005 41

Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without writtenpermission of Idea Group Inc. is prohibited.

Compensatory Adaptationto Media Obstacles:An Experimental Study ofProcess Redesign Dyads

Ned Kock, Texas A&M International University, USA

ABSTRACT

Past research on electronic communication media suggests that those media pose obstacles tocommunication in collaborative tasks when compared with the face-to-face medium. Yet, pastresearch also points at mixed findings in connection with the quality of the outcomes ofcollaborative tasks, generally suggesting that the use of electronic communication media hasno negative effect on those outcomes. A new theoretical framework building on human evolutiontheory, called compensatory adaptation theory, has been proposed to explain these contradictoryfindings. This study provides a review and test of compensatory adaptation theory. It investigatesthe impact of the use of an electronic communication medium on 20 business process redesigndyads involving managers and professionals at a large defense contractor, with a focus oncognitive effort, communication ambiguity, message preparation, fluency, and task outcomequality. The study suggests that even though the use of electronic communication media seemedto increase cognitive effort and communication ambiguity, it had a neutral impact on taskoutcome quality. These results appear to be an outcome of compensatory adaptation, wherebythe members of the dyads interacting through the electronic communication medium modifiedtheir behavior in order to compensate for the obstacles posed by the medium, which is suggestedby a decrease in fluency and an increase in message preparation. The results generally supportpredictions based on compensatory adaptation theory.

Keywords: biological influences; communication media; compensatory adaptation; computer-mediated communication; electronic communication; human factors

INTRODUCTION

Research on the effects of technolo-gies on people in business settings has along history. Within that research tradition,few research topics have received so muchsustained attention over such a long period

of time as “electronic communication” —that is, the study of communication throughelectronic media created by artifacts suchas the telephone, fax, and computer. Thisarea of inquiry has taken different formsand different names over the years, suchas computer-supported cooperative work,

42 Information Resources Management Journal, 18(2), 41-67, April-June 2005


computer-mediated communication,groupware, group support systems, andmore recently, a variety of “e” combina-tions (where “e” stands for “electronic”)such as e-collaboration and e-commerce(Grudin, 1994; Davenport, 2000; Dennis,Carte, & Kelly, 2003; Fingar, Aronica, &Maizlish, 2001; Kock, Hilmer, Standing, &Clark, 2000; Kock, Davison, Ocker, &Wazlawick, 2001; Standing & Benson,2000). While these different varietiespresent unique characteristics that identifythem as distinct “research schools,” theyall share the same common interest in elec-tronic communication tools and their effectson human behavior. The advent of theInternet, e-business, and the proliferationof low-cost computer networks and elec-tronic communication tools have led to in-creased interest in research on how elec-tronic communication media affect collabo-rative work in organizations.

The above interest is shared by theU.S. Department of Defense (DoD), whereInternet-based computer networks haveremoved geographical and time constraintsto collaboration among distributed processteams engaged in defense acquisition ac-tivities. With the growth of distributed ac-quisition process teams also comes the chal-lenge of improving defense acquisition pro-cesses in a distributed manner, since newtechnologies, regulatory modifications, andother change drivers constantly push theDoD into rethinking and redesigning the wayit procures, purchases, and internally dis-tributes products and services. This can beaccomplished through distributed and asyn-chronous process redesign groups sup-ported by Internet-based electronic com-munication tools. Yet little is known aboutthe effects of electronic communicationmedia on process redesign groups, particu-larly in the defense sector. This study triesto fill this gap by conducting a preliminary

investigation of the effects of electroniccommunication media on process redesigndyads (i.e., pairs) targeting defense acqui-sition processes.

This article is organized as follows. Itfirst reviews different theoretical perspec-tives that seem contradictory, and that per-tain to the adequacy of electronic commu-nication as a full or partial replacement toface-to-face communication in organiza-tional settings. Next, it discusses a newtheory that tries to address those contra-dictions, namely compensatory adaptationtheory. The theory is discussed particularlyin connection with its two main theoreticalprinciples of media naturalness and com-pensatory adaptation. The article thendevelops a set of hypotheses that are em-pirically tested through a field study of 20business process redesign dyads involvingmanagers and professionals at a large de-fense contractor, with a focus on cognitiveeffort, communication ambiguity, messagepreparation, fluency, and task outcome qual-ity. The article concludes with a discussionof the findings and implications for practi-tioners.

RESEARCH BACKGROUND

It has long been theorized that theface-to-face communication medium pos-sesses inherent characteristics that makeit more appropriate for the conduct of avariety of collaborative tasks (Daft &Lengel, 1986; Graetz, Boyle, Kimble, Th-ompson, & Garloch, 1998; Sallnas,Rassmus-Grohn, & Sjostrom, 2000; Short,Williams, & Christie, 1976; Warkentin,Sayeed, & Hightower, 1997). This has ledto the conclusion that the use of electroniccommunication media, which usually do notincorporate all of the elements present inthe face-to-face communication medium(e.g., synchronicity, ability to convey tone



of voice and facial expressions) will leadto decreased effectiveness in communica-tion interactions and thus to decreased qual-ity of outcomes in collaborative tasks.

In the human-computer interaction lit-erature, one field study in the early 1990s(Nardi et al., 1993) and two more recentexperimental studies (Basdogan, Ho,Srinivasan, & Slater, 2000; Sallnas et al.,2000) provide compelling evidence support-ing in part the above conclusions. The fieldstudy conducted by Nardi et al. (1993) onthe use of video and audio conferencingsystems in the operating room of a hospitalis particularly interesting because it shedsnew light on previous claims that the addi-tion of a video channel (usually in the formof what the authors refer to as “talkingheads”) to an existing audio channel in ane-communication medium usually does notsignificantly enhance performance in col-laborative tasks (see their paper for a re-view of previous research on the topic).Nardi et al. (1993) suggest that the resultsof previous studies might have been biasedby technology operation and experimentaldesign problems, and provide unequivocalevidence of the usefulness of video in ad-dition to audio as a “cognitive effort re-ducer” in a variety of complex and fast-paced interactions between individuals in-volved in surgical procedures.

The experimental studies conductedby Basdogan et al. (2000) and Sallnas etal. (2000) provide additional support for thenotion that the face-to-face communica-tion medium possesses characteristics thatmake it particularly appropriate for the con-duct of a variety of collaborative tasks.They examined the role of touch, or “hap-tic,” feedback in the execution of collabo-rative tasks in distributed virtual environ-ments. Both studies involved data collec-tion and analysis regarding several pairs ofindividuals collaborating through a shared

virtual environment to perform simple taskswith and without haptic feedback. Sallnaset al.’s (2000) study involved more subjectsand higher task variety and complexity thanBasdogan et al.’s (2000) study. The simi-larity of their findings is remarkable. Bothstudies found that haptic feedback signifi-cantly improved task performance, withSallnas et al.’s (2000) study offering addi-tional evidence linking a reduction in cog-nitive effort with haptic feedback: “[Theanalysis of the perceptions by subjects]suggests that it was easier to manipulateand understand the interface when the in-teraction was supported by haptic forcefeedback” (p. 474).

Two theories are well aligned with theabove conclusions in connection, with theadvantages offered by the face-to-facemedium over electronic media (particularlymedia that suppress many of the elementsfound in face-to-face communication) inconnection with variety of collaborativetasks. Those two theories are the socialpresence theory (Short et al., 1976) andthe media richness theory (Daft & Lengel,1986; Daft, Lengel, & Trevino, 1987).

Social presence theory (Short et al.,1976) conceptualizes different communi-cation media along a one-dimensional con-tinuum of “social presence,” where thedegree of social presence is equated to thedegree of “awareness” of the other per-son in a communication interaction. Accord-ing to social presence theory, communica-tion is effective if the communication me-dium has the appropriate social presencerequired for the level of interpersonal in-volvement required for a task. On a con-tinuum of social presence, the face-to-facemedium is considered to have the most so-cial presence, whereas written, text-basedcommunication, the least.

Similarly to the social presence theory,media richness theory (Daft & Lengel,



1986; Daft et al., 1987) classifies commu-nication media along a continuum of “rich-ness,” where richness is based on the abil-ity of media to carry non-verbal cues, pro-vide rapid feedback, convey personalitytraits, and support the use of natural lan-guage. A reasonable interpretation of themedia richness theory’s core argument isthat decisions regarding matching media tocollaborative tasks are based on the needto reduce discussion ambiguity. The face-to-face communication medium is gener-ally considered as among the richest andmost effective media for reducing discus-sion ambiguity (Daft & Lengel, 1986). Incontrast, electronic communication mediain general are not considered rich becauseof their inherent limitations in, for example,carrying non-verbal cues (Daft et al., 1987;Lee, 1994).

Many past empirical findings, particu-larly from the organizational research lit-erature, have supported in part the socialpresence and media richness theories (Daftet al., 1987; Fulk, Schmitz, & Steinfield,1990; Rice, 1993; Rice & Shook, 1990).However, among the key problems with thesocial presence and media richness theo-ries was that they proposed theoretical linksbetween low social presence and low rich-ness, respectively, in communication me-dia, and either: (a) avoidance by users touse those media for collaborative tasks; or(b) low quality of the outcomes of col-laborative tasks, if the users decide to usethose communication media (Daft et al.,1987; Lengel & Daft, 1988; Short et al.,1976).

