iv THE ROLE OF COMPUTER MEDIA TED COMMUNICATI ON COMPETENCE ON UNIQUE INFORMATION POOL ING AND DECISION QUALITY IN VIRTUAL TEAMS By Heng Chen Xie A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of PsychologyŠMaster of Arts 2015 iii ABSTRACT THE ROLE OF COMPUTER MEDIA TED COMMUNICATI ON COMPETENCE ON UNIQUE INFORMATION POOL ING AND DECISION QUALITY IN VIRTUAL TEAMS By Heng Chen Xie Past studies and reviews on the role of Computer Mediated Communication (CMC) mediums on unique information pooling and decisi on quality in virtual teams has largely been inconsistent and, at times, cont radictory. In this study I argue that the inconsistency in past findings is due to a flawed interpretation of CM Cs as being equivalent forms of communication, when they should instead be viewed as distinct mediums varying in media richness. In this study I present a new model of decision-making where by the media richness of the CMC being used and the team™s level of CMC competence wi ll predict decision quality through unique information pooling. Results indicated that th ere were significant differences in unique information pooled between high and low media richness conditions, with teams in the high media richness condition pooling significantly mo re unique information. However, team CMC competence was not found to predict unique in formation pooling, and unique information pooling was not found to predict decision quality. iii TABLE OF CONTENTS LIST OF TABLES ................................................................................................................ ........ iv LIST OF FIGURES ............................................................................................................... ....... v Introduction ................................................................................................................................... 1 Information Sharing and Decision-Making in Virtual Teams ...................................................... 4 Media Richness Theory ......................................................................................................... ... 9 Media Naturalness Hypothesis ................................................................................................. 13 Media Compensation Theory ................................................................................................... 1 5 CMC Competence and Individual Differences ........................................................................ 21 Model & Hypotheses ............................................................................................................ ........ 26 Method ........................................................................................................................ .................. 31 Participants .................................................................................................................. ............. 31 Measures ...................................................................................................................... ............. 31 Apparatus/Materials ................................................................................................................. 34 Procedure .................................................................................................................................. 37 Results ....................................................................................................................... .................... 40 Discussion .................................................................................................................... ................. 52 Limitations and Future Directions ............................................................................................. ... 60 APPENDICES .................................................................................................................... .......... 63 Appendix A: Initial Measures .................................................................................................. 64 Appendix B: Exploratory Factor Analysis Results Œ Principal Axis Factor, Direct Oblimin Rotation .................................................................................................................................... 68 Appendix C: Revised Measures ............................................................................................... 71 Appendix D: Task Information Packets ................................................................................... 74 Appendix E: Checklist Form .................................................................................................... 87 REFERENCES .................................................................................................................... ......... 90 iv LIST OF TABLES Table 1. Distribution of Informa tion in Hidden Profile Task ....................................................... 36 Table 2. Means, Standard Deviations, and Correlations among Individual -Level Variables ....... 41 Table 3. Regression Analysis Predicti ng Engagement (Individual-level) .................................... 42 Table 4. Means, Standard Deviations, and Correlations among Team-L evel Variables .............. 45 Table 5. Logistic Regression Analysis of Unique Information P ooling on Decision Quality ...... 46 Table 6. Means, Standard Deviations, and Correlations among Individual-L evel Variables (Alt.)....................................................................................................................................................... 49 Table 7. Means, Standard Deviations, and Correlations among Team-Level Variables (Alt.) .... 50 Table 8. Logistic Regression Analysis of Unique Information Pooling on Decision Quality (Alt.) ....................................................................................................................................................... 51 v LIST OF FIGURES Figure 1. Media Richness/Naturalness Continuum ...................................................................... 15 Figure 2. Model of Team Decision-Making ................................................................................. 26 1 Introduction The organizational landscape has been rapidly evolving the past few decades. Globalization has increased both the reach and span of organizations, leading to a proliferation of multinational and transnational workforces that transcend the boundaries of time and geography (Martins, Gilson, & Maynard, 2004). The work team is increasingly becoming the focal point of work structures due to its agility in dealing with rapidly changing environments without having to await for orders from top management, and also because of its ability to leverage the skills and expertise of its members to interpret a divers e range of information to deal with complex decision-making tasks(Mesmer-Magnus & De Church, 2009). Advances in communication technology have allowed for the decentralization of teams, allowing for the creation of virtual teams that can access the exper tise and skills of members locat ed throughout the world, while lowering costs associated with relocating or importing personnel. With so many potential benefits, it™s no surprise why mo re and more organizations are a dopting the use of virtual teams (Copeland, 2006). However, it would be folly to see only the upsides without considering the potential downsides of having people work and communicate virtually rather than naturally in a face-to-face (FTF) context. An important consid eration is that when teams use computer- mediated communication (CMC) mediums, or a ny type of communication medium that is facilitated through a computer-interface (e.g. instant messaging, e-mail, Skype), the manner in which they share and interpret information changes. In fact, there has been considerable research arguing that virtual collaboration using CMCs often lead to slower decision making, lower group effectiveness and lower satisfaction (Baltes, Dickson, Sherman, Bauer, & LaGanke, 2002; Kiesler, Siegel, & McGuire, 1984; Kiesler & Sprou ll, 1992). Furthermore, research findings have shown that information sharing, a central process of effective team sharing, may be hampered by 2 the limitations of the CMC in its capacity to transmit informa tion, the degree to which it is similar to face to face interactions, and the virt uality of the team using the CMC (Daft & Lengel, 1984; Kock, 2004; Mesmer-Magnus, DeChurch, Ji menez-Rodriguez, Wildman, & Shuffler, 2011). As such, research aimed at understandin g how the unique context of the virtual team changes the way in which people interact and collaborate is critical in maximizing the potential benefits of virtual teams, while minimizing the downsides. (Stahl, Maznevski, Voigt, & Jonsen, 2010). Unfortunately, empirical findings regard ing communication technology have been inconsistent at best and a recent meta-analy sis by Lu, Yuan, and McLeod (2012) actually found no significant effects of communication me dium (CMCs vs. Face- to-Face) on unique information pooling and decision quality. However, this conclusion has to be interpreted cautiously as they noted many of the studies they looked at were decades old and both the technology and user base has mature d since then. Back in the 90™s personal computers were still relatively new and likewise, the user base was also less familiar and inexperienced with using CMCs to communicate. However, technology and virtuality are now ingrained into our everyday lives (e.g., e-mail, texting, Facebook, Twitter, Skype) and it would not be difficult to argue that today™s society is more technologi cally literate than ever before. Thus it would only make sense for us to re-examine the effects of CMC tec hnology using a sample experienced with virtual communication and more representative of the modern workforce. As such, in this study I will be using a sample of undergraduate students who should be sufficiently familiar and adept with CMC technology. Furthermore, there is an important defici ency in how information sharing and team performance is studied in virtua l teams today. Researchers in the information sharing literature 3 have largely neglected the characteristics of the specific CMC be ing utilized; lumping together CMCs ranging from e-mail, to audio-conferenci ng, to group decision support systems to video conferencing into an all-encompassing CMC category for meta-analyses(Lu et al., 2012; Mesmer-Magnus et al., 2011). This questionable practice has led to predictably ambiguous and contradictory findings within the literature, as it should seem rather obvious that communicating with e-mail would be very diffe rent from communicating through a video conference call. It is time for researchers to clearly look at the distinct differences between types of CMCs, so that we can understand how these differences can influence information exchange and decision making. One important distinction between types of CMCs I will be discussing in this study is the CMC™s media richness, or the medium™s capacity for immediate feedback, the number of cues and channels utilized, personalization, and language variety (Daft & Wiginton, 1979). As a critical process of team decision-making is the ability to share information, it is expected that since virtual teams communicate exclusively through CMCs, the CMC™s media richness will have a significant impact on how quickly information ca n be disseminated within the group and what type of information can be shared within the group. Lastly, I introduce the concept of CMC competence, or an individual™s belief in his or her ability to communicate effectively using CMCs, to show that not only is it the CMC being used that can affect information sharing and decision- making in virtual teams, but also how the CMC is being used and by whom. Many studies have s hown that there are inte r-individual differences in communication efficacy and personality (J oinson, 2004; Spitzberg, 2006; Tosun & Lajunen, 2010) that can affect how people communicate and build relationships virtually. As such, it is likely that a member™s self-perceived comp etence in communicating virtually can have a significant impact on successful information sharing within the team . It is expected that, as a 4 result of different experiences and predilections, individuals are likely to vary in the degree to which they are competent in using CMCs. A team with competent users that are familiar with the CMC being utilized would very likely outperform a team with users that are technologically illiterate or wary of communicating virtually. As such, the role of CMC competence should be a critical component of effective info rmation sharing in virtual teams. Information Sharing and Decision-Making in Virtual Teams Virtual teams are teams that rely on comput er mediated communica tions (CMCs) and can cross boundaries of geography, time and organiza tion (Martins, Gilson & Maynard, 2004). There are thought to be many perks associated with thes e types of teams such as allowing collaboration between members who are in different countries or in different organizations (can fihirefl outside assistance or a specialist), or saving on the co sts of relocating member s (regular face to face teams require physical proximity). However, be cause virtual teams comm unicate and interact primarily using CMCs rather than through face-to -face interactions, virtual teams can encounter significant barriers to effective team performance such as difficulty in establishing trust between members, forming relationships, and may exchange /interpret information differently than face- to-face teams (Mesmer-Magnus & DeChurch, 2009; Mesmer-Magnus et al., 2011; Rosen, Furst, & Blackburn, 2007; Walther & Bunz, 2005). As a critical process of effective decision-making in teams is the effective sharing of information, there has been extensive research on how information sharin g is influenced by the context of the virtual team. One important avenue of research has focused on the effects of CMC on the hidden profile and the shared informa tion bias (Stasser & Titus, 1985, 1987). Hidden profile research has shown that decision-maki ng teams are more likely to make incorrect 5 decisions when members share common information that is biased towards an incorrect profile (Lu et al., 2012). This finding is attributed to the shared information bias where shared information is more likely to be menti oned, re-mentioned and supported, whereas unique information (information that is not known by a ll members) is often not discussed at all or unsupported by others and thus discounted. Studies on the shared information bias have shown that teams must not only posse ss the knowledge to make the co rrect decision, but also possess the ability to share that know ledge between members so that appropriate action can be taken. Thus, it is not surprising that unique information pooling has been found to be the strongest predictor of decision quality (Lu et al., 2012; Mesmer-Magnus & DeChurch, 2009) and we would expect that teams that are able to overc ome the shared information bias will be more likely to reach the correct decision. As previously mentioned, research looking at the effect of co mmunication medium on unique information pooling and decision quality has been somewhat conflicted. While Lu et al. (2012) found no overall effect of communication medium on unique information pooling and team decision quality, there are some studies th at do suggest a more pos itive note for CMC use. For example, Lam and Schaubroeck (2000) f ound that teams using group decision support systems (GDSS) were much more likely to shar e unique information than FTF teams, and that overall GDSS teams significantly outperformed FT F teams in hidden profile decision-making tasks. On the other hand, studies such as Kerr and Murthy (2009) found that teams utilizing a CMC chat tool were less successful in exchangi ng and processing information in a hidden profile task and thus less successful than FTF teams in correctly solving the hidden profile task. Yet Crede and Sniezek (2003) found no difference be tween video-conferencing groups and face-to-face groups in regards to decision accuracy, ove rconfidence or commitment to group decisions. 6 At face value it may seem that perhaps Li Lu and her colleagues were right, perhaps these contradictory findings mean that there is no consistent effect of CMC on either unique information pooling or decision quality. However, if one were to look carefully at the studies, it becomes clear that their findings are based on the assumption that all CMCs are equal. One important distinction is that in each of the th ree studies mentioned a di fferent CMC (GDSS, chat tool, video conferencing) is being compared to f ace-to-face teams. However, it would be folly to consider all CMCs to be equivalent. For example, when using a video conferencing CMC you are able to hear and see the person you are communi cating with, as compared to using a chat tool where you are unable to perceive visual and audio information. In f act, it could be argued that video conferencing is more similar to face to face communication than it is to a chat tool. Unfortunately, I have not been able to find any empirical studies comparing different types of CMC and their influence on team unique inform ation pooling and/or de cision quality and will seek to remedy that problem in this study. One potential manner in which predictions can be made regarding how differences between CMCs may affect unique information pooling and decision quality comes from the cues filtered out perspective (Culnan & Markus, 1987). This perspective argues that because CMCs transmits less nonverbal, contextual and social cues compared to face to face communication, the way in which members perceive and interact with one another become markedly different from face-to-face communication. It was thought that due to the lack of contextual and social cues, communication between members in virtual teams may become less personal and/or result in a lack fisocial presencefl (Short, Williams, and Christie, 1976). As a result, it would be expected that communication using CMCs that severely limit the amount of nonverbal, contextual and social cues (e.g. e-mail) may become more ta sk-focused and less affectively laced whereas 7 CMCs that allow for the transmission of additi onal cues will be more similar to face-to-face communication (e.g. video conferencing). However, this increased focus on the task at hand and the blocking of extraneous information/backgrou nd noise may in some cases actually facilitate better task performance. One school of thought on how information shari ng differs in a virtual context is that virtual teams benefit from its lack of socia l/contextual/visual cues through social/status equalization (Driskell, Radtke, & Salas, 2003; Siegel, Dubrovsky, Kiesler, & McGuire, 1986). Typically in small group discussion, there are st rong group norms and pressures at play leading to inhibition to share information. Those viewed as having higher status and prestige will be afforded more opportunities to lead conversation, to be recognized and agreed with (Driskell & Mullen, 1990), whereas low status members may fear to speak out against those high status members for fear of censure or reprisal. Expect ation states theory (Berger, Cohen, & Zelditch, 1972) suggests that people attribute expectations to status characteristics perceived to be salient in the situation (i.e. deference would be given to a doctor during a medical emergency). These attributions can create an informal hierarchy based on power and prestige within the group. However in some virtual teams these status char acteristics are suppressed or made less salient because the CMC limits what type of informati on can be transmitted. For example, oftentimes seniority can be a powerful status characteristic (e.g. younger workers may defer to the veterans as they are thought to be more knowledgeable), but when visual cues are suppressed (such as when using e-mail as the CMC) this hierarchy ma y fail to appear and afford more opportunities to those who may traditionally be in a filowerfl st atus. As such, we would expect the amount of contextual information conveyed through the co mmunication medium to affect not only how information is shared and accepted/rejected, but also who is able to step up to the plate and share. 8 This suggests that teams using CMCs that limit the type of information shared to the bare minimum (i.e. only text information) will not on ly be more task-focused and unencumbered by extraneous information, but may also promote more equal contribution amongst team members, leading to more unique information pooled. Some support for this conc lusion comes from a meta-a nalysis conducted by Mesmer- Magnus et al. (2011) on the effect of team virtuality on information sharing and team performance. Similar to the argument presented in this study, they argued that it makes little sense to compare only face-to-face teams to virtual te ams as there are degrees of virtuality within teams. Instead, they classified virtual teams as either high or low on virtuality, drawing from Kirkman and Mathieu (2005), which presented three dimensions of virtuali ty: the proportion of communication that was done exclusively through CMCs, the extent to which CMC transmits data that is valuable for team effectiveness, and the synchronicity of the CMC (real time vs. lagged response). High virtuality teams would be teams that communicate almost exclusively through CMCs that had a heavy delay and had severe limitations on what type of information could be transmitted (e.g. virtual teams communi cating via e-mail). Low virtuality teams would be teams that communicate using both CMC and f ace-to-face, and/or use CMCs that are high in synchronicity and allow for many different types of information to be shared (e.g. co-located virtual team that occasionally vi deo conferences meetings). Thei r findings revealed that teams high in virtuality were more successful in pooling unique information as compared to low virtuality teams and face-to-face teams which were more successful in openly sharing high volumes of information. However, one must keep in mind that while team virtuality is in part derived from media richness, they ar e still distinct concepts. Team vi rtuality is a characteristic of 9 a team and can be altered (e.g. meeting more frequently in person, changing the CMC being used), whereas media richness is an intrinsic property of the CMC. Media Richness Theory The Media Richness Theory (Daft & Lengel, 1984; Daft & Lengel, 1986) was initially developed to describe how different communication mediums differ in their capacity to exchange information and it is where concept of media richness was first coined. To understand their theory, one must first clarify what they contend are the three fundamental issues at play in situations requiring information exchange a nd also what differentiates one communication medium from another. Information and richness: Firstly, they cont end that information differs in richness . Richness of information is defined as its ability to change understanding within a time interval by overcoming different frames of referen ce or clarifying ambiguous issues. Within organizations and the workgroups, information n eeds to be shared, pooled and then processed before a decision can be made. Information™s main purpose is twofold, to reduce uncertainty (i.e. absence of information) and lim it equivocality (i.e. ambiguity or conflicting views regarding a situation). Situations that organizations a nd workgroups face can vary in the amount of uncertainty and equivocality present. In situations of high uncertainty, there exists a gap between the group™s information and the amount of information necessary for the gro up to make a quality decision. For example, a situation characterized by high uncertainty might be one where a promotion committee examining performance reviews is missing a few m onths™ worth of performance data. Situations that are characterized instead by high equivoca lity are ones where there are conflicting views on 10 what is correct or most appropriate, where th e problem space is poorly defined. This would be especially prominent in diverse groups where members hold differing values/view and interpret the information differently. For example, the Su preme Court of the United States must make decisions based on their interpre tation of the Constitution. However, the manner in which the Justices of the Supreme Court in terpret the Constitution can vary dramatically, and oftentimes there is no simple clear-cut corr ect answer because the Constitution was not meant to deal with many modern issues. Information that is high in richness possesse s the ability to quickl y change/persuade one to accept an alternate view, or strength en one™s support of a view by reducing uncertainty/ambiguity. It needs to be noted here , however, that just because the information is rich does not necessarily mean it will be accepted. It simply means that rich information should be more effective in changing opi nions than low richness information that is hard to understand or unconvincing. For example, let us say you are a jury member in a murder trial. Both the prosecutor and defense attorney present you w ith different pieces of evidence/testimony accompanied with explanations seeking to persuade your judgment. Some evidence or testimony may be more effective in helping you reach a decision, and some will be less effective. Strong, rich evidence (e.g. a written confession or a vide otape of the crime in action) will help you to quickly come to a decision; whereas weaker, less rich evidence (e.g. witness testimony from someone that seems highly unreliable, blurry vide otape) may be ineffective in helping you come to a conclusion. Ability to convey rich information: Secondl y, they contend that different types of communication media differ in their richness, or ability to convey rich information. Media richness is defined as the medium™s capacity for immediate fee dback, the number of cues and 11 channels utilized, personalization, and language variety (Daft & Wiginton, 1979). As this theory was first proposed prior to the digital age, th eir ordering of mediums was limited, with face to face being the highest in richness followed by the telephone, letters/memos, impersonal written memos, and numeric documents. Face to face was thought to the be richest medium because it happens in real-time, allowing for instant comm unication and feedback while also allowing for additional, supportive information su ch as nonverbal cues, tone and gestures to come into play as well. Documents were considered less rich becaus e there was no instant feedback (they took time to be sent and read), and were less fipersuas ivefl because they were not supplemented with nonverbal cues etc. CMCs, like all other forms of communication medi ums, also vary in their richness. Rich mediums such as video conferencing are nearly equal to face-to-face communication since they are able to transmit a signifi cant amount of contextual, verbal and audio information and is nearly as synchronous as face to face communica tion. On the lower end of the media richness spectrum we would place instant messaging, as it is severely lim ited in its ability to convey visual/audio/contextual informati on. At its core the Media Richness Theory has five main tenets: a. Situations vary in uncertainty and equivocality that must be addressed before a decision can be made. b. Information varies in its richness, its ability to reduce uncertainty and resolve equivocality. c. Communication mediums vary in their ability to transfer rich information. Media Richness is determined by the medium™s capacity for immediate feedback, the number of cues and channels utilize d, personalization and language variety. 