Several empirical studies have sug-gested that these hypothesized theoreticallinks are wrong, particularly because otherfactors such as social influences and geo-graphic distribution can both lead users tochoose “lean” communication media and

modify their behavior in ways that are in-dependent of the degree of social presenceor richness of those media and that com-pensate for problems associated with me-dia “leanness” (Fulk et al., 1990; Lee, 1994;Markus, 1994; Ngwenyama & Lee, 1997).Other empirical studies led to mixed find-ings. For example, Baker’s (2002) study,which compared the performance of 64virtual teams utilizing four different typesof communication media (text-only, audio-only, text-video, and audio-video), found nosignificant difference in the quality of deci-sions reached by teams interacting throughtext-only and audio-only media. However,the same study found that the addition ofvideo to audio-only communication resultedin a significant improvement in the qualityof teams’ strategic decisions.

Inconsistencies such as the ones men-tioned above led, over the years, to severalattempts to develop more robust theoreti-cal frameworks combining theoretical ele-ments of the social presence and mediarichness theories with theoretical elementsof theories that take into account other fac-tors, such as social influences (Carlson &Zmud, 1999; Trevino, Webster, & Stein,2000; Webster & Trevino, 1995). Otherattempts tried to review the foundations ofthe social presence and media richness theo-ries, and create more robust theories thatcould provide an alternative to the socialpresence and media richness theories. Onesuch attempt led to the development ofcompensatory adaptation theory.

Compensatory Adaptation Theory

Compensatory adaptation theory(Kock, 1998, 1999, 2001a, 2001b) has beendeveloped in part to explain the inconsis-tent findings discussed in the previous sec-tion, and in part to provide an alternative



theory that could overcome the limitationsof the social presence and media richnesstheories. Compensatory adaptation theoryargues that electronic communication me-dia in general offer certain advantages, suchas that of allowing for asynchronous anddistributed group interaction, and at thesame time, pose obstacles for communica-tion in groups. In this respect, compensa-tory adaptation theory is similar to the so-cial presence and media richness theories.The key difference is that compensatoryadaptation theory attempts to provide ascientific basis for the existence of thoseobstacles, by arguing that they are prima-rily due to the fact that our biological com-munication apparatus, which includes spe-cialized organs and brain functions, has beenoptimized by Darwinian evolution for face-to-face communication, incorporating fivemain elements — co-location, synchronicity,the ability to convey body language, facialexpressions, and speech.

Evidence about the evolution of ourbiological communication apparatus sug-gests that during over 99% of our evolu-tionary cycle, our ancestors relied on co-located and synchronous forms of commu-nication through facial expressions, bodylanguage, and sounds (including speech,which uses a large variety of sound combi-nations) to exchange information andknowledge among themselves (Boaz &Almquist, 1997; Cartwright, 2000). Accord-ing to evolutionary principles, a plausibleconclusion from this is that our biologicalcommunication apparatus was designedprimarily to excel in face-to-face commu-nication (Kock, Hassell, & Wazlawick,2002). That is, since we have communi-cated only face-to-face during the vast ma-jority of our evolutionary cycle, then ourbiological communication apparatus (whichincludes the brain modules involved in the

sensing and control of our communicationorgans) must have been designed for face-to-face communication.

The above conclusion is supported bythe presence of obvious face-to-face com-munication adaptations in our biologicalcommunication apparatus. For instance,evolution endowed human beings with acomplex Web of facial muscles (22 on eachside of the face; more than any other ani-mal) that allow them to generate over 6,000communicative expressions; very few ofthese muscles are used for other purposes,such as chewing (Bates & Cleese, 2001;McNeill, 1998). The existence of such acomplex Web of muscles would appear tohave been a major waste, had we not beendesigned to use them extensively whilecommunicating with others.

The evolutionary path that led to ourspecies also suggests a noticeable evolu-tionary direction over millions of years to-ward the development of a biological com-munication apparatus that supported evenmore sophisticated forms of speech, cul-minating with the development of complexspeech approximately 100,000 years ago.The advent of complex speech was enabledby the development of a larynx located rela-tively low in the neck and an enlarged vo-cal tract — key morphological traits thatdifferentiate modern humans from theirearly ancestors and that allow modern hu-mans to generate the large variety ofsounds required to speak most modern lan-guages (Laitman, 1984, 1993; Lieberman,1998). The morphology of the human earalso suggests a specialized design to de-code speech (Lieberman, 1998; Pinker,1994). The adaptive importance of speechfor human beings is highlighted by the factthat our enlarged vocal tract also makes usthe most likely among all primates to chokeon food and ingested liquids.



The Media Naturalness Principle

The evolution of closely matched brainand body functions, which follows from thewidely held brain-body co-evolution law ofmodern evolution theory (Lieberman, 1998;Wills, 1989, 1993), provides a scientific ba-sis for the apparent bias toward face-to-face communication hypothesized by thesocial presence and media richness theo-ries, and is reflected in compensatory ad-aptation theory’s media naturalness prin-ciple. The principle states that individualswho choose to use electronic communica-tion tools experience increased cognitiveeffort and communication ambiguity pro-portionally to the degree to which the toolssuppress elements that are present in face-to-face communication (e.g., synchronicity,ability to convey/perceive non-verbal com-munication cues). The principle is task in-dependent, that is, it applies to all collabo-rative tasks, even though it acknowledgesthat the link is less noticeable in tasks thatdo not involve intense communicationwhich are seen as tasks that involve littleknowledge sharing among collaborators(Kock, 2001b).

The media naturalness principle linksthe use of electronic communication me-dia with high cognitive effort and commu-nication ambiguity, but not necessarily withspecific task-related outcomes. In doing so,it explains empirical findings that supportedin part the social presence and media rich-ness theories (Daft et al., 1987; Fulk et al.,1990; Rice, 1993; Rice & Shook, 1990),while at the same time avoiding the prob-lems associated with making predictionsabout media choice or task-related out-comes based on communication mediatraits, which led to criticism by social re-searchers (El-Shinnawy & Markus, 1998;Kinney & Watson, 1992; Kock, 1998; Lee,1994; Markus, 1994). For example, the

media naturalness principle is compatiblewith the notion that social influences canlead users to modify their behavior (Markus,1994; Lee, 1994; Ngwenyama & Lee, 1997)in ways that are independent of the appar-ent degree of naturalness of a medium, eventhough the cognitive effort required for thisto happen will be higher than if a more“natural” communication medium (i.e., theface-to-face medium) were used.

The above discussion is compatiblewith the notion that electronic communica-tion tools are artifacts developed to solvespecific problems, but they also add newproblems of their own (Ellis, Gibbs, & Rein,1991; Nunamaker, Dennis, Valacich, Vogel,& George, 1991; Nunamaker, Briggs,Mittleman, Vogel, & Balthazard, 1997;Walther, 1996). There seems to be a gen-eral pattern of gains due to the use of elec-tronic communication tools to relate togroup efficiency, such as higher numberof ideas generated per unit of time (Den-nis, Haley, & Vanderberg, 1996;Nagasundaram & Bostrom, 1995) andlower costs associated with running groups(Kock, 2000), even though sometimeslosses occur in connection with obstaclesto group effectiveness, such as less com-mitment toward group decisions due tolower group cohesiveness and satisfaction(Ellis et al., 1991; Nunamaker et al., 1991;Walther, 1996).

While it is intuitive to think that ob-stacles to high group effectiveness invari-ably lead to lower quality of group out-comes, there is a wealth of evidence fromfields as diverse as biological anthropology(Dobzhansky, 1971) and analytical psychol-ogy (Jung, 1968) suggesting that humanbeings voluntarily and involuntarily compen-sate for obstacles posed to them, in somecases overcompensating for those ob-stacles and achieving even better outcomesthan if the obstacles were not present



(Kock, 1998). This compensatory adapta-tion phenomenon has the potential to con-tradict deterministic predictions linkingnegative communication media influenceson group effectiveness with low group out-come quality. Kock (1998, 1999) obtainedempirical evidence of this compensatoryadaptation phenomenon in the context ofelectronic communication in a study thatcompared groups performing complex andknowledge-intensive tasks over e-mail andface-to-face. The e-mail medium was con-sistently seen by group members as lessappropriate than the face-to-face mediumto accomplish the tasks. Yet, the tasks ac-complished through e-mail yielded outcomesthat were perceived as being of slightlybetter quality than those produced by theface-to-face groups.

Compensatory Adaptation Principle

Compensatory adaptation theory ar-gues that users of electronic communica-tion tools present two common patterns ofreaction toward those tools. Firstly, usersof electronic communication tools in gen-eral perceive those tools as creating com-munication media that pose cognitive ob-stacles to communication when comparedwith the face-to-face medium (Kock,2001a, 2001b), as proposed by the medianaturalness principle discussed above. Thatis, even though electronic communicationtools may reduce or eliminate physical ob-stacles to face-to-face communication —for example, e-mail and instant messagingallow people to communicate instantly overlong distances — they also increase thecognitive effort required to communicateinformation and knowledge. The secondcommon pattern of reaction is one of com-pensation for the obstacles posed by themedia (Kock, 1998, 1999), which is em-bodied in the theory’s compensatory ad-

aptation principle. The principle statesthat individuals who choose to use elec-tronic communication media tend to com-pensate for the cognitive obstacles theyperceive as associated with the lack ofnaturalness of those media, which leadsthose individuals to generate, when en-gaged in collaborative tasks, outcomes ofthe same or better quality than if they hadinteracted solely face-to-face.

So, in summary, compensatory adap-tation theory argues, in an apparently para-doxical way, that obstacles posed by elec-tronic communication media will have nonegative effect on the quality of group out-comes as individuals engaged in collabora-tive tasks attempt to (often involuntarily)and often succeed in compensating forthem. Since electronic communication me-dia offer some advantages over the face-to-face medium, such as the possibility ofasynchronous and non-collocated interac-tion, the compensatory adaptation theoryargument, based on the two complemen-tary theoretical principles of media natu-ralness and compensatory adaptationjust discussed, seems to support the para-doxical notion that “less can be more,” soto speak.

Hypotheses

As it can be inferred from the reviewof compensatory adaptation theory pre-sented in the previous section, the theorytakes a somewhat general view of collabo-rative tasks, and does not refer to any col-laborative task in particular. This placescompensatory adaptation theory in a goodposition to be tested, and potentially falsi-fied, according to Stinchcombe’s (1968)principle of empirical testing variety. Thatepistemological principle maintains that atheory’s validity can be best tested when avariety of predictions regarding related



empirical results can be derived from thetheory. In other words, the more indepen-dent corroborations are made of thetheory’s basic propositions under differentcircumstances, the more likely it is that thosepropositions are generally correct.

The application of Stinchcombe’s(1968) principle of empirical testing vari-ety is illustrated in this study by the choiceof two related collaborative tasks as thebases for an empirical test of compensa-tory adaptation theory, which should be seenas an initial step in testing the theory, andone of many (future) empirical tests of thetheory. The related tasks are those of pro-cess modeling and redesign, whereby indi-viduals respectively: a) create a model of aparticular organizational process; and b)based on the model created, conceptuallyredesign the process so that improvementsin quality and productivity can be achieved.For the sake of simplicity in formulatingtestable hypotheses, the two related tasksare referred in this section as one main“process redesign” task. This is consistentwith current organizational practices, whereprocess modeling is usually seen as part ofthe larger task of process redesign.

Compensatory adaptation theory’smedia naturalness principle states that elec-tronic communication media in general,which are assumed to suppress some ofthe elements of face-to-face communica-tion, cause higher cognitive effort in con-nection with communication interactionsthan the face-to-face medium. The reasonfor this increase in cognitive effort is, ac-cording to compensatory adaptation theory,that evolutionary forces shaped our biologi-cal communication apparatus so that weexcel in face-to-face communication. Be-cause of the suppression of elements thatare found in typical face-to-face commu-nication interactions, in electronic commu-

nication our brain cannot make use of cer-tain hardwired communication circuits (e.g.,those aimed at speech generation and rec-ognition), and thus must rely on other, lessefficient learned brain circuits (e.g., thosedeveloped through use-induced learning ofelectronic communication tools). This ar-gument, which is explored in more detailby Kock (2002) and is only summarizedhere, leads us to hypothesis H1.

H1: The use of an electronic communicationmedium by process redesign dyads willincrease the level of cognitive effortexperienced by the members of the dyads.

When individuals are brought up indifferent cultural environments, they invari-ably develop different information process-ing schemas over their lifetimes. Differentschemas make individuals interpret infor-mation in different ways, especially wheninformation they expect to receive is notactually provided. This phenomenon wasoriginally demonstrated by Bartlett (1932)through a series of experiments he con-ducted involving the American Indian folktale “The War of The Ghosts,” which iswell known among experimental psycholo-gists for its strange gaps and bizarre causalsequences. The experiments yielded evi-dence that subjects who held different in-formation processing schemas interpretedthe tale in substantially different ways andaccording to their specific cultural schemas.Individuals were expecting certain piecesof information to be provided to them inBartlett’s (1932) experiments. When theywere not provided with the information theyexpected to receive, those individuals “filledin the gaps” based on their existing infor-mation processing schemas and the limitedinformation that they were given (see alsoGardner, 1985). This conclusion provideda solid foundation on which to explain key



differences in the way different individualsinterpreted the tale.

Several generic problems have oc-curred recurrently during the millions ofyears that led to the evolution of the hu-man species (e.g., how to identify a fruitthat is rich in certain minerals), and thehuman brain possesses a series ofhardwired information processing schemasthat are designed to solve those problems(Cosmides & Tooby, 1992; Tooby &Cosmides, 1992). Several of those prob-lems have been addressed by evolutionaryadaptations that are associated with thecommunication process (Pinker & Bloom,1992). Those adaptations have led to thedevelopment of hardwired schemas tied tothe communication process that make ussearch for enough information to effec-tively interpret the message being commu-nicated, and that information comes to usthrough several of the stimuli that arepresent in actual face-to-face communica-tion (Lieberman, 2000), such as contextualcues (available in co-located communica-tion) and immediate feedback (available insynchronous communication) in the formof facial expressions, body language, andvoice intonations. When many of thosestimuli are absent, which may be causedby their selective suppression through e-communication technologies, one wouldexpect that e-communication technologyusers will “fill in the gaps” in a manner simi-lar to what the subjects in Bartlett’s (1932)experiments did.

However, in the absence of informa-tion-giving stimuli, “filling in the gaps” islikely to lead to a higher proportion of mis-interpretations, and thus ambiguity, than ifthe stimuli were not suppressed — asBartlett’s (1932) and other studies show(see, e.g., Gardner, 1985; Pinker, 1997).Given the general similarity of the biologi-

cal apparatus shared by different individu-als, it is likely that they will look for thesame types of communicative stimuli. Yet,given their different information process-ing schemas, their interpretation of the mes-sage being communicated in the absenceof those stimuli is likely to differ from theinterpretations reached at by other individu-als. This leads to the conclusion that a de-crease in medium naturalness, caused bythe selective suppression of media natural-ness elements in a communication medium,is likely to lead to an increase in the prob-ability of misinterpretations of communica-tive cues and thus an increase in commu-nication ambiguity. This leads us to hy-pothesis H2.

H2: The use of an electronic communicationmedium by process redesign dyads willincrease the level of communication ambiguityexperienced by the members of the dyads.

The compensatory adaptation prin-ciple of compensatory adaptation theoryargues that individuals that choose to useelectronic communication media for com-munication try to compensate for the cog-nitive obstacles they perceive as associ-ated with the lack of naturalness of thosemedia. This behavior is posited by the theoryto be a natural and often involuntary reac-tion to the perception of cognitive obstaclesposed by electronic communication toolsin general, as stated in the media natural-ness principle. Previous exploratory re-search (Kock, 1998, 1999, 2001c) suggeststhat this compensatory adaptation behav-ior is indicated by at least two behavioralpatterns — more careful preparation ofcommunication messages and decreasedcommunication fluency (which is definedas the number of words per unit of timeconveyed through the medium). This takesus to hypotheses H3 and H4.



H3: The use of an electronic communicationmedium by process redesign dyads will leadto increased preparation of communicationmessages by the members of the dyads.

H4: The use of an electronic communicationmedium by process redesign dyads willdecrease the fluency displayed by the membersof the dyads.

Finally, compensatory adaptationtheory argues that compensatory behaviorusually leads individuals to compensate forthe obstacles posed by electronic commu-nication media of low naturalness, whichoften leads to outcomes of the same orbetter quality than those achieved throughmore natural media. This provides the ba-sis for hypothesis H5.

H5: The use of an electronic communicationmedium by process redesign dyads will haveno negative effect on the quality of theoutcomes produced by the members of thedyads.

In summary, we could say that theset of hypotheses above comprises the es-sence of compensatory adaptation theoryand thus provides the basis for a valid testof the theory. Underlying the theory is thenotion that communication media that sup-press key face-to-face communication el-ements pose obstacles to communication,leading to increased cognitive effort (H1)and communication ambiguity (H2). Nev-ertheless, the theory also argues that theseobstacles will trigger compensatory adap-tation mechanisms, indicated by increasedpreparation of communication messages(H3) and decreased fluency (H4). This will,according to the theory, lead to compensa-tion and, in what appears to be acounterintuitive twist, no negative effect onthe quality of outcomes in connection witha process redesign task (which arguably

requires intense communication) performedelectronically (H5).

Research Method

The hypotheses were tested througha field experiment employing a repeatedmeasures design where the communica-tion medium used varied according to twoexperimental conditions: face-to-face andelectronic. The impact of changes in thecommunication medium factor on a set ofdependent variables was assessed bymeans of multiple ANOVA tests (Green,Salkind, & Akey, 1997; Rosenthal &Rosnow, 1991).

Participants andTreatment Conditions

The research study involved subjectswith substantial hands-on experience inprocess redesign in the defense sector, re-cruited from management and engineeringranks of a large defense contractor. All ofthe subjects were college educated, andmost held undergraduate degrees in busi-ness, computer science, or engineering. Thesubjects were familiar with each other andwith the electronic communication mediumused prior to their participation in this fieldexperiment. However, they had no priorexperience using the electronic communi-cation medium for the collaborative comple-tion of tasks of the same type as, or evensimilar complexity to, the experimental task.Their ages ranged from 23 to 60, with amean age of 35. Fifty-nine percent of thesubjects were males.

The subjects were randomly assignedto dyads and to communication media con-ditions. Each dyad completed two similarprocess redesign-related tasks using differ-ent communication media for each task.Half of the dyads (i.e., 10 dyads) completed



one of the tasks face-to-face, while theother half completed the same task elec-tronically. After this, all dyads moved on tothe next task, using different media thanthey had used in the previous task — thatis, the dyads previously interacting face-to-face now interacted electronically andvice versa. This led to the collection of re-search data in connection with 40 cases ofdyads performing the same type of collabo-rative task.