12 d. Certain situations, such as those characterized by high equivocality and that require debate/rapid exchange to reach consensus, will benefit more from high richness mediums whereas other situations can be re solved with low richness mediums. Thus improperly matching the communication medium to the situation will lead to sub- optimal performance outcomes. e. People will learn to match, and prefer co mmunication mediums that meet the optimal level of richness. More richness, while always better, is not always necessary. The Media Richness Theory has met with mi xed empirical results. While there were a number of papers that found supporting evidence that people performed better and were more satisfied when the type of medium matched th e situation (Daft & Lengel, 1986; Graetz, Boyle, Kimble, Thompson, & Garloch, 1998; Rice, 1993; Sp roull & Kiesler, 1986), there were also those that found little differences in performan ce outcomes when using low richness mediums vs. high richness mediums (Crede & Sniezek, 2003 ; Dennis & Kinney, 1998; ElShinnawy & Markus, 1997; Hiltz, Johnson, & Turoff, 1986). Additionall y, managers and workers commonly utilized and preferred e-mail on a variety of tasks and si tuations where it would be deemed as a sub- optimal match in terms of medi a richness (Markus, 1994). However, despite the mixed results of media richness on group performance outcomes, the FTF medium remained the preferred medium of choice for participants in most studies, especially when the task was more complex or required higher levels of coordination or di scussion (Adrianson & Hjelmquist, 1991; Straus & McGrath, 1994). Additionally, even in situations where performance was the same, or when Media Richness Theory would argue a richer me dium was not necessary, people consistently rated higher satisfaction when using mediums highe r in richness. To address this particular phenomenon, and to provide an alternative expl anation for why richer mediums are typically 13 preferred over lower richness mediums, the Media Naturalness Hypothesis and, subsequently, the Media Compensation Theory was develope d (Hantula, Kock, D'Arcy, & DeRosa, 2011; Kock, 2004, 2005). Media Naturalness Hypothesis Kock™s (Kock, 2004) basic argument wa s that people preferred face-to-face communication the most because it is the most natural way for humans to communicate. If we consider that the face-to-face communication ha s been facilitated by millions of years of evolution, such as the development of special fa cial muscles dedicated to facial expression or specialized brain circuits dedica ted to deciphering facial and speech recognition, one can hardly be surprised why we feel most comfortable with our de facto mode of communication. We must also consider that FTF communicat ion brings with it the ability to communicate in synchronicity, complemented by a vast array of facial expre ssions, body gestures, speech tones and inflections that even the most advanced vide o-conferencing software fails to completely capture. To that end, Kock created the term media naturalness, or the ability of communication me dia to support co-located and synchronous communication employi ng facial expressions, body language and speech. He defined seven elements of media na turalness, and CMCs possessing more would be considered more natural and those possessing less would be considered less natural. The seven elements are: a) individuals are co-located and can scan, see and h ear one another, b) there is a high degree of synchronicity that allows individuals to quickly interact with each other, c) individuals have the ability to observe and convey facial expressions, d) individuals are able to observe and convey body language, e) individuals can convey and listen to oral speech, f) individuals are able to engage in mutual gaze; making and holdi ng (or avoiding) eye contact and seeing where other people are loca ted, and finally g) individuals are able to use and sense subtle 14 olfactory and tactile stimuli such as pheromones or a light touch. In a sense, when we remove certain aspects of our natural communication, such as synchronicity or facial expressions, we are essentially ficripplingfl the way we communicate. He theorized that when naturalness was low it would lead to an increase in cognitive effort (increased neural activity in th e brain), increased communication ambiguity (misinterpretation due to missing cues), and lower physiological arousal (lower task engagement). It was expected that these factors would mediat e the relationship between media naturalness and performance outcomes and that the less natu ral the medium, the worse its impact on communication and thus performance. However, hi s theory has not been tested empirically. At first glance, it would seem that Media Naturalness and Media Richness are highly similar, as after all, mediums that contain more elements of media naturalness would also be higher in media richness (e.g. vide o conferencing is both more na tural and rich than e-mail). However, the distinction lies in that Media Ri chness Theory, at its core, argues that more richness is always better as it facilitates the rapid exchange of information and resolution of uncertainty and ambiguity, whereas the Media Naturalness Hypothesis would argue mediums that add too much information could be count er-productive and cause cognitive overload at a point (because they are not natural). An exampl e of how a media richness continuum would look relative to media naturalness is shown in the following Figure 1 (Hantula et al., 2011) below. Media richness increases as the continuum move s from the left to the right, but media naturalness decreases as it move s from the center (center is always highest naturalness). 15 Figure 1. Media Richness/Naturalness Continuum Media Compensation Theory It is frequently found that virtual teams utilizing CMCs are able to communicate and collaborate effectively, and have been shown to be able to genera te decisions with equal or better accuracy/quality as FTF teams (Crede & Snieze k, 2003; Lu et al., 2012). In fact, there is evidence that shows managerial preference for me dia lean mediums such as e-mail for a variety of tasks that both Media Richne ss Theory and the Media Naturalness Hypothesis would deem to be a poor fit (Rice & Shook, 1990). In order to address this seemingl y paradoxical finding for why people would choose and efficiently utilize potentially poorly fitting and unnatural CMCs, the Media Compensation Theory (Hantula et al ., 2011) was developed. This theory is an expansion of Kock‚s original Media Naturalness hypothesis and contains seven additional principles beyond the Media Naturalness principle and seeks to explain how humans communicate naturally, and how they adapt to n on-natural CMCs. The eight principles are a) media naturalness, b) learned schema diversity, c) innate schema similarity, d) evolutionary task relevance, e) compensatory adaption, f) medi a humanness, g) cue removal, and h) speech imperative. 16 The media naturalness principle is basically a transplant of the media naturalness hypothesis by Kock, and has already been thoroughly discussed beforehand. At its core, it argues that CMCs that are more similar to face to face communication will be considered more natural, resulting in less effort to interpret me ssages, and increase physiological arousal. The learned schema diversity principle argu es that fiIndividuals learn and acquire communication schemas through interaction with the environment; individual differences are a result of learning.fl The idea of schemas was first introduced by Jean Piaget (Piaget & Cook, 1952) and defined as an organized pattern of thought and behavior that are organized categories of information and the relationships among them. Originally used for characterizing development in children, Piaget argued that as children develop by encountering new information and experiences, they will repeatedly acquire new sc hemas or modify existing ones to organize their understanding of the world. Sim ilarly here, the Media Compensation Theory argues that the manner in which people communicate are also base d off communication schemas that they have acquired through their development, and that ther e will be inter-individual differences in the amount of schemas acquired because individuals will have had different experiences. For example, the communication schemas that the av erage teenager possesses now might include communicating through tweets on Twitter, textin g on phones, or posting comments on Facebook. These teens may possess substant ial knowledge about commonly us ed abbreviations in CMCs such as fibrb,fl fittyl,fl or filol.fl On the other ha nd, the average 80 year old will likely not have had much experience with these t ypes of communication schemas a nd have difficulty communicating through Twitter or Facebook and/or fail to recogn ize what those abbreviations mean. Instead they might still remember the days when you sent regular mail or sent beeper messages rather than e-mail. Thus, individuals should differ in their ability to communicate via CMC depending 17 on how much exposure and experience they have had with different mediums in the past. However, it is important to understand that communication schemas are not necessarily tied down to one medium. For example, the use of em oticons or emojis to express feelings is a communication schema that can be applied to mu ltiple types of CMCs including e-mail, Twitter, instant messaging, Facebook and so on. Another exampl e would be starting letters/e-mails with a fiDear Mr.fl or fiHello,fl and ending with fiSincerely.fl As such, it would be expected that schemas attained through using one form of communication can sometimes be transferrable to other forms of communication. I will connect th is principle later on with CMC competence to show that individuals will be more competent at certain CMCs due to their different experiences that have led to acquisition of different communication schemas. The innate schema similarity principle ar gues that there are universally shared communication schemas that exist between all humans as all humans have evolved to communicate in a similar fashion. At its core, this principle argues that despite cultural, geographical, and linguistic differences between individuals nowadays, there still exist some communication commonalities that can be interpretable by all. Examples include facial expressions such as smiling or frowning. When app lied to the virtual context, it implies that despite people using new communication schemas such as CMCs to communicate, we will still incorporate classic schemas that should be recogn izable by all (e.g. smiling in a video call). This principle has limited applicability to my study be yond that participants will likely be able to recognize smiles and facial expressions of each other in the video conferencing condition. The evolutionary task relevant principle argues that modern tasks that are functionally similar to ancient tasks (e.g. foraging, hunting) will require less effort to complete than tasks that are not functionally similar at all to ancient tasks. This study will only be using one task, a 18 hidden-profile decision-making task, and thus this principle should not have any direct implications on the study as it mainly argues that there may be differences between tasks in how much effort is needed. Furthermore, communica ting and sharing information should be a simple task that has been present since ancient times and should not require significantly more effort to accomplish. The compensatory adaptation principle argues that fiIndividuals using media that suppress elements of face-to-face communication do not accept the obstacles posed by unnatural media passively. Instead they compensate by changing their communication behavior...fl(Hantula et al, 2011, p.347). An example would be using emoticons a nd emojis in e-mails and text messages to express feelings in a medium that would normally prevent that type of information from being exchanged. This principle, when used in conjunction with the learned schema diversity principle, suggests that although humans are evolutionary predisposed to FTF communication, through experience and schema acquisiti on, we can adapt ourselves to the limitations of the CMC by altering the manner in which we co mmunicate in those mediums. Furt hermore, this means that it is not only an issue of what CMC is being used, but also how effectively you are able to adapt to the strengths and limitations of the CMC. This al lows for the possibility that one individual may be more adapted to using a leaner medium thr ough frequent use, such as e-mail, and perform better with it, but fail to communicate effectiv ely with a richer medium such as video- conferencing because he/she has not had an opport unity to acquire experi ence with the medium. This suggests that there may be a powerful individual quality that can influence success in virtual teams. One stream of research th at supports this view is the research associated with the Social information Processing Theory (SIPT) (Walther , 1996). Walther found that while the manner in which we exchange and receive information may be limited by the virtual context (such as lack 19 of contextual/social cues ), given adequate time to communi cate and interact, individuals were able to develop meaningful relationships with one another by adapting the manner in which they exchange and interpret information (Tidwe ll & Walther, 2002; Walther, 1992, 1996; Walther & Burgoon, 1992). The media humanness principle argues that when the computer interface of the CMC incorporates elements that make them filook and feelfl more human, they will also be perceived as more natural. This would suggest that when we design programs or tools that are similar to humans (e.g. giving the name fiSirifl to the arti ficial intelligence (A.I.). on the Iphone) they become more natural and we trea t them as another social actor. In the context of this study, participants will be using Skype which is not designed to emulate humans, and thus this principle should not come into play. However, this prin ciple would suggest that if a computer A.I. confederate was used in a study, it would beneficial if it was de signed to resemble human speech patterns and responses. The