Electronic Communication Medium

A set of Web-based threaded onlinediscussion boards created the electroniccommunication media employed in the ex-periment. The online discussion boards weredeveloped using Microsoft FrontPage 2000and Active Server Pages. They were usedin a quasi-synchronous manner — that is,they were “refreshed” at short time inter-vals, creating an online chat-like environ-ment where the discussions were threaded.One Web-based online discussion boardwas created for each dyad. All online dis-cussion boards were identical.

Tasks

Two experimental tasks were used,which can be seen as being of the samegeneral type and also as subtasks of a largerprocess redesign task. Both tasks weredeveloped based on a “real” process rede-sign project previously reported by Kockand Murphy (2001) that targeted the pro-cess whereby the U.S. Department ofDefense procured and purchased complexsoftware development services. Since thatreal project was a successful process re-design project and experts evaluated itsoutcomes and found them to be optimal, itwas assumed that those outcomes shouldserve as a basis for comparison with the

outcomes produced by the participants inour field experiment. In other words, weemployed two hidden-profile tasks in ourfield experiment (see, e.g., Dennis, Kinney,& Hung, 1999).

In the first task, which involved pro-cess modeling, participants were given dif-ferent pieces of information about a de-fense acquisition process and were askedto develop a full graphical model of the pro-cess using a set of pre-defined symbols (seeAppendix A for a more detailed descrip-tion of the process modeling task). Bothpieces of information initially received byeach of the members of the dyad werenecessary for the generation of the fullgraphical model of the process (see Ap-pendix C for the expected outcome).

In the second task, which involvedprocess redesign, participants were givendifferent sets of guidelines about how toredesign the process modeled in the previ-ous task and were asked to develop agraphical model of the process after theapplication of all the redesign guidelines,using the same set of symbols employed inthe previous task (see Appendix B for amore detailed description of the processredesign task). Both sets of guidelines ini-tially received by each of the members ofthe dyad were necessary for the genera-tion of the graphical model of the rede-signed process (see Appendix D for theexpected outcome).

Procedure

After the participants were randomlyassigned to dyads and each dyad was ran-domly assigned to one of the two commu-nication media conditions, the participantsreceived a general orientation about thetasks and went through a 15-minute “re-fresh” session on process modeling andredesign.



The dyads were then given 40 min-utes to complete the process modeling task,after which each member of the dyadshanded in their initial process model(sketch) to an experiment facilitator andcompleted a questionnaire. After a shortbreak, the dyads (now interacting throughdifferent communication media) were given40 minutes to complete the second processredesign task, after which each memberof the dyads handed in their redesignedprocess model (sketch) to an experimentfacilitator and completed a questionnaire(the same questionnaire completed at theend of the process modeling task). No dyadwas able to complete its task in less thanthe time allocated (40 minutes); a few dy-ads appeared to have completed their workin slightly less than 40 minutes, but never-theless decided to use the remaining timeto perfect their sketches.

Measures

The dependent variables were cog-nitive effort (H1), communication ambigu-ity (H2), message preparation (H3), fluency(H4), and task outcome quality (H5). Thefollowing variables were measured at theindividual level of analysis (thus based on40 data points) through one-item, percep-tion-related questions (see Appendix F):cognitive effort, communication ambiguity,message preparation, and task outcomequality. The remaining variable, fluency,was measured at the dyadic level of analy-sis (based on 20 data points).

The question-statement and scale forthe cognitive effort variable was based onNASA’s task load index (a.k.a., NASA-TLX) developed by Hart and Staveland(1988). The question-statements for com-munication ambiguity and message prepa-ration were based on a focus group dis-cussion with the participants, conducted

prior to the experiment, in which the mean-ing of the construct’s communication am-biguity and message preparation and theproper wording of related questions (to beanswered on a seven-point Likert-typescale) were agreed upon. Rosenthal andRosnow (1991) suggest the test-retestmethod as an alternative method for reli-ability assessment, which is a convenientalternative when single-item measures areused since component reliability cannot becomputed. Following that suggestion, theinstrument comprising the single question-statements for each variable was assessedthrough the test-retest method with twosimilar “dummy” process redesign projectsconducted two weeks apart. That assess-ment yielded acceptable results (alpha =.88).

As proposed by Kock (1998), fluencywas measured by counting the number ofwords exchanged by the members of thedyads and dividing it by the number of min-utes each dyad took to complete the task(40 minutes for all dyads), yielding a mea-sure of the number of words per minuteexchanged by the dyads. These measureswere obtained based on four videotapedface-to-face dyad sessions (two for theprocess modeling and two for the processredesign task) and 10 electronic dyad ses-sions (five for the process modeling andfive for the process redesign task).

Task outcome quality was measuredby comparing the process sketches gener-ated by the dyad members with the “cor-rect” models (see Appendix C and Appen-dix D)—that is, the models generated basedon the successful process redesign projectteam studied by Kock and Murphy (2001).Two different coders generated these “simi-larity scores” used to assess task outcomequality independently. The scores weregenerated based on criteria addressing syn-tactic as well as semantic correctness of



the sketches. Inter-coder reliability was high(alpha = .91).

Analysis Results

Table 1a summarizes one-wayANOVA statistics as well as means andstandard deviations in connection with allvariables but fluency. Since fluency datadid not conform to assumptions underlyingANOVA (e.g., the number of data pointswas different for each treatment condition),a Mann-Witney U test (a nonparametrictechnique — see, e.g., Siegel & Castellan,1998) was employed; its results are sum-marized in Table 1b.

The analyses of variance yielded sta-tistically significant results in connectionwith cognitive effort: F(1, 38) = 9.94, p <.01; communication ambiguity: F(1, 38) =12.10, p < .01; and message preparation:F(1, 38) = 6.34, p < .05. Participants gen-erally perceived cognitive effort, commu-nication ambiguity, and message prepara-tion to be higher in the electronic commu-nication than in the face-to-face condition,

which provides general support for H1, H2,and H3.

The Mann-Whitney U test yieldedstatistically significant results in connectionwith fluency: Z = -2.83, p < .01. Fluencywas significantly lower in the electroniccommunication than in the face-to-facecondition, falling below what the “typing-versus-speaking effect” would allow us toexpect (i.e., typing is inherently slower thanspeaking; see McQueen, Payner, & Kock,1999, for a review of studies that addressedthis effect). Given that the participants usedcomputers for preparation of written docu-ments on a daily basis, the “typing-versus-speaking effect” would allow us to expectthe fluency over the electronic communi-cation medium to be, on average, no lessthan half the fluency face-to-face (or about36 words per minute; see McQueen et al.,1999). Thus, the actual fluency in the elec-tronic communication medium (16.58words per minute) provides general sup-port for H4.

The analyses of variance yielded sta-tistically insignificant results in connection

Table 1a. Descriptive statistics and one-way ANOVA results for all variables but fluency

Table 1b. Descriptive statistics and Mann-Whitney U test results in connection with fluency

Variable Mean face-to-face

St. dev. face-to-face

Mean electronic

St. dev. electronic

F p

Cognitive effort

55.01 20.23 77.76 25.12 9.94 < .01

Communication ambiguity

2.47 1.55 4.44 2.00 12.11 < .01

Message preparation

2.99 1.78 4.41 1.79 6.34 < .05

Task outcome quality

4.06 1.69 3.91 1.61 .09 .77

Variable Mean face-to-face

St. dev. face-to-face

Mean electronic

St. dev. electronic

Z p

Fluency

71.01 15.75 16.58 5.93 -2.83 < .01



with task outcome quality: F(1, 38) = .08, p= .77. That is, there was no significant dif-ference in the quality of the outcomes gen-erated by the members of the dyads acrossdifferent communication media conditions,which provides general support for H5.

DISCUSSION

This study suggests that the use of aWeb-based threaded online discussionboard by dyads redesigning a defense ac-quisition process, when compared with theface-to-face medium, increased perceivedcognitive effort by about 41%, perceivedcommunication ambiguity by about 80%,and perceived message preparation byabout 47%, while at the same time reduc-ing actual fluency by approximately 77%.The study also suggests that the use of theWeb-based threaded online discussion boardhad no significant impact on the quality ofthe outcomes generated by the dyads.

Since the hypotheses tested throughthis study were derived from compensa-tory adaptation theory, and were generallysupported by the evidence, we can con-clude that the study overall supports com-pensatory adaptation theory. In particular,the marked increases in perceived cogni-tive effort and communication ambiguityprovide support for the theory’s media natu-ralness principle. At the same time, the sig-nificant increase in perceived messagepreparation and the drastic reduction in“fluency,” coupled with the neutral impacton the quality of the outcomes generatedby the dyads, provide strong support forthe compensatory adaptation principle.

As previously discussed, many doubtshave been raised by researchers (e.g., Fulket al., 1990; Lee, 1994; Markus, 1994;Ngwenyama & Lee, 1997) in connectionwith media richness theory. Given those

doubts, it seems plausible to explore thepossibility that compensatory adaptationtheory can replace media richness theory,perhaps with some advantages. This wouldprobably require an expansion of compen-satory adaptation theory so that specificpredictions could be made in connectionwith particular types of tasks (Zigurs &Buckland, 1998), and could prove to be-come a fertile line of theoretical researchgiven the influence that media richnesstheory has continually exerted on electroniccommunication research (Allen & Griffeth,1997; Carlson & Zmud, 1999; Dennis &Kinney, 1998; Dennis & Valacich, 1999;Jackson & Purcell, 1997).

The potential for compensatory ad-aptation theory to replace media richnesstheory is highlighted by the fact that it ex-plains findings of this study, which argu-ably supports one of the notions seen byresearchers (Carlson & Zmud, 1999; Jack-son & Purcell, 1997; Rice, 1992) as vali-dating in part media richness theory (i.e.,that the face-to-face medium is perceivedby users as generally “richer” than elec-tronic communication media); at the sametime, compensatory adaptation theory canbe used as a basis on which one can ex-plain findings that generally contradict oneof the key predictions of media richnesstheory — that groups interacting through arelatively “lean” medium invariably produceoutcomes of lower quality than groups in-teracting through “richer” media.

Some researchers may argue that thefindings of this study may be interpreteddifferently, and be seen as supporting in partmedia richness theory and pointing at waysin which that theory could be revised toincorporate such notions as that of socialinfluence (Fulk et al., 1990) and commonground (Clark & Brennan, 1991). Fulk etal.’s (1990) social influence model providesthe basis from which to predict that social



influences, such as peer pressure, can leadto the use of a “lean” medium, even thoughmedia that are perceived as richer (e.g.,face-to-face, telephone) may be available.That could lead individuals to develop com-mon ground (Clark & Brennan, 1991) ele-ments (e.g., mutual knowledge, shared be-liefs, goals and attitudes, etc.) in connec-tion with the use of a “lean” medium tosupport a particular collaboration task, or atask of collaborative type. This could, inturn, lead to a “channel expansion” phe-nomenon (Carlson & Zmud, 1999), whereboth the perceived and actual richness ofthe medium would be increased by the de-velopment of common ground elements.Such plausible alternative interpretation isanalogous to that proposed by Carlson andZmud (1999) in connection with their chan-nel expansion theory. Therefore, future re-search should consider the possibility thatcompensatory adaptation theory is not thebest theoretical framework to explain thisstudy’s findings, even though there aresome indications that this may be the case.Such skepticism and theoretical neutralismis likely to lead to greater progress than thenarrower view of compensatory adaptationas a sort of “grand theory” (which it is un-likely to be).

Much of the past research on behav-ior toward electronic communication toolshas focused on individual choice of com-munication media and, to some extent, theoutcomes produced by individuals engagedin collaborative tasks. While that researchhas led to mixed findings, it neverthelesssuggests that behavior toward electroniccommunication tools is both complex and,notwithstanding much research done in thearea, somewhat unpredictable (DeSanctis,Poole, Dickson, & Jackson, 1993; Postmes,Spears, & Lea, 1998; Sallnas et al., 2000).Research findings that appear to be some-what contradictory have supported predic-

tions based on theories that emphasize char-acteristics of the communication medium(Short et al., 1976; Daft & Lengel, 1986),as well as theories that emphasize socialinfluences (Fulk et al., 1990; Lee, 1994;Markus, 1994; Ngwenyama & Lee, 1997),which have often been seen as competingtypes of theories (Trevino et al., 2000;Webster & Trevino, 1995). Underlying thisdebate between advocates of communica-tion media and social influence theories isa set of puzzling findings, which can besummarized into two main groups of find-ings: a) that the face-to-face medium isconsistently perceived by individuals as avery appropriate communication mediumfor a variety of collaborative tasks (Daft etal., 1987; Rice, 1992; Markus, 1994; Rice& Shook, 1990; Walther, 1996); and b) thatthis perception has often been contradictedby the choice of communication media dif-ferent from face-to-face by individuals con-ducting collaborative tasks (Lee, 1994;Markus, 1994; Ngwenyama & Lee, 1997)and by those individuals sometimes produc-ing the same or better quality task outcomesthan individuals interacting primarily face-to-face (Kock, 1998).

The study described in this articlemakes an important theoretical contribu-tion that can be used as a basis for recon-ciling the competing findings above. It doesso by providing evidence that generally sup-ports a new theory, compensatory adapta-tion theory, which builds on the contempo-rary version of Darwin’s (1859) evolutiontheory. In spite of the caveat presentedabove regarding the possibility of alterna-tive theoretical explanations that are dif-ferent from those of compensatory adap-tation theory, it is not unreasonable to ar-gue that this article shows beyond muchdoubt that compensatory adaptation theoryhas the potential to explain the puzzling andcontradictory findings discussed above, and



provide a new basis for future research onelectronic communication.

CONCLUSION

Much of the past research on elec-tronic communication media suggests thatthose media pose obstacles to communi-cation when compared with the face-to-face medium. Yet, past research also pointsat mixed findings in connection with thequality of the outcomes of tasks, suggest-ing that the use of electronic communica-tion media has no negative effect on them.A new theoretical framework, called com-pensatory adaptation theory, has been pro-posed to explain these contradictory find-ings. This study provides a test of compen-satory adaptation theory by investigatingthe impact of the use of an electronic com-munication medium on 20 business processredesign dyads involving managers and pro-fessionals at a large defense contractor,with a focus on cognitive effort, communi-cation ambiguity, message preparation, flu-ency, and task outcome quality.

This study suggests that even thoughelectronic communication media use in pro-cess redesign dyads involving managers andprofessionals seemed to increase cognitiveeffort and communication ambiguity, thatuse had a neutral impact on task outcomequality. These results appear to be an out-come of compensatory adaptation,whereby the members of the dyads inter-acting through the electronic communica-tion medium modified their behavior in or-der to compensate for the obstacles posedby the medium, which finds confirmatorysupport from a marked decrease in fluencyand an increase in message preparation.The results generally support predictionsbased on compensatory adaptation theory.

The findings above provide a newbasis on which users of electronic commu-

nication tools can understand why tools thatseem to make communication more diffi-cult can still lead to no impact on the effec-tive use of those tools for communicationabout complex issues. It is important forelectronic communication tool users to un-derstand the phenomenon of compensatoryadaptation, particularly because its para-doxical nature may lead those users to be-lieve that outcomes of collaboration tasksare not negatively affected by the use ofelectronic communication tools because ofthe tools’ effectiveness in supportingcommunication interactions. This will probablylead to frustration when those users realizethat “good [electronic] communication requireshard work” (see Bartlett, 2001, p. 1).

Nevertheless, one implication of thisstudy for practice is that individuals col-laborating electronically to accomplish com-plex and knowledge-intensive tasks suchas process redesign can expect to be suc-cessful, even when the tasks are conductedentirely electronically. In light of the emer-gence of the Internet as a key enabler ofcommunication in organizational processes,and the consequent multiplication of orga-nizational forms characterized by their lowdependence on physical structures for em-ployee interaction, such as the so-called“virtual organizations” (Barnatt, 1995;Davidow & Malone, 1992), this is not onlygood news for organizations, but also pro-vides the basis on which to call for increas-ing use of electronic communication mediato support a variety of types of group tasks,ranging from routine group tasks, wherethe use of electronic communication me-dia is already relatively common, to moread-hoc (or project-based) ones, where theuse of electronic communication media isstill rare.

However, the extra “cost” imposedon individuals who wish to communicateelectronically, rather than face-to-face,



about complex issues is an important issuethat must be addressed in further research.As posited by the media naturalness prin-ciple and indicated by this study, compen-satory adaptation is associated with “work-ing hard” to compensate for increases incognitive effort and communication ambi-guity. That is, compensatory adaptation has“a price.” One possible negative conse-quence of this finding, not addressed by thisstudy, could be avoidance by group mem-bers to participate in future electronicgroups after their initial experience, as theywould become increasingly aware of theextra cognitive effort required of them. Thiscould have an overall negative impact onthe use of electronic media by organiza-tions to support organization-wide initiatives,such as process redesign programs aimedat enabling organizations to obtain qualityor productivity certifications. Perhaps theeducation about users of electronic com-munication technologies about compensa-tory adaptation would allow them to un-derstand the extra “cognitive price” thatthey have to pay, and thus mitigate theirnegative perceptions about the use of thosetechnologies to accomplish complex andknowledge-intensive collaborative tasks.

Another alternative to address the is-sue above is to break down complex taskssuch as process redesign into subtasks, anduse different communication media to sup-port those subtasks. Prior research hasshown that the amount of knowledge trans-fer involved in a communication interac-tion correlates the perceived difficulty ininteracting through non-face-to-face me-dia (Kock, 1998). Therefore, subtasks couldbe classified according to the amount ofknowledge transfer involved, and assigneddifferent communication media, where thedegree of similarity of each medium to theface-to-face medium should be matchedwith the amount of knowledge transfer

needed for effective completion of eachsubtask. That is, for high knowledge trans-fer subtasks, media that is very face-to-face-like (e.g., videoconferencing) shouldbe used, whereas for low knowledge trans-fer subtasks, media that incorporate fewof the elements found in face-to-face com-munication (e.g., e-mail) could be used.Following Kock and Davison’s (2003) con-ceptual formulation, high knowledge trans-fer subtasks are defined here as those thatinvolve process-related explanations andrelated associative assertions (e.g., expla-nations of why a certain car assembly pro-cess is designed the way it is), rather thanonly descriptions of process attributes (e.g.,the current production capacity of a carassembly process). This matching ofsubtasks and media is likely to become in-creasingly common in industry, as new elec-tronic communication tools are developedto support specific types of tasks. In fact,from a long-term perspective, this may be-come one of the most common approachesto effectively employ electronic communi-cation tools in organizations in general.

REFERENCES

Allen, D.G. & Griffeth, R.W. (1997). Ver-tical and lateral information processing:The effects of gender, employee classi-fication level, and media richness oncommunication and work outcomes.Human Relations, 50(10), 1239-1260.