Cue Removal principle contends that fimed ia that provide stimuli (or cues) but block people from sensing the information accompanyi ng those cues will require more effort and adaptation than media that do not provide such cues at all.fl (Hantula et al., 20011, p. 349). This principle is important as it argues that more is not always better, as extraneous information can increase cognitive load without necessarily prov iding important information. More importantly, in the Hidden Profile paradigm where being able to maximize unique information is key, CMCs higher in richness may actually use up more cognitive resources for attending to information that is not necessary to solving the task at hand. For example, instances where the video aspect of the video conferencing was not needed all for the task and ended up only as a distraction, especially if the video started chopping up or cutting out. In decision-making tasks, the content of the 20 information is more important than how it™s presented. Being able to perceive a teammate™s eye color or shirt color does not add any incremental value to the task at hand. As such, extraneous information channels in the CMC may actually detract from the efficiency of the team™s information sharing by increasing cognitive load and forcing team members to attend to irrelevant information. Thus it would make sense that in decision-making tasks, the addition of audio and/or video information may detract from the experien ce rather than add to it. The final principle of the Media Compensation Theory is the speech imperative principle, which argues a medium™s ability to convey speec h is significantly more important than the medium™s ability to convey faci al expressions or body language. This principle was derived from evolutionary literature that suggests more costly adaptations (evolutionary changes in our body) are also more important for the underlying tasks th ey support. As evolving the larynx to allow us the ability to speak also significantly increased our susceptibility to choking, it is thought that this ficostfl we exchanged for the ability to spea k represents how important speech is to us. This suggests individuals may be more accustomed to and prefer CMCs that include speech as a component, although the authors failed to make any claims on this principle™s effect on performance and communication. Unfortunately, the Media Compensation Theory has never been tested empirically, but some of its principles are useful to this study by bringing to light that individuals can vary in their understanding and familiarity with differe nt communication schemas (learned schema diversity), that individuals are able to overcome limitations of CMCs by adapting their behavior (compensatory adaptation) and that sometimes CMCs that provide unnecessary cues may actually hinder performance by causing more effortful processing (cue removal). 21 CMC Competence and Individual Differences One issue with all of the central CMC theories introduced earlier (Media Richness Theory, Media Naturalness Theory, and Media Compensation Theory) are that they largely ignore the role of the individuals utilizing the CMCs. Media richness theory generally argues that the richer the medium, the more it would benefit all users as it should provide more types of information to be communicated and more qui ckly. Thus, everyone should prefer richer mediums. Media naturalness hypothesis would argue th at the more natural th e better, and that all individuals will prefer CMCs that are more sim ilar to face to face communi cation because that is what we have evolved to prefer. Media compensation theory does provide some acknowledgement of differences betw een individuals in their learned schema diversity principle, acknowledging that some people may be more skilled in using certain CMCs as a result of acquiring different communication schemas, but their general premise is that people will need to use more effort to adapt to le ss natural mediums and should generally be more efficient using more natural CMCs. These theories have largely ignored the issue of variability between individuals in both their preference for CMCs and competence in using CMCs. However, there is a body of research that has shown that individual differences (e.g. personality, self-esteem, self-efficacy) can have significant effects on how people approach and utilize CMCs. For example, Joinson (2004) noted that low self-esteem internet users preferred e-mail communications much more than high self- esteem users, and that increased chances of rejection in a scenar io led to much higher preferences for virtual communicat ion than face to face communication. This shows that people may have individual preferences to certain CMCs that have little to do with whether the CMC is fit for it or how similar it is to face to face communication 22 CMC competence is a term coined by Spitzberg (2006) and represents an individual™s competence and effectiveness in using CMCs. The main facets of CMC competence are CMC motivation, CMC knowledge and CMC skills thought to correspond to their parallels in FTF communication (e.g. composure, attentiveness, coordination). The origins of this conceptualization stem from work done by Ring and colleagues using a dramaturgical perspective in conceptualizing an actor™s performance (Ring, Braginsky, & Braginsky, 1966; Ring, Braginsky, Levine, & Braginsky, 1967). They argue that an actor needs to be motivated to give a good performance, but motivation by itself is insufficient if the actor does not have the script for how the play should go (knowledge). However, even possessing both the motivation and knowledge is insufficient if the actor lacks the skills to translate that motivation and knowledge into competent action. Using this broa d conceptual model of competence as being a function of the motivation, knowledge and skills of the individual, Spitzberg translated it to the CMC context to develop his conceptualization of CMC competence. CMC motivation is meant to capture the range of constructs that would endear a person to look favorably upon CMC such as willingnes s to adopt new communication technologies, satisfaction, gratifications, and positive attitudes toward such technologies. Individuals with high CMC motivation are characterized by confidence and comfort in using CMCs whereas negative motivation towards CMC use is characterized by anxiety, apprehension, apathy or even disinterest towards using CMCs. Spitzberg formally defined CMC motivation as fithe ratio of approach to avoidance attitudes, beliefs, and va lues in a given CMC context.fl (p. 640). As such, it would suggest that individuals may differ in th eir willingness to use CMCs, independent of the richness or naturalness of the CMC. 23 CMC knowledge is formally defined fias th e cognitive comprehension of content and procedural processes involving in conducting appropriate and effective interaction in the computer-mediated context.fl (p. 641). Thus , an individual possessing a high amount of knowledge regarding CMCs would be expected to be able to effectively adapt their communication style to the CMC context, and also possess the procedural knowledge needed for utilizing different CMCs (e.g. understanding the role of emoticons in messages, knowing that fitweetsfl have a 140 character limit). While all knowledge and skill acquisition must be acquired through some type of learning, it is relatively rare for individuals to learn to use CMCs through formal training or lecture. One manner in which individuals may come to acquire CMC knowledge is through experiences and repeated interactions with CMC. As such, one manner in which we can view CMC knowledge is by the breadth and depth of communication schemas attained through past experiences (i.e. learned schema dive rsity principle). Spitzberg defined skills as fithe repeatable, goal-oriented behavioral tactics and routines that people employ in the service of their mo tivation and knowledge.fl (p. 638). In a previous study, Spitzberg and Cupach (2002) identified over 100 distinct skills in the communication competence literature, but ultimately were able to refine them into 4 central skill clusters: attentiveness (i.e., displaying concern for, interest in, and attention to the other person or persons in the interaction), composure (i.e., displaying assertiveness, confidence, being in control), coordination (i.e., displaying deft management of timing, initiation and closure of conversations, topic management), and expressiveness (i.e., di splaying vividness and animation in verbal and nonverbal expression). It is thought that these skills reflect basic principles of effective communication, and thus an individual high in CMC competence s hould be able to adapt these skills into the CMC context. Several studies have shown evidence that these skills exist in the 24 CMC context and are beneficial towards effective communication. For example, Bunz and Campbell (2004) found that participants were more likely to reply to e-mails politely when there were politeness cues embedded within the e-mail, suggesting that showing concern/interest in others is likely to be reciprocated in a CMC context. Castella, Abad, Alonso, and Silla (2000) found that familiar individuals communicating vi rtually adapted their messages to be more informal, including emoticons and humor to be tter express themselves to their friends. All in all, it is expected that an individual highly competent in using CMCs must not only possess the knowledge to effectively communicate using CMCs, they must al so possess the skills to apply that knowledge into the CMC context. Additionally, an individual competent in CMC must also have the motivation to use CMCs, ot herwise they will be unable to leverage their knowledge and skills. For example, an indi vidual may possess high knowledge regarding Formula 1 racing, extensively studying videos about how to properly corner and have read books on how to shift gears efficiently. However, just possessing the knowledge is insufficient for competency. If the individual lack s the skills to actually transfer that knowledge (e.g. having the motor coordination to shift gears in time, havi ng the hand-eye coordination necessary to properly corner) then all that knowledge would be useless. Likewise, the opposite is also true. Possessing high motor skills and being able to shift gear s smoothly, does not make one a competent F1 driver by default if one lacks the knowledge on wh en it is appropriate to apply these skills (e.g. randomly shifting gears when it is an inappropriate time to so ma y damage the vehicle). Lastly, both knowledge and skills come to naught if the pe rson lacks the motivation to put them to use. A racer that has suffered an accident may still reta in the knowledge and skills necessary to race at a top level, but has lost the motivation and confidence to race again and cannot be called a competent F1 driver any longer. 25 As decision-making teams typically require i nput from all members, it would be expected that the average CMC competence of the team , or team CMC competence, would be one indication of whether a virtual team would be e xpected to successfully pool unique information. I argue that CMC competence can be aggregated into a compositional team variable and can be used for meaningful comparisons between teams. Specifically, it would be expected that teams with a lower team CMC competence score would perform more poorly than teams with higher team CMC competence because its members will be less motivated, possess less knowledge and less skills than members in the high CMC competence team. While CMC competence is predicted to be one of the two main predictors of unique information pooling in this model, an alternativ e perspective might argue that CMC competence is actually just self-efficacy w ith a different label. Self-efficacy was first coined by Bandura, and is commonly defined as one™s be lief in one™s own ability to complete tasks and reach goals (Bandura, 1977, 1982). The self-efficacy literature has shown that one™s perception of self- efficacy is a strong predictor for behavior and numerous self-efficacy measures have been developed ranging from exercise to internet usage, to breast feeding (Eastin & LaRose, 2000; Kingston, Cindy-Lee, & Sword, 2007; Marcus, Selby, Niaura, & Rossi, 1992). It is highly likely that the two will be highly corre lated as CMC competence, is overall, a measure of perceived ability to successfully communicate using a CMC. However, the difference lies in that a truly CMC competent individual must possess all three components of motivation, knowledge and skills to effectively utilize CMC, whereas a highly self-efficacious individual only needs to believe they can do well, regardless of thei r actual skills and knowledge. As such, CMC competence is a more specific construct that s hould provide explanations for when outcomes do 26 not match perceptions of ability (e.g. may have the motivation and confidence to do well, but lack the knowledge and skills to back up that confidence). Model & Hypotheses Figure 2. Model of Team Decision-Making Presented above is my model for the propos ed relationships between media richness, unique information pooling, decision quality a nd CMC competence. I predict that media richness is negatively related to team unique information pooling, and that generally lower media richness CMCs will out pool higher media richness CMCs. I predict that team CMC competence is positively related to team unique information, and that teams with higher CMC competence will pool more unique information than teams with lower CMC competence. Lastly Media Richness Team Decision Quality Team Unique Information Pooling H2Mediation H3Team CMC Competence Mediation H5 27 I predict that the unique information pooling mediates the relationship between both media richness and team CMC competence to team decision quality. Contrary to what has been suggested by Media Richness Theory, I argue that more is not always better. Specifically, lower media richness teams should be more task focused via Cues- Filtered Out Theory leading to mo re effective utilization of time, receive less cognitive burden via the Cue Removal Principle from Media Comp ensation Theory, and also promote more equal participation from team members via Status Generalization; allo wing for more unique information to be shared and pooled. For ex ample, in the high media richness condition, individuals may be reluctant to speak out of turn when a higher status member is speaking, allowing for one person to dominate the conve rsation. Conversely, in the instant messaging condition, people can type and input text without interrupting one another as the transmitting of one text does not prevent the transmission of another™s text. Furthermore, instant messaging requires very little bandwidth and there are less issues with filagfl such as screen blurring/freezing and/or audio cutting off intermittently, wh ich in video conference calls, may frustrate user collaboration and hamper the sharing/understanding information being discussed (i.e. cue removal principle). While some res earchers have previously found no effects for communication technology on unique information pooling and decision quality (Lu et al., 2012), I contend that previous research was handicapped by the time period in which they were conducted as virtual communication was still in its relative infancy and participant familiarity with the technology relatively low. Additionally, previous st udies have largely ignored the differences between CMCs and inappropriately grouped together multiple CMCs into one category. I contend that there are differences in media richness, the speed of which information can be communicated and the type of informati on which can be communicated, within the broad 28 family of CMCs. Furthermore, some individual s may be more competent with CMCs as opposed to others because of differences in motiva tion, knowledge and skill regarding CMCs that resulted from different experiences and communication schema s acquired. For the purpose of this study I will be comparing a low media ri chness CMC, instant messaging, against a high media richness medium, video conferencing, to showcase the differences that are inherent between different types of CMCs. Thus I propose the following: H1: Media richness of CMC is negatively related to team unique information pooling such that teams using CMCs with high richness w ill pool less unique information than teams using CMCs with low media richness. Unique information pooling is one of the most significant predictors of decision accuracy within the information shari ng literature (Lu et al., 2012; Mesmer-Magnus & DeChurch, 2009). Teams that are able to discover and share more unique pieces of information are more likely to correctly solve the hidden profile. As such, I would expect my findings to fall in line with the rest of the field in this regard, that there is a positive relationship between unique team sharing and team decision quality (whether or not the team makes the correct choice in the hidden profile task). The level of analysis for this is necessari ly at the team level because decision quality will be assessed by the team™s decision. H2: Team unique information pooling posi tively predicts team decision quality. In virtual teams, media richness is exp ected to influence team decision quality by changing the manner in which team members communicate, limiting the type and speed of information able to be conveyed. The medium by itself does not lead directly to changes in 29 decision quality; rather it should aff ect the degree to which the team is able to successfully pool unique information which may then impact deci sion quality. As such I propose the following: H3: Team unique information pooling mediates the relationship between media richness of CMC and team decision quality. Next I propose that team CMC competence also positively predicts unique information pooling in conjunction with media richness. This predicted effect is supported by the learned schema diversity principle and the compensato ry adaptation principle of Media Compensation Theory(Hantula et al., 2011); the Social Information Proc ess Theory(Walther & Burgoon, 1992); and also broadly by motivation and self-e fficacy research. The introduction of this relationship is meant to clarify the conflicting findings of virtual team performance by looking at beyond just the capabilities of the medium, but also the inter-indi vidual differences in skills, knowledge, and motivation of the teams using the CMCs. Just as one would not expect a novice violinist to perform better simply by handing th em a Stradivarius, one would not expect an individual to be able to effectively utilize a medium he/she has no knowledge of or unconfident in using. A virtual team that is motivated, know ledgeable and skillful in using CMCs is much more likely to be able to leverage the capabilities of the CMC they are utilizing. Thus, on average, a team with higher CMC competence should be more likely to correctly solve the hidden profile task than a team with a lowe r CMC competence score. And thus I propose: H4: Team CMC competence positively p redicts unique information pooling. 30 Similar to H3, I predict that team CMC co mpetence influences te am decision quality by changing the amount of unique information the team is able to pool; teams filled with motivated, skillful and knowledgeable members are much more likely to overcome the shared information bias and thus I propose my final hypothesis: H5: Team unique information pooling medi ates the relationship between team CMC competence and team decision quality. 31 Method Participants Participants were undergraduate college students recruited from a large Midwestern university through the psychology department™s experimental rese arch website. Participants were given credits for participation that were either required for psyc hology courses or could be used as extra credit for certain courses. There were minimal restricti ons for participation in the study besides the requirement to be able to speak and read English fluently. After accounting for teams with missing da ta and mechanical failures (1 team was removed because the session could not be comple ted due to mechanical failure, 1 was removed because the chat log was lost), the final sample size consisted of 234 participants spread across 78 teams (38 instant messaging, 40 video conferen cing). The average age of participants was 19.33 (SD = 1.57). Participants were predominantly female (184 females to 50 males) and Caucasian (168 Caucasian/234 total). Measures The primary constructs of interest in my model are CMC competence, unique information pooling and team decision quality. However, I also measured virtual decision- making self-efficacy to show that CMC competen ce is a distinct construct from self-efficacy. Additionally, I also included a m easure of engagement in my st udy to address concerns of an alternative model of decision-making quality where it is th ought that participant engagement would predict decision quality rather than unique information poo ling. In my study I believe that decision quality is predicted by media richness and team CMC competence, mediated through unique information pooling, rather than engagement in the task. 32 CMC Competence: To measure CMC Competence, I us ed a sub-scale of the IMPACCT measure (Spitzberg, 2011) that was develope d as an advancement of the initial CMC Competence Measure proposed in Spitzberg (2 006). The IMPACCT meas ure was originally developed to survey student communication and cr itical thinking skills a nd has been empirically tested to be reliable ( a=.96) using a 1,880 student sample. This sub-scale was developed specifically to measure individual competence re garding the appropriate and effective use of CMC technologies for communication. I updated the measure slightly to account for changes in technology (e.g. providing updated examples of common CMCs as referents and removing some attention check items that were unneeded in our study, resulting in a 25-item measure (see Appendix A for items). Participants indicated th e extent to which each statement on the measure accurately described them using a 7-point scale ranging from 1 (Not at all true of me) to 7 (Very true of me). Individual CMC competence was com puted by calculating the average of each individual™s score and team CMC competence was computed by averaging the CMC competence of each individual within the team (compositional va riable). The internal consistency reliability of the measure is .92. During the analysis phase I conducted an exploratory factor analysis (full results shown in Appendix B) with the CMC competence scale and the virtual decision-making self-efficacy measure to see if the constructs clearly mapped onto separate factors. In this process I also identified 7 items that showed poor loadings a nd that seemed to map onto a different construct (all the CMC adaptability skill items mapped se parate from the other CMC competence items, and were dropped). The final measure containe d 18-items and had an internal consistency reliability of .92 (see Appendix C for final item set). 33 It should be noted that this CMC competence measure is targeted at the broad level of competence towards all CMCs, rather than to wards a specific CMC (i.e. competence toward video conferencing). It was not expected that there would be significant differences in competence toward either CMC in this study as they are both components of the same program (i.e. Skype) and commonly used in conjunction. Virtual Decision Making Self-Efficacy: To measure virtual decision making self- efficacy, I adapted a 12-item measure created by (Howard, 2014) that was originally created to measure an individual™s self-efficacy when using a computer. The measure was adapted such that each item referred to CMCs instead of computer s where appropriate, and the items were framed to a virtual decision making cont ext (i.e. fiWhen I am in a virt ual decision-making team using CMC, I am confident that–fl). Participants indi cated their agreement with each statement using a 5-point scale ranging from 1 ( Strongly Disagree) to 5 (Strongly Agree ). An individual™s score on the measure was calculated by computing the average of their responses across the 12 items, and the team-level virtual decision making self-efficacy was the average of each team members™ scores (see Appendix A for items). After dropping 1 item that loaded poorly from the exploratory factor analysis, the final measur e contained 11 items and had an in ternal consistency reliability of .86 (see Appendix C for final item set). Engagement: To measure participant engagement in the task, I adapted a portion of the User Engagement Scale (UES) (O'Brien & To ms, 2013). In their study they found that the original 28-item measure loaded cleanly on 4 di fferent underlying factors, one of which was theorized to represent user engagement and expe rience (the other factor s were concerned with usability/utility, aesthetics and focused attentio n). I slightly adapted the items from their proposed sub-scale to create a 9-item measur e on engagement (see Appendix A for items). 34 Participants indicated their agreement with each statement using a 5-point scale ranging from 1 (Strongly Disagree) to 5 (Strongly Agree). An individual™s engagement score was calculated by computing the average of their responses across the 9-items and team-level engagement was computed by averaging each team members™ score. The internal consistency reliability of the measure is .87. Unique Information Pooling: Unique information pooling was computed via text and audio coding of the number of pieces of unique information mentioned during the discussion phase of the experiment. There were a total of 31 pieces of unique information (12 regarding Company A, 19 regarding Company B) that could be mentioned by each team. All initial coding was done by two undergraduate research assistants who were trained on how to code by the primary investigator. They were trained togeth er for 4 hours using pilot study data and initial inter-rater reliability based on the pilot study was .91 which was deemed sufficient for data collection for the main study to commence. Inter -rater reliability for the main study was .61 as calculated using Cohen™s kappa, which is considered moderate agreement. Disagreements were resolving in subsequent re-coding sessions through rater consensus. Decision Quality: Decision quality was operationalized as whether the team made the correct decision in the experiment (coded dichotomous ly as either 0 = Incorrect or 1 = Correct). Apparatus/Materials The hidden-profile task was chosen because th e group outcomes of task types that require persuasion or consensus are thought to be mo re highly influenced by communication mediums (Straus & McGrath, 1994). For this study the ACME Hidden Profile Task created by Dr. Poppy L. McLeod was used. It is a standard hidden profile task in which the participants are given the 35 role of a top management team tasked with deci ding on a firm to acquire out of three potential firms (see Appendix D). Each information packet contained information about three companies that were potential acquisition targets as well as the criteria on which they should evaluate each company on. The criteria were 1) which company had the most promising future and would give the highest return over the long run, 2) the probab ility of you getting a return on your investment and whether these projections are accurate, 3) potential growth of the market in the future, 4) self-sufficient management team that does not require micro-managing, and 5) overall general strategy and business policies. Each company pr ofile contained information about their name, products, location, size, age, financials (e.g. investment return rate , sales growth rate), strategic assets (e.g. management team, market share, product) and labor (e.g. labor costs, training, turnover). The instructions were adapted slight ly to allow for single-choice rather than rank-ordering. As with other typical, solvable hidden profile tasks, there is one company that is the fibestfl choice (in our scenario, Company A), but that choice is not immediately obvious. Each member received a different packet of inform ation that contained some common information regarding the companies, but also some uniquely held information about the companies. The full information set contained 95 items 13 positive items, 7 neutral items, and 13 negative items on Company A (Net Score = 0); 11 positive items, 8 neutral items, and 19 negative items on Company B (Net Score = -8); 5 positive items, 7 neutral item s and 11 negative items on Company C (Net Score = -6). There were three sepa rate packets used for this task, each differing in their information composition. A breakdown of how the information was distributed is shown below in Table 1. 