Avison, D., Lau, F., Myers, M.D., &Nielson, P. (1999). Action research.Communications of the ACM, 42(1),94-97.

Baker, G. (2002). The effects of synchro-nous collaborative technologies on deci-sion making: A study of virtual teams.Information Resources ManagementJournal, 15(4), 79-94.



Barnatt, C. (1995). Office space,cyberspace and virtual organization.Journal of General Management,20(4), 78-92.

Bartlett, F. (1932). Remembering: A studyin experimental and social psychol-ogy. Cambridge, MA: Cambridge Uni-versity Press.

Bartlett, M. (2001). Good e-mail commu-nication requires hard work. Washing-ton Post. Retrieved fromw w w. n e w s b y t e s . c o m / n e w s / 0 1 /168549.html.

Basdogan, C., Ho, C.H., Srinivasan, M.A.,& Slater, M. (2000). An experimentalstudy on the role of touch in shared vir-tual environments. ACM Transactionson Computer-Human Interaction,7(4), 443-460.

Baskerville, R. (1997). Distinguishing ac-tion research from participative casestudies. Journal of Systems and Infor-mation Technology, 1(1), 25-45.

Baskerville, R. (1999). Investigating infor-mation systems with action research.Communications of the AIS, 2(19), 32.Retrieved from cais.isworld.org.

Baskerville, R. & Wood-Harper, T. (1996).A critical perspective on action researchas a method for information systemsresearch. Journal of InformationTechnology, 11(3), 235-246.

Baskerville, R. & Wood-Harper, T. (1998).Diversity in information systems actionresearch methods. European Journalof Information Systems, 7(2), 90-107.

Bates, B. & Cleese, J. (2001). The humanface. New York: DK Publishing.

Boaz, N.T. & Almquist, A.J. (1997). Bio-logical anthropology: A synthetic ap-proach to human evolution. UpperSaddle River, NJ: Prentice-Hall.

Burke, G. & Peppard, J. (Eds.). (1995).Examining business process re-engi-neering. London: Kogan Page.

Carlson, J.R. & Zmud, R.W. (1999). Chan-nel expansion theory and the experien-tial nature of media richness perceptions.Academy of Management Journal,42(2), 153-170.

Cartwright, J. (2000). Evolution and hu-man behavior: Darwinian perspec-tives on human nature. Cambridge,MA: MIT Press.

Champy, J. (1995). Reengineering man-agement. New York: Harper Business.

Choi, T.Y. (1995). Conceptualizing continu-ous improvement: Implications for orga-nizational change. Omega, 23(6), 607-624.

Choi, T.Y. & Behling, O.C. (1997). Topmanagers and TQM success: One morelook after all these years. The Acad-emy of Management Executives, 11(1),37-47.

Choi, T.Y. & Liker, J.K. (1995). BringingJapanese continuous improvement ap-proaches to U.S. manufacturing: Theroles of process orientation and commu-nications. Decision Sciences, 26(5),589-620.

Clark, H.H. & Brennan, S.E. (1991).Grounding in communication. In L.B.Resnick, J. Levine, & S.D. Teasley(Eds.), Perspectives on sociallyshared cognition (pp. 127-149). Wash-ington, DC: American PsychologicalAssociation.

Cosmides, L. & Tooby, J. (1992). Cogni-tive adaptations for social exchange. InJ.H. Barkow, L. Cosmides, & J. Tooby(Eds.), The adapted mind: Evolution-ary psychology and the generation ofculture (pp. 163-228). New York: Ox-ford University Press.

Creswell, J.W. (1994). Research design:Qualitative and quantitative ap-proaches. Thousand Oaks, CA: SagePublications.

Creswell, J.W. (1998). Qualitative inquiry



and research design: Choosingamong five traditions. Thousand Oaks,CA: Sage Publications.

Daft, R.L. & Lengel, R.H. (1986). Orga-nizational information requirements,media richness and structural design.Management Science, 32(5), 554-571.

Daft, R.L., Lengel, R.H., & Trevino, L.K.(1987). Message equivocality, mediaselection, and manager performance:Implications for information systems.MIS Quarterly, 11(3), 355-366.

Davenport, T.H. (1993). Process innova-tion. Boston: Harvard Business Press.

Davenport, T.H. & Stoddard, D.B. (1994).Reengineering: Business change ofmythic proportions? MIS Quarterly,18(2), 121-127.

Davidow, W.H. & Malone, M.S. (1992).The virtual corporation. New York:HarperCollins.

Deming, W.E. (1986). Out of the crisis.Cambridge, MA: Center for AdvancedEngineering Study, Massachusetts Insti-tute of Technology.

Dennis, A.R & Valacich, J.S. (1999). Re-thinking media richness: Towards atheory of media synchronicity. Athens,GA: Terry College of Business, Univer-sity of Georgia.

Dennis, A.R. & Kinney, S.T. (1998). Test-ing media richness theory in the newmedia: The effects of cues, feedback,and task equivocality. Information Sys-tems Research, 9(3), 256-274.

Dennis, A.R., Carte, T.A., & Kelly, G.G.(2003). Breaking the rules: Success andfailure in groupware-supported businessprocess reengineering. Decision Sup-port Systems, 36(1), 31-47.

Dennis, A.R., Haley, B.J., & Vanderberg,R.J. (1996). A meta-analysis of effec-tiveness, efficiency, and participant sat-isfaction in group support systems re-search. In J.I. DeGross, S. Jarvenpaa,

& A. Srinivasan (Eds.), Proceedings ofthe 17th International Conference onInformation Systems (pp. 278-289).New York: The Association for Com-puting Machinery.

Dennis, A.R., Kinney, S.T., & Hung, Y.C.(1999). Gender differences and the ef-fects of media richness. Small GroupResearch, 30(4), 405-437.

DeSanctis, G., Poole, M.S., Dickson, G.W.,& Jackson, B.M. (1993). Interpretiveanalysis of team use of group technolo-gies. Journal of Organizational Com-puting, 3(1), 1-29.

Dobzhansky, T. (1971). Mankind evolv-ing: The evolution of the human spe-cies. New Haven, CN: Yale UniversityPress.

Ellis, C.A., Gibbs, S.J., & Rein, G.L. (1991).Groupware: Some issues and experi-ences. Communications of ACM,34(1), 38-58.

El-Shinnawy, M. & Markus, L. (1998).Acceptance of communication media inorganizations: Richness or features?IEEE Transactions on ProfessionalCommunication, 41(4), 242-253.

Fingar, P., Aronica, R., & Maizlish, B.(2001). The death of “e” and the birthof the real new economy. Tampa, FL:Meghan-Kiffer Press.

Fulk, J., Schmitz, J., & Steinfield, C.W.(1990). A social influence model of tech-nology use. In J. Fulk & C. Steinfield(Eds.), Organizations and communi-cation technology (pp. 117-140).Newbury Park, CA: Sage Publications.

Gardner, H. (1985). The mind’s new sci-ence. New York: Basic Books.

Graetz, K.A., Boyle, E.S., Kimble, C.E.,Thompson, P., & Garloch, J.L. (1998).Information sharing in face-to-face, tele-conferencing, and electronic chat groups.Small Group Research, 29(6), 714-743.

Green, S.B., Salkind, N.J., & Akey, T.M.



(1997). Using SPSS for Windows: Ana-lyzing and understanding data. Up-per Saddle River, NJ: Prentice-Hall.

Grudin, J. (1994). Groupware and socialdynamics: Eight challenges for develop-ers. Communications of the ACM,37(1), 93-105.

Hammer, M. & Stanton, S.A. (1997). Thereengineering revolution. GovernmentExecutive, 27(9), 2-8.

Hart, S.G. & Staveland, L.E. (1988). De-velopment of NASA-TLX (Task LoadIndex): Results of empirical and theo-retical research. In P. Hancock & N.Meshkati (Eds.), Advances in psychol-ogy (pp. 139-184). Amsterdam: NorthHolland.

Ishikawa, K. (1986). Guide to qualitycontrol. Tokyo: Asian Productivity Or-ganization.

Jackson, M.H. & Purcell, D. (1997). Poli-tics and media richness in World WideWeb representations of the formerYoguslavia. Geographical Review,87(2), 219-239.

Jung, C.G. (1968). Analytical psychology:Its theory and practice. New York:Vintage Books.

Kinney, S.T. & Watson, R.T. (1992). Dy-adic communication: The effect of me-dium and task equivocality on task-re-lated and interactional outcomes. In J.I.DeGross, J.D. Becker, & J.J. Elam(Eds.), Proceedings of the 13th Inter-national Conference on InformationSystems (pp. 107-117). Dallas, TX: TheAssociation for Computing Machinery.

Kock, N. (1998). Can communication me-dium limitations foster better group out-comes? An action research study. In-formation & Management, 34(5), 295-305.

Kock, N. (1999). Process improvementand organizational learning: The roleof collaboration technologies.

Hershey, PA: Idea Group Publishing.Kock, N. (2000). Benefits for virtual orga-

nizations from distributed groups. Com-munications of the ACM, 43(11), 107-113.

Kock, N. (2001a). The ape that used e-mail: Understanding e-communicationbehavior through Evolution Theory. Com-munications of the AIS, 5(3), 29. Re-trieved from cais.isworld.org.

Kock, N. (2001b). Behavior toward e-com-munication tools: A new theoreticalmodel based on Evolution Theory. In E.Cohen & A.J. Zaliwski (Eds.), Proceed-ings of the 2001 Informing ScienceConference (pp. 310-317). Krakow,Poland: Cracow University of Econom-ics.