36 Table 1. Distribution of Inform ation in Hidden Profile Task InformationDistributionPacketA(Full) PacketBPacketCCommonCompanyA21(2+,6=,13)21(2+,6=,13)21(2+,6=,13)CommonCompanyB19(11+,7=,1)19(11+,7=,1)19(11+,7=,1)CommonCompanyC23(5+,7=,11)23(5+,7=,11)23(5+,7=,11)UniqueCompanyA(Aonly)13(11+,1=,1)UniqueCompanyBSet1(Aonly)9(1=,8)UniqueCompanyBSet2(A&Bonly) 5(5)5(5)UniqueCompanyBSet3(A&Conly) 5(5)5(5)TotalItems956868NetScore CompanyA01111NetScore CompanyB855NetScore CompanyC644Note: (+) denotes positive, (=) denotes neutral, (-) denotes negative Packet A contained the full information set, and also served as a manipulation check for the solvability of the hidden profile. Provided the full information set, it was expected that the participants would be able to discern the objectively best Company (Company A) for investment (Net score of 0 vs. -8 and -6). However, the common information was biased towards selection of Company B, and participants with incomple te information sets (Packets B and C) should prefer Company B (Net score of 5 vs. -11 and -4). This sets up a situation where a majority of the team members should prefer selec tion of the Company B, meaning that it would necessitate the use of unique information to overcome the sh ared information bias for Company B. Participants were required to use a windows-based personal comput er, a headset and a webcam for the task. The study used the Skype communication program, with certain functions disabled, as the CMC medium that teams comm unicated through. All measures were presented virtually through online surveys hosted on the Qualtrics survey website and the information packets were stored in Mi crosoft Word documents. 37 Procedure There were two conditions in my research st udy to represent low vs. high media richness: an instant messaging condition (low media rich ness) and a video conferencing condition (high media richness). In both conditions the participan ts used the Skype program to communicate, but were limited to certain functions within th e program. In the instant messaging condition participants were only allowed to communicate via a 3-way instant messaging chat room and were not allowed to use the program™s audio or video capabilities. In this condition messages were not transmitted until the message author pressed the enter key on the keyboard. In the Video-Conferencing condition participants were provided a microphone and a webcam and were forced to communicate through a 3-way video conferencing call. They were only allowed to communicate through audio and not allowed to type to one another. Upon arrival at the research lab, participants were greeted by the experimenter and told that they would be participating in a study of virtual team decisi on-making and that they will be asked to work together to solve a team decision-making task. They were assigned a participant code for confidentiality purposes, and then given an informed consent form detailing what they could expect from the study, requesting their perm ission to be recorded, and informing them that they would have different pieces of information in their information sets later on in the task. Participants were then grouped into ad-hoc teams of 3 participants and randomly assigned into a condition. Each team had 3 members that would participate in the main task, while the 4 th participant was given an alternative task and se rved as an additional manipulation check for the solvability of the hidden profile task. The 4th participant was given the full information set and used as a reference to assess how difficult the task would be to solve if all information was readily available. 38 After the consent forms were signed, participants in the team task were randomly assigned to a station, each of which contained a different informa tion packet (A, B or C) while the individual participant was told to await furt her instructions. Each station faced a different side of the wall so that no participant would be in view of each other throughout the rest of the experimental session. Additionally, each member was given a noise-canceling headphone to wear for the discussion phase so that cues out side of the CMC they were using would be minimized. Participants in the team task were given approximately 5 minutes to complete an online survey containing the CMC competen ce, virtual decision making self-efficacy and personality measures. After everyon e was finished with the online survey, they were instructed to maximize the word document containing th e information set and given 20 minutes to memorize as much information as they could about each company and make a decision for which company they prefer prior to team discussion. At this time, the 4 th participant was given a hard copy of the full information data set to study and likewise given 20 minutes to study the information and come to a decision. During the memorization phase they were instructed to study independently without taking notes, relying only on their memorization. They were not allowed to communicate with each other during th is phase and were told to pay particular attention to what criteria they should be basing their decision on and that they should do their best to remember specific facts and/or num bers to use in their discussion phase. At the conclusion of the 20 minute study phase, participants were required to close the word document containing the information set and write down what their pre-discussion preference was. This was used as a manipulation check that the common information biased the decision towards Company B (for packets B and C) and to assess the solvability of the hidden profile (packet A and individual participant). Afterwards, participants in the team task were 39 instructed to switch over to Skype and given 25 minutes to discuss their preferences and come to a consensus about which company they would ac quire. Participants in the instant messaging condition were only allowed to type to each other in the 3-way chat, while participants in the video conferencing condition were only allowed to communicate verbally (with video) to each other. After collecting the individual participants ™ choice, they were allowed to leave as they would not participate in the discussion phase. During the discussion phase the participants were instructed that they must utilize the full 25 minutes of discussion time and were reminded periodically that they should do th eir best to use statistics or specific points to argue their choices and that they should try to recall the criteria used for evaluating the companies. At the end of the 25 minutes, participants were asked if they had reached a consensus. If they had, they were told to quit Skype and then finish another online survey with the engagement measure and demographics questionnaire. It should be noted he re that every team reached consensus, but if they had not reached consensus the data would have been retained for alternative analyses. After completing the online survey, participants were asked to come together to fill out a form where they were instructed to copy their pre-discussion choice from the slip they wrote earlier, their choice as a team, and also a checklis t (see Appendix E) where they were asked to indicate which pieces of unique information they had mentioned. Originally the idea had been to use the checklist as a self-report measure of unique information pooled, but subsequent analyses revealed that the checklist showed very poor reliability with rater coding in the pilot study and thus unique information pooling was ultimat ely assessed only through rater coding. 40 Results Means, standard deviations, and correlations for individual level variables are presented in Table 2. The individual mean of CMC competence was 5.18 ( SD = .72), suggesting that most participants considered themselves to be comp etent in using CMCs (a rating of 5 on the CMC competence scale represented fiSomew hat true of mefl) and the low SD suggests that very few people considered themselves to be incompetent at CMCs. Virtual decision-making self-efficacy (M = 3.64, SD = .49), likewise suggested that most participants consid ered themselves to be efficacious in virtual contexts without too many participants straying far from the mean. Engagement ( M = 3.82, SD = .58) suggested that most participants were moderately interested and engaged in the task. CMC competence was significantly correlated with age ( r(231) = -.23, p < .01) and virtual decision-making self-efficacy ( r(232) = .58, p < .01). These results suggested that younger participants were more likely to rate themselves as being competent in CMCs and that as CMC competence increased, so did one™s perceptions of virtual decision-making self- efficacy. Engagement was found to be positiv ely related to CMC competence ( r(232) =.28, p < .01) and virtual decision-making self-efficacy ( r(232) = .24, p < .01), suggesting that participants who were more competent and effica cious were the ones more engaged in the task. Participants were not told whether their team ma de the correct decision prior to the collection of the engagement ratings, so it was unlikely that en gagement ratings were influenced by feedback on actual performan ce on the task. 41 Table 2. Means, Standard Deviations, and Correlations among Individu al-Level Variables Variables M SD 1 2 3 4 5 1. Age 19.33 1.67 ___ 2. Gender 0.21 0.41 .14 ** ___ 3. CMC Competence 5.18 0.72 -.23 ** -.20 ** ___ 4. Virtual Decision-Making Self-Efficacy 3.64 0.49 -.09 ** .02 ** .58 ** ___ 5. Engagement 3.82 0.58 -.09 ** .10 ** .28 ** .24 ** ___ Note: Overall N = 234.Correlation coefficients marked with an asterisk were statis tically significant (* = p < .05, ** = p < .01) Gender was coded as 0=Female, 1= Male. CMC-Competence was rated on a 1-7 scale, Virtual Decision-Making Self-Efficacy and Engagement on a 1-5 scale. 42 To further clarify the findings, an exploratory linear multiple regression was run with media richness (condition), CMC competence, virtual decision-making and self-efficacy predicting engagement. While not a part of my initial hypotheses, it was thought that this exploratory analysis would provide useful insight into how my constructs might impact engagement, an important construct in the performance literature. The results are presented in Table 3. The analyses revealed that the overall model was significant and accounted for 13% of the variance in engagement, R2 = .13, F(5, 228) = 6.92, p < .01. However, only CMC competence ( = .17, p < .05) and media richness ( = -.17, p < .01) emerged as significant predictors. The results suggest that although enga gement was correlated with multiple variables, it was mainly media richness and CMC competence accounting for the variance between individuals in engagement. Table 3. Regression Analysis Predic ting Engagement (Individual-level) Note: N = 234. Regression coefficients marked with an asterisk were statistically significant (* = p < .05, ** = p < .01) Variables Engagement Predictors B SE(B) Media Richness -.20 ** .07 ** -.17** CMC Competence .14 ** .06 ** .17* * Virtual Decision-Making Self-Efficacy .12 ** .09 ** .10 ** ** R2** ** .13 ** 43 The results of the manipulation check reveal ed that the task showed relatively high solvability, with 64% of participants in the i ndividual task solving the task correctly and 78% percent of participants given Packet A (full-information set) selecting Company A (the correct answer) as their pre-discussion choice. Furthermor e, the manipulation check also revealed that 80% of participants receiving Packet B and 82% of participants receiving Packet C selected Company B as their pre-discussi on choice, providing evidence that the manipulation for biasing the common information towards Company B succeeded. Means, standard deviations, and correlations are presented for team level variables in Table 4. Mirroring the individual-level relationships, CMC competence, virtual decision-making self-efficacy and engagement were significantly rela ted to one another. Once again, this suggests teams that were competent in CMCs also indica ted higher virtual decision-making self-efficacy and engagement in the task. The mean of unique information pooling was 9.65 ( SD = 4.18), which was quite concerning as there were a tota l of 48 pieces of unique information available. This means that teams generally pooled less than a quarter of the unique information available to them. This provides some context as to why team s fared so poorly in making the correct decision (M = .33, SD = .47). With a 33% accuracy rate and 3 possible choices, the teams™ decision accuracy was effectively at chance. A significant correlation emerged between me dia richness and team-level engagement (r(76) = -.30, p < .01), and unique information pooling ( r(76) = .26, p < .05). As media richness was operationalized via condition (low media richness vs. high-media richness condition) and coded as a binary variable, these correlations are point-biserial correl ations. The significant negative correlation between media richness and engagement falls in line with the regression analysis conducted earlier to parse out the main predictors of engagement, showing that teams 44 were more likely to have lower engagement in the video conferencing condition. The significant positive correlation between media richness and unique information pooling, however, was unexpected as Hypothesis 1 had predicted a negative relationship between media richness and unique information pooling (i.e. teams in the instant messaging condition will pool more unique information that teams in the video conferenci ng condition). Results indicated that the video conferencing condition pooled more unique information ( M = 10.73, SD = 4.14) than the instant messaging condition ( M = 8.53, SD = 3.98). As such, Hypothesis 1 was not supported. Team CMC competence did not significantly correlate with media richness ( r(76) = .00, p > .05), unique information pooling ( r(76) = -.02, p > .05) or decision quality ( r(76) = -.17, p > .05). The lack of a significant relationship between media ri chness and CMC competence was reassuring as this meant there we ren™t any significant differences in competence between the two conditions. Not finding a significant correlation between team CMC competence and unique information pooling did suggest that there was no main effect of team CMC competence on unique information pooling, and as such Hypothesis 4 was not supported. 45 Table 4. Means, Standard Deviations, and Correlations among Team-Level Variables Note: Correlation coefficients marked with an asterisk were statistically significant (* = p < .05, ** = p < .01) CMC Competence was rated on a 1-7 scale. Virtual Decision-Making Self-Efficacy, and Engagement on a 1-5 scale. Media richness was coded dichotomously as 0 = Instant Messaging, 1 = Video conferencing. Variables M SD 1 2 3 4 5 6 1. Media Richness (Condition) ___ ___ ___ 2. CMC Competence 5.18 0.38 .00 ** ___ 3. Virtual Decision-Making Self-Efficacy 3.64 0.26 -.23 ** .46 ** ___ 4. Engagement 3.82 0.36 -.30 ** .34 ** .25 ** ___ 5. Unique Information Pooling 9.65 4.18 .26 ** -.02 ** .07 ** -.03 ** ___ 6. Decision Quality 0.33 0.47 .04 ** -.17 ** -.12 ** -.03 ** -.03 ** ___ 46 To test for Hypothesis 2, which stated that team unique information pooling positively predicts decision quality, I conducted a logistic regression where decision quality was coded as a binary variable with 0 = incorrect, and 1 = correct. The result s are presented in Table 5. Interpretation of the model chi-Square statistic revealed that the ove rall model with unique information pooling as the main predictor, was not significant p = .77. Additionally, unique information pooled was not a significan t predictor of decision quality (p = .77) with an odds ratio (e) of .98, meaning Hypothesis 2 was not supported. The e represents the change in probability of the team reaching the correct decision for each one unit change in unique information pooling. An e of .98, if it had been significant, would have suggested that for each additional piece of unique information pooled, the team was 2% more likely to make the incorrect decision. Table 5. Logistic Regression Analysis of Unique Information Pooling on Decision Quality Predictors SE Wald™s X2 df p e Constant .53 * .61 * .77 * 1** .38 * .59 * Unique Information Pooling -.02* .06 * .08 * 1** .77 * .98 * X2 df p 2ll Cox & Snell Nagelkerke Overall Model .08 * 1** .77 * 99.21 .00 * .00 * Note: N = 78. Regression coefficients marked with an asterisk were statistically significant (* = p < .05) Hypothesis 3 predicted that team unique information pooling would mediate the relationship between media ric hness and decision quality. However, because there was no significant relationship between unique informati on pooling and decision quality (H2), there was 47 no possibility of a mediation effect occurring and thus Hypothesis 3 was also unsupported. Likewise, Hypothesis 5 predicted that team unique informa tion pooling would mediate the relationship between team CMC competence and decision quality. However, because there was no significant relationship between neither unique information pooling and decision quality (H2) or between team CMC competence and unique information pooling, Hypothesis 5 was also unsupported. As all of my hypotheses were unsupported, even those that have been reliably replicated in literature (i.e. H2), I re-examined my data to see if perhaps I had retained too many teams. I decided to rerun my analyses using a more stri ngent cut-off to remove teams that were very unlikely to have been actively participating. Unfortunately, I also could not remove too many teams as that might too severely limit my power to detect effects. Ultimately, I settled on removing teams that pooled less than 5 pieces of unique information pooled, ultimately removing 6 teams that were all from the instant messaging condition. Comparing tables 6 and 7 to ta bles 2 and 3 we see that ther e was only one major change, the positive relationship between media richness and unique information pooling disappeared. Otherwise, decision quality remained at 33% a nd the mean unique information pooled per team went up to 10.19, but otherwise did not significantly change any other relationships. Unfortunately, this new set of analyses also failed to support my original hypotheses. The correlation between media richness and unique information pooled was not significant ( r(70) = - .07, p > .05) and thus Hypothesis 1 was not s upported. The correlation between team CMC competence and unique information pool ed was also not significant (r(70) = .15, p > .05) and thus Hypothesis 4 was not supported. The binary logistic regression results shown in table 8 48 reveal that unique information pooling still failed to predict decision quality p = .76 and thus Hypothesis 2 remained unsupported. In the following discussion I will be mainly referring to the initial findings as these subse quent analyses did not reveal anything new with the exception of the relationship between media richness and uni que information pooling which disappeared due to the removal of the 6 low performing teams th at were all in the in stant messaging condition. 49 Table 6. Means, Standard Deviations, and Correlations among Individual-L evel Variables (Alt.) Note: Overall N = 218.Correlation coefficients marked with an asterisk were statis tically significant (* = p < .05, ** = p < .01) Gender was coded as 0=Female, 1= Male. CMC-Competence was rated on a 1-7 scale, Virtual Decision-Making Self-Efficacy and Engagement on a 1-5 scale. Variables M SD 1 2 3 4 5 6 1. Media Richness (Condition) ___ ___ ___ 2. CMC Competence 19.33 1.71 -.02 ** ___ 3. Self-Efficacy 0.20 0.40 -.01 ** .13 ** ___ 4. Engagement 5.20 0.83 .00 ** -.19 ** -.16 ** ___ 5. Unique Information Pooling 3.65 0.53 -.12 ** -.08 ** .04 ** .60 ** ___ 6. Decision Quality 3.82 0.58 -.18 ** -.09 ** .11 ** .30 ** .22 ** ___ 50 Table 7. Means, Standard Deviations, and Correlations among Team-Level Variables (Alt.) Note: Correlation coefficients marked with an asterisk were statistically significant (* = p < .05, ** = p < .01) CMC Competence was rate d on a 1-7 scale. Virtual Decision-Making Self-Efficacy, and Engagement 1-5 scale. Media richness was coded dichotom ously as 0 = Instant Messaging, 1 = Video conferencing. Variables M SD 1 2 3 4 5 6 1. Media Richness (Condition) ___ ___ ___ 2. CMC Competence 5.20 0.44 -.01 ** ___ 3. Self-Efficacy 3.63 0.27 -.01 ** .47 ** ___ 4. Engagement 3.82 0.38 -.31 ** .37 ** .26 ** ___ 5. Unique Information Pooling 10.19 3.88 .15 ** -.07 ** .12 ** -.03 ** ___ 6. Decision Quality 0.33 0.47 .04 ** -.19 ** -.14 ** -.01 ** -.04 ** ___ 51 Table 8. Logistic Regression Analysis of Unique Information Pooling on Decision Quality (Alt.) Predictors SE Wald™s X 2 df p e Constant -.49 * .71 * .48 * 1** .48 * .61 * Unique Information Pooling -.02 * .07 * .09 * 1** .76 * .98 * X2 df p 2ll Cox & Snell Nagelkerke Overall Model .09 * 1** .76 * 91.57 .00 * .00 * Note: N = 72. Regression coefficients marked with an asterisk were statistically significant (* = p < .05) 52 Discussion My study proposed a model of virtual decisi on-making whereby team decision quality in virtual teams would be influenced by the media ri chness of the CMC being utilized by the team and the CMC competence of team members. Pu lling from the Media Naturalness Hypothesis, Media Compensation Theory and Cues-Filtered Out Theory among others, I had predicted teams using CMCs lower in media richness would pool more unique information than teams using CMCs with higher media richness (H1) because it was thought that teams using low media richness CMCs would suffer from less cognitive lo ad, be more-task focused and have more equal participation amongst team memb ers. Additionally, I sought to replicate the common finding in the decision-making literature that unique information pooling would positively predict decision quality (H2). Combining H1 and H2 together, I had hypothesized that the relationship between media richness on decision quality would be me diated by the amount of unique information pooled (H3). Furthermore, I had also hypothesi zed that team CMC co mpetence would likewise positively predict unique information pooling (H4) as it was thought that teams possessing higher CMC competence, characterized by highe r motivation, knowledge and skills, would be more suited to communicating via CMC when co mpared to team™s low in CMC competence. Likewise I had expected this relationship team CMC competence would thus indirectly influence team decision quality through unique information pooling (H5). In the analyses, it was discovered that age was found to have a significant negative correlation with CMC competence. This rela tionship suggests that younger people view themselves as being more technologically lite rate and competent, reinforcing the rationale provided earlier that past findings on CMC technology need to be interpreted with caution society becomes increasingly competent with tec hnology as time passes. As the effect emerged 53 even within a relatively homogeneous sample in terms of age, one might expect a much stronger effect when comparing individuals in organi zations which generally contain much more variation in age. If the relationship were to be replicated in a worker sample, it would suggest that older workers may possess lower CMC competence, possessing lower motivation to collaborate virtually and lacking the skills and knowledge to utilize CMC to its full effect. Furthermore, CMC competence had a fairly larg e correlation with virtual decision-making self- efficacy. The relationship between CMC competen ce and virtual decision-making self-efficacy makes quite a bit of sense intuitively; a person that view himself/herself as being highly skilled in communicating via CMCs will also likely consider themselves to be highly efficacious in regards to communicating and collaborati ng virtually. Likewise, a person that views himself/herself as being incompetent in using CMCs would feel powerless and apprehensive in situations where they have to collaborate virtua lly, resulting in low virtual decision-making self- efficacy. This would have powerful implications in regards to team comp osition and selection of the appropriate communication medium for virtua l teams. Teams that are more homogeneous in age may share similar levels of CMC comp etence, meaning younger workers may be more familiar and confident in using newer technolog ies, whereas older workers may benefit more from less virtual communication. When teams are comprised of workers varying vastly in age, and CMC use is necessary, it may be beneficial to select a CMC that everyone has familiarity with (e.g. e-mail). Otherwise, effective team pe rformance may be hampered by a lack of self- efficacy and the skills/knowledge necessary to co llaborate virtually through CMC use. Overall, this suggests that age may be an important vari able of interest in studying the relationships between CMC use and team perf ormance for future studies. 54 However, it should also be noted that there was high levels of range restriction within my sample in regards to age, gender, CMC compet ence, self-efficacy, and engagement. This range restriction could be a key factor in not finding the expected relationships especially given that teams did not access enough information to make an informed decision. I also discovered a counter-intuitive relati onship between media richness and engagement whereby participants in the video conferencing co ndition rated themselves as being less engaged in the task. One would typically imagine that be ing able to see and hear your teammates should increase one™s engagement and enjoyment of the task, especially when the contrasting condition was where one typed to each other in silence. This finding also runs counter to the Media Naturalness Hypothesis which, in part, states that more natural mediums should lead to increased physiological arousal and thus engagement in the task. However, I do want to stress that this finding should be interpreted with caution as it may not necessaril y be only the characteristics of the medium that has caused this relationship. Medi a richness is defined, in part, by the capacity and speed of which information can be transm itted through the medium, thus richer mediums should naturally be able to communicate more quickly. From my observations, participants typically reached consensus rather quickly in the video conferencing condition, perhaps as a byproduct of being able to exchange information so rapidly. However, they were not allowed to end their discussion until the end of the allotted time period, resulti ng in long pauses and awkward silences in the conversa tion. Additionally, being able to see your teammates sitting in silence with bored expressions might have detr acted from the experien ce, resulting in lower engagement scores. On the other hand, in the instant messaging condition the amount of perceivable cues was decreased and breaks in communication were more normal since participants had to take time to think and type up their responses. It may be prudent to measure 55 time to decision to see whether it was the amount of fidead airtimefl after reaching a consensus that may have influenced engagement scores. However, if the relationship was in fact due to media richness, one possible interpretation is that extraneous information provided in the video conferencing CMC led to lowered engagement, perhaps through cognitive overload or fatigue. In the video conferencing condition, participants have to constantly attend to visual and audio cues over an extended period of time which may drain cognitive resources. Furthermore, when participants know that they are being visually observed they may feel ill at ease, requ iring them to continually self-regulate their behavior and expressions. On the other hand, pa rticipants in the instant messaging condition may have felt less need to regulate their behaviors/expressions as there are no visual or audio cues information being exchanged (e.g. body posture, facial expressions, eye gaze, tone, etc.), resulting to less drain in cognitive resources. If the decrease in engagement is in fact due to the cognitive overload/fatigue, then it might be prudent for future studies to measure cognitive overload/fatigue and how it is influenced by time (a s it is unlikely for a 5 minute conversation to be very taxing cognitively). Basically, how long is too long and what characteristics of the CMC are more taxing than others? Broader implications for the workplace might be that workers will have difficulty being engaged in long virtual m eetings where they have to attend to multiple sources of information, and that perhaps there is a good reason for why past studies have shown that many managers prefer e-mails over other forms of communication(Markus, 1994). Another possible interpretation is that this finding reflects the cue-removal principle of Media Compensation Theory, whereby the visual/audio cues presented in the video conferencing condition are imperfect and/or not matched with other expected cues (i.e. perceiving someone speaking from the video, but having the sound coming out delayed; expecting to see someone 56 making gestures when they are debating heated ly). Hantula et al. (2011) theorized that sometimes rich mediums such as video-confer encing might actually be more cognitively taxing than less media rich mediums like instant-messa ging if they present information and cues imperfectly, or fail to supplement the information with expected cues. An example might be an instance where latency causes a discrepancy between the video stream and the audio stream. In the visual stream of information you see your teammate speaking, providing you with a visual cue that you should expect audio information as well. However, the delay has mismatched the audio such that sometimes their mouth is m oving but no audio is coming through, or conversely where you hear words but the speaker™s mouth is not moving in the video stream. The cue- removal principle argues it requires more effortfu l processing of information when individuals have to actively suppress the confusion over why the information associated with certain cues is not occurring. This may, once again, lead to a draining of cognitive resources and/or cognitive fatigue, leading to less engagement in the task. Unfortunately, none of my main hypothe ses were supported in the study. I found no evidence that groups using a lower media richness CMC medium pooled more unique information than groups using a higher media richness CMC medium (H1). In fact, I found an effect in the opposite direction of what I had predicted, with groups in the video-conferencing condition (high media richness) pooling significa ntly more unique information. This finding, while not in line with my initial prediction, is important as it suggests that media richness can have significant effects on unique information pooling and that there are differences between CMCs. Previous studies have focused almost exclusively on an inappropriate FTF vs. CMC dichotomization (Lu et al., 2012). By lumping toge ther all CMCs into one general category we have lost sight of the significant differences that are present between CMCs, especially in 57 regards to its media richness. Furthermore, this finding reaffirms the need for researchers to stop generalizing the effects of one CMC across all CMCs and serves as a call to arms for researchers to pinpoint the key characteristics within CMCs that may be impacting information pooling and decision making (e.g. media richness). Support for Hypothesis 2 was also not found. There was no effect of unique information pooling on decision quality. This was surprising, especially considering the robustness of the relationship found in the literature. However, this might have been in part due to the overall low amount of unique information pooling witnessed across conditions. With the average team failing to pool even a quarter of the total available unique inform ation, it is highly possible that the failure to replicate the effect might have been due to the low performance of the sample. Additionally, it may be possible th at my stringent operationalization of decision quality (right vs. wrong) decreased my ability to detect small incr eases in performance. Some previous studies have included multiple dimensions of decision quality such as performance across multiple trials, time to decision or confidence in decision, which might have made it easier detect increases in performance(Kerr & Murthy, 2009; Mesmer-Magnus et al., 2011). Furthermore, there is a small likelihood that the task possessed a ficritical thresholdfl of unique information that needed to be reached for the correct decision to be made; mean ing any increases in unique information pooled below the critical amount would not have impacted decision quality as it was operationalized in the study. It may be prudent for fu ture studies to include multiple dimensions of decision quality and perhaps simplify the task so that gains in pe rformance can be seen, even with relatively low pooling involved. There is also the possibility that some part icipants were simply unable to accurately interpret the unique information being pooled. Fr om my observations, many times participants 58 would erroneously interpret some clues (e.g., viewing aggressive labor unions as a positive when it should be viewed as a negative for investme nt purposes) or bring along personal bias into interpreting clues (e.g., fioil spills are not a big deal, look at BP they™re still doing just finefl). Additionally, participants would frequently forget about what criteria they were supposed to be evaluating the companies on or introduced their own criteria for evaluation such as age of company or industry (e.g., many participants view ed Company B much more favorably as they viewed oil companies as being very lucrativ e companies). Some may also wonder about the difficulty or suitability of the task, since participant sample was predominantly students in the social sciences whereas the task was a busine ss decision task (company acquisition). However, the manipulation check showed that given a comple te information set, most participants were able to reach the correct decision; suggesting that overall the task was quite solvable, even in lieu of personal bias. Finally, it may have been that the compos ition of unique information pooled mattered more than the overall amount of unique inform ation pooled. The pieces of unique information were divided between companies A and B, all of wh ich could be objectively interpreted as either positive or negative towards the evaluation of the company. Since the common information was biased towards the selection of company B, it would requires teams to pool more positive information about A and negative information on B to overcome the bias from the common information. Furthermore, some teams may have pooled unique information equally about both companies and ficanceled outfl the positives and nega tive clues. Since overall pooling was so low, even if the majority of the unique information pieces pooled were towards the correct choice, it might not have been enough to overcome the co mmon information. Lastly, as participant A was given the full information set, he/she had th e most cognitive load and was also the only one 59 biased towards Company A, so the onus of reach ing the correct decision was largely in their hand. In future studies it may be necessary to reduce the overall initia l bias present in the common information, so that th ere is no majority opinion eff ect. Additionally, it might be necessary to reduce the cognitive load on participant A so that he/she does not need to attempt to remember so many pieces of information at once. Since I failed to find an effect for Hypothesis 2, my mediation hypotheses (Hypothesis 3 and 5) were also rejected as a consequence. Overall the analyses showed no significant difference between conditions in decision accuracy, with teams in both condition selecting the correct decision effectively at chance (33%). Inte restingly, the results showed that teams in the high media richness condition pooled more uniqu e information and also had higher average ratings of engagement in the task when compar ed to teams in the low media richness condition. If Hypothesis 2 had been supported it might have suggested that teams using high media richness CMCs pools more unique information and makes better decisions than teams using low media richness CMCs. This finding would have fallen in line with Media Richness Theory, which argues for a positive linear effect of increased media richness on unique information pooling and decision-making quality. Finally, although CMC competence was not f ound to predict unique information pooling, there were significant correlations between CM C competence and virtual decision-making self- efficacy and engagement. This suggests that while CMC competence may not influence actual performance, it definitely does influence percepti ons of self-efficacy and engagement in the task. In the post-hoc analyses it was discovered that CMC competence accounted for significant variance over and above media richness in pred icting engagement. This suggests that CMC competence may have relationships with other important outcomes beyond what was originally 60 hypothesized and merits future study. For example, employee engagement is commonly thought to be associated with higher productivity, job satisfaction and overall performance in organizations (Harter, Schmidt, & Hayes, 2002). Measuring CMC competence in highly virtualized workplaces may increase our unde rstanding of how CMCs influence worker engagement and provide us cues for when training interventions may be warranted to increase CMC competency and bolster engagement. Limitations and Future Directions While I failed to find support for my main hypotheses, I was able to find significant differences between conditions for unique information pooled and engagement. However, this relationship needs to be assessed with caution as it disappeared with the removal of 6 teams from the data set. Also, while CMC competence did not significantly predict unique information pooling, post-hoc analyses revealed that it did significantly predict engagement. These findings suggest that CMC competence may be a meani ngful construct to examine in regards to improving workplace engagement, which is often an important correlate of job performance and satisfaction. While I had argued for a negative relati onship between media richness and unique information pooling, after some consideration it would actually make more sense for the relationship to be curvilinear, such that at ex tremely low levels of me dia richness (e.g. sending letters) teams would likely perform more poor ly than teams using teleconferencing or videoconferencing. I would predict that instant messaging would be near the apex of the relationship between media richness and informa tion pooling, with extremely low CMCs (e.g. e-mail) performing similarly to videoconferencing. 61 One major issue in the study was the overall low performance of the student sample on the task. With an accuracy rate at chance (33% ) and the average team pooling only a quarter of the total unique information available, changes are warranted for future data collection attempts. It may be prudent to institute some type of incentive for performance, although the high engagement scores do suggest that lack of engage ment with the task was not the main issue. It may also be beneficial to include a forced recall test so that students would be more motivated to memorize specific points of information. This study specifically looked at decision-ma king teams utilizing CMCs as their main form of communication. Using an ad-hoc student sample I was una ble to analyze the effects of member familiarity, which have been known to bolster the effects of performance and satisfaction in virtual teams(Adams, Roch, & Ayman, 2005), but it would be recommended for future studies to see how team member familiarity may decrease the need for high team competence. For example, familiar teams may already have specific Transactive Memory Systems in place to facilitate info rmation sharing, and also foster a safe psychological climate, further increasing the chances of participation even from members lower in CMC competence. Similarly, at this time I am unable to compare teams with the same mean CMC competence but different compositions (e.g. one extremely competent member with two incompetent members, vs. three average competence members). Unfortun ately, with such low standard deviations within the sample in regards to CMC competence, it is unlikely that I would be able to attain this type of information using a student sample. Si nce a significant relationship between age and CMC competence was discovered, a study using an organizational sample comprising of workers from many different generations would be ideal for exploring compositional differences as well as further clarifying the relationship between age and CMC competence. 62 Additionally, in this study I used a broad measure of CMC competence that was not limited to a specific CMC, however, it is possible th at some individuals are much more skilled in certain types of CMCs as opposed to others. As such, future studies may benefit from