Kock, N. (2001c). Compensatory adapta-tion to a lean medium: An action re-search investigation of electronic com-munication in process improvementgroups. IEEE Transactions on Profes-sional Communication, 44(4), 267-285.

Kock, N. (2002). Evolution and media natu-ralness: A look at e-communicationthrough a Darwinian theoretical lens. InL. Applegate, R., Galliers, & J.L.DeGross (Eds.), Proceedings of the23rd International Conference on In-formation Systems (pp. 373-382). At-lanta, GA: The Association for Informa-tion Systems.

Kock, N. & Davison, R. (2003). Can leanmedia support knowledge sharing? In-vestigating a hidden advantage of pro-cess improvement. IEEE Transactionson Engineering Management, 50(2),151-163.

Kock, N. & McQueen, R.J. (1997). A fieldstudy of effects of asynchronousgroupware support on process improve-ment groups. Journal of InformationTechnology, 12(4), 245-259.

Kock, N. & Murphy, F. (2001). Redesign-



ing acquisition processes: A new meth-odology based on the flow of knowl-edge and information. Fort Belvoir, VA:Defense Acquisition University Press.

Kock, N., Davison, R., Ocker, R., &Wazlawick, R. (2001). E-collaboration:A look at past research and future chal-lenges. Journal of Systems and Infor-mation Technology (Special Issue onE-Collaboration), 5(1), 1-9.

Kock, N., Hassell, L., & Wazlawick, R.S.(2002). Tacit knowledge sharing andhuman evolution: A framework for de-veloping “natural” collaboration technolo-gies. In G.M. Hunter & K.K. Dhanda(Eds.), Proceedings of the 1st

ISOneWorld Conference. Las Vegas,NV: ISOneWorld.

Kock, N., Hilmer, K., Standing, C., & Clark,S. (2000). Supporting learning processeswith collaboration technologies: A brieflook at past research and challenges thatlie ahead. Journal of Systems and In-formation Technology, 4(4), 1-8.

Kock, N., Jenkins, A., & Wellington, R.(1999). A field study of success and fail-ure factors in asynchronous groupware-supported process improvement groups.Business Process Management, 5(3),238-253.

Laitman, J.T. (1984). The anatomy of hu-man speech. Natural History, 20(7),20-27.

Laitman, J.T. (1993). The anatomy of hu-man speech. In R.L. Ciochon & J.G.Fleagle (Eds.), The human evolutionsource book (pp. 56-60). EnglewoodCliffs, NJ: Prentice-Hall.

Lee, A.S. (1994). Electronic mail as a me-dium for rich communication: An em-pirical investigation using hermeneuticinterpretation. MIS Quarterly, 18(2),143-157.

Lengel, R.H. & Daft, R.L. (1988). Theselection of communication media as an

executive skill. Academy of Manage-ment Executive, 2(3), 225-232.

Lieberman, P. (1998). Eve spoke: Humanlanguage and human evolution. NewYork: W.W. Norton & Company.

Lieberman, P. (2000). Human languageand our reptilian brain: The subcor-tical bases of speech, syntax, andthought. Cambridge, MA: Harvard Uni-versity Press.

Markus, M.L. (1994). Electronic mail asthe medium of managerial choice. Or-ganization Science, 5(4), 502-527.

Maxwell, J.A. (1996). Qualitative re-search design: An interactive ap-proach. London: Sage Publications.

McNeill, D. (1998). The face: A naturalhistory. Boston: Little, Brown & Com-pany.

McQueen, R.J., Payner, K., & Kock, N.(1999). Contribution by participants inface-to-face business meetings: Impli-cations for collaborative technology.Journal of Systems and InformationTechnology, 3(1), 15-33.

Myers, M.D. (1997). Qualitative researchin information systems. MIS Quarterly,21(2), 241-242.

Nagasundaram, M. & Bostrom, R.P.(1995). The structuring of creative pro-cesses using GSS: A framework for re-search. Journal of Management In-formation Systems, 11(3), 87-114.

Nardi, B.A., Schwarz, H., Kuchinsky, A.,Leichner, R., Whittaker, S., & Sclabassi,R. (1993). Turning away from talkingheads: The use of video-as-data in neu-rosurgery. Proceedings of the Confer-ence on Human Factors in Comput-ing Systems (pp. 327-334). New York:The Association for Computing Machin-ery.

Ngwenyama, O.K. & Lee, A.S. (1997).Communication richness in electronicmail: Critical social theory and the



contextuality of meaning. MIS Quar-terly, 21(2), 145-167.

Nunamaker, J.F., Dennis, A.R., Valacich,J.S., Vogel, D.R., & George, J.F. (1991).Electronic meeting systems to supportgroup work. Communications of ACM,34(7), 40-61.

Nunamaker, J.F., Briggs, R.O., Mittleman,D.D., Vogel, D.R., & Balthazard, P.A.(1997). Lessons from a dozen years ofgroup support systems research: A dis-cussion of lab and field. Journal ofManagement Information Systems,13(3), 163-207.

Olesen, K. & Myers, M.D. (1999). Tryingto improve communication and collabo-ration with information technology: Anaction research project which failed. In-formation Technology & People,12(4), 317-332.

Patton, M.Q. (1980). Qualitative evalua-tion methods. Beverly Hills, CA: SagePublications.

Patton, M.Q. (1987). How to use qualita-tive methods in evaluation. NewburyPark, CA: Sage Publications.

Pinker, S. (1994). The language instinct.New York: William Morrow & Co.

Pinker, S. (1997). How the mind works.New York: W.W. Norton & Co.

Pinker, S. & Bloom, P. (1992). Natural lan-guage and natural selection. In J.H.Barkow, L. Cosmides, & J. Tooby(Eds.), The adapted mind: Evolution-ary psychology and the generation ofculture (pp. 451-493). New York: Ox-ford University Press.

Postmes, T., Spears, R., & Lea, M. (1998).Breaching or building social boundaries?Side-effects of computer-mediated com-munications. Communication Research,25(6), 689-715.

Rice, R.E. (1992). Task analyzability, useof new media, and effectiveness: Amulti-site exploration of media richness.

Organization Science, 3(4), 475-500.Rice, R.E. (1993). Media appropriateness:

Using social presence theory to com-pare traditional and new organizationalmedia. Human Communication Re-search, 19(4), 451-484.

Rice, R.E & Shook, D.E. (1990). Relation-ship of job categories and organizationallevels to use of communication channels,including electronic mail: A meta-analy-sis and extension. Journal of Manage-ment Studies, 27(2), 195-230.

Rosenthal, R. & Rosnow, R.L. (1991).Essentials of behavioral research:Methods and data analysis. Boston:McGraw-Hill.

Sallnas, E.L., Rassmus-Grohn, K., &Sjostrom, C. (2000). Supporting pres-ence in collaborative environments byhaptic force feedback. ACM Transac-tions on Computer-Human Interac-tion, 7(4), 461-476.

Schacter, D.L. (2001). The seven sins ofmemory: How the mind forgets andremembers. New York: HoughtonMifflin.

Short, J., Williams, E., & Christie, B. (1976).The social psychology of telecommu-nications. London: John Wiley & Sons.

Siegel, S. & Castellan, N.J. (1998). Non-parametric statistics for the behav-ioral sciences. Boston: McGraw-Hill.

Soles, S. (1994). Work reengineering andworkflows: Comparative methods. In E.White & L. Fischer (Eds.), Theworkflow paradigm (pp. 70-104).Alameda, CA: Future Strategies.

Sommer, B. & Sommer, R. (1991). A prac-tical guide to behavioral research.New York: Oxford University Press.

Standing, C. & Benson, S. (2000). An ef-fective framework for evaluating policyand infrastructure issues for e-com-merce. Information Infrastructure &Policy, 6(4), 227-237.



Stinchcombe, A.L. (1968). Constructingsocial theories. New York: HarcourtBrace.

Tooby, J. & Cosmides, L. (1992). The psy-chological foundation of culture. In J.H.Barkow, L. Cosmides, & J. Tooby(Eds.), The adapted mind: Evolution-ary psychology and the generation ofculture (pp. 19-136). New York: OxfordUniversity Press.

Trevino, L.K., Webster, J., & Stein, E.W.(2000). Making connections: Comple-mentary influences on communicationmedia choices, attitudes, and use. Or-ganization Science, 11(2), 163-182.

Walther, J.B. (1996). Computer-mediatedcommunication: Impersonal, interper-sonal, and hyperpersonal interaction.Communication Research, 23(1), 3-43.

Walton, M. (1991). Deming managementat work. London: Mercury.

Warkentin, M.E., Sayeed, L., & Hightower,R. (1997). Virtual teams versus face-to-face teams: An exploratory study ofa Web-based conferencing system. De-cision Sciences, 28(4), 975-996.

Webster, J. & Trevino, L.K. (1995). Ratio-nal and social theories as complemen-tary explanations of communicationmedia choices: Two policy-capturingstudies. Academy of ManagementJournal, 38(6), 1544-1573.

Weick, K.E. (1995). Sensemaking in or-ganizations. Thousand Oaks, CA: SagePublications.

Wills, C. (1989). The wisdom of the genes:New pathways in evolution. New York:Basic Books.

Wills, C. (1993). The runaway brain: Theevolution of human uniqueness. NewYork: Basic Books.

Zigurs, I. & Buckland, B.K. (1998). Atheory of task/technology fit and groupsupport systems effectiveness. MISQuarterly, 22(3), 313-334.

APPENDIX A

Process Modeling Task

The U.S. Department of Defense(DoD) routinely acquires software devel-opment services from several companies,including Lockheed Martin and ComputerSciences Corporation. Our task involves theredesign of a hypothetical business process,called the Software Services AcquisitionProcess (SSAP), through which a branchof the DoD contracts the development ofcomputer-based defense systems. For thesake of illustration, one such computer-based defense system contracted in the pastis the Amphibious Tank Radar Manager(ATRM—a fictitious name), which auto-mates the operation of the radar of an am-phibious tank.

You and your partner are receivingdifferent pieces of information about theSSAP process (listed in bullet items below),which starts with the DoD issuing a re-quest for proposals (RFP) and ends withthe DoD receiving a proposal (or severalproposals) from a bidder (or a group of bid-ders). You will now first discuss these dif-ferent pieces of information with each otherand each develop a sketch of the entireprocess. Please bear in mind that eachof you will develop a separate sketchbased on your discussion. You may dis-cuss the sketch as much as you want,but you must not show your sketch toeach other.

Information Provided to theFirst Member of the Dyad

• The SSAP process starts with DoD’sprogram manager e-mailing a requestfor proposals (RFP) document to the bidconsultant of a company calledWebmasters.



• The DoD program manager then callsWebmasters’ bid consultant over thephone to provide him with extra infor-mation about the RFP, including thedeadline for receipt of proposals andtypes of organizations eligible to bid.

• Webmasters’ bid consultant uploads theRFP from his e-mail inbox onto DoD’sRFP’s Web site using a system calledSecureWorkflow, which automaticallynotifies potential bidders by e-mail aboutthe RFP.

• The contracts manager of a bidder re-ceives the RFP notification in his e-mail’sinbox folder, downloads the RFP fromDoD’s RFP Web site usingSecureWorkflow, and places it in his per-sonal folder.

• As soon as he has some time, thebidder’s contracts manager e-mails theRFP to the bidder’s technical lead.

Information Provided to theSecond Member of the Dyad

• Once a day, the bidder’s technical leadchecks her e-mail’s inbox folder forRFPs. Once an RFP is found, she en-ters the cover page information aboutthe RFP into a control spreadsheet andsaves the RFP document in her RFPsfolder.

• As soon as she has some time, thebidder’s technical lead e-mails the RFPto the bidder’s project manager.

• Once a week, the bidder’s project man-ager checks his e-mail inbox for newRFPs and, once he finds one, preparesa related proposal in consultation withhis team.

• The bidder’s project manager e-mails theprepared proposal to the bidder’s tech-nical lead.

• The bidder’s technical lead enters thecover page information about the pro-posal into the same control spreadsheetin which she previously entered infor-mation about the RFP, and uploads theproposal onto DoD’s proposals Web siteusing the SecureWorkflow system,which automatically notifies DoD’s pro-gram manager of the receipt of a newproposal.

APPENDIX B

Process Redesign Task

You will now apply the process rede-sign techniques discussed below to theSSAP process, which should lead to a sim-plification of the process. You and yourpartner are receiving only half of the rede-sign techniques each, so you will have todiscuss the other techniques with each otherin order to be able to apply all of the pro-cess redesign techniques. Finally, you willeach develop a sketch of the simplified pro-cess. Please bear in mind that each ofyou will develop a separate sketchbased on your discussion. You may dis-cuss the sketch as much as you want,but you must not show your sketch toeach other.

Redesign Guidelines Provided to theFirst Member of the Dyad

• Foster Asynchronous Communication.When people exchange information, theycan do it synchronously, i.e., interactingat the same time, or asynchronously, i.e.,interacting at different times. One ex-ample of synchronous communication isa telephone conversation. If the conver-sation takes place via e-mail, it then be-comes an example of asynchronous



communication. It has been observed,especially in formal business interaction,that in the vast majority of cases, asyn-chronous communication is more effi-cient than synchronous communication.For example, synchronous communica-tion often leads to wasted time (e.g.,waiting for the other person to be found)and communication tends to be less ob-jective. Asynchronous communicationcan be implemented with simple artifactssuch as in-boxes and out-boxes, faxmachines, and billboards. Asynchronouscommunication can also be implementedwith more complex artifacts such ascomputer files. These artifacts work asdynamic information repositories.

• Reduce Information Flow. Excessiveinformation flow is often caused by anover-commitment to efficiency to thedetriment of effectiveness. Informationis perceived as an important componentof processes which drives people to anunhealthy information hunger. Thiscauses information overload and the cre-ation of unnecessary information pro-cessing functions within the organization.Information overload leads to stress and,often, the creation of information filter-ing roles. These roles are normally thoseof aides or middle managers, who areresponsible for filtering in the importantbit from the information coming from thebottom of, and from outside, the organi-zation. Conversely, excessive informa-tion flowing top-down forces middlemanagers to become messengers, to thedetriment of more important roles. In-formation flow can be reduced by se-lecting the information that is importantin processes and eliminating the rest, andby effectively using group support anddatabase management systems.

Redesign Guidelines Provided to theSecond Member of the Dyad

• Reduce Control. Control activities donot normally add value to customers.They are often designed to prevent prob-lems from happening as a result of hu-man mistakes. In several cases, how-ever, control itself fosters neglect, witha negative impact on productivity. Forexample, a worker may not be carefulenough when performing a process ac-tivity because he knows that there willbe some kind of control to catch his mis-takes. Additionally, some types of con-trol, such as those aimed at preventingfraud, may prove to be more costly thanno control at all. Some car insurancecompanies, for example, have found outthat the cost of accident inspections, fora large group of customers, was muchmore expensive than the average costof frauds that that group committed.

• Reduce the Number of Contact Points.Contact points can be defined as pointswhere there is interaction between twoor more people, both within the processand outside. This involves contacts be-tween functions, and between functionsand customers. Contact points generatedelays and inconsistencies and, when inexcess, lead to customer perplexity anddissatisfaction. In self-service restau-rants and warehouses, for example, thepoints of contact were successfully re-duced to a minimum. Additionally, it ismuch easier to monitor customer per-ceptions in situations where there are asmall number of contact points. Thismakes it easier to improve process qual-ity.



APPENDIX C

Correct Answer for Process Modeling Task

DoD’s program manager

-RFP

Webmaster’s bid consultant’s email

inbox folder

Prepare proposal(Bidder’s

project manager)

-extra RFP information

-RFPDoD’s RFP’s

Web site

-RFP

Download RFP(Bidder’s contracts

manager)

-RFP

Bidder’s contractsmanager’s

personal folder

-RFP

Bidder’s contracts manager email

inbox folder-RFP notification

-RFP notification

Bidder’s technical lead’s emailinbox folder

Bidder’s contractsmanager

-RFP

-RFP

Maintain control spreadsheet(Bidder’s

technical lead)

-RFP

Bidder’s technical lead

control spreadsheet

Bidder’s technical lead email inbox

folder

-RFP

-RFP cover page details

Bidder’s technical lead

-RFP

Bidder’s project

manager email inbox folder

-RFP

Upload RFP(Webmaster’s bid

consultant)

-RFP

-proposal

-proposal

-proposal cover page details

-proposal

Upload proposal(Bidder’s

technical lead)

DoD’sproposal’sWeb site

-proposal

-proposal notification

-proposal

APPENDIX D

Correct Answer for Process Redesign Task

DoD’s RFP’s Web site

-RFP

Download RFP(Bidder’s contracts

manager)

-RFP

Bidder’s contracts manager email

inbox folder-RFP notification

-RFP notificationUpload RFP

(DoD’s program manager)

Bidder’s RFP’s and proposals

-RFP

Prepare proposal(Bidder’s

project manager)

-RFP

-proposal

Upload proposal(Bidder’s contracts

manager)

DoD’s proposals Web site

-proposal

-proposal

-proposal notification



APPENDIX E

Questionnaire Measures

Ned Kock is associate professor and chair of the Department of MIS and Decision Science atTexas A&M International University. He holds degrees in electronics engineering (BEE),computer science (MSc), and management information systems (PhD). Dr. Kock has authoredseveral books and published in a number of journals including Communications of the ACM,Decision Support Systems, IEEE Transactions on Professional Communication, Information &Management, Information Systems Journal, Information Technology & People, Journal ofOrganizational Computing and Electronic Commerce, Journal of Systems and InformationTechnology, MIS Quarterly, and Organization Science. He is editor-in-chief of the InternationalJournal of e-Collaboration, associate editor of the Journal of Systems and Information Technology,and associate editor for information systems of the journal IEEE Transactions on ProfessionalCommunication. His research interests include action research, ethical and legal issues intechnology research and management, e-collaboration, and business process improvement.

Cognitive effort

Indicate how much effort it took for you to complete the process redesign task by marking anywhere

on the continuous scale below with an “X”.

0 ----- 25 ----- 50 ----- 75 ----- 100 ----- 125 ----- 150

Absolutely Extreme no effort effort

Communication ambiguity

Communication between my partner and myself was ambiguous.

1 ------- 2 ------- 3 ------- 4 ------- 5 ------- 6 ------- 7

Strongly Strongly Disagree Agree

Message preparation

I had to spend a lot of time preparing my contributions (e.g., group of sentences summarizing facts

or conveying ideas) in my discussions with my partner.

1 ------- 2 ------- 3 ------- 4 ------- 5 ------- 6 ------- 7

Strongly Strongly Disagree Agree

compensatory adaptation to media obstaclescits.tamiu.edu › kock › pubs › journals ›...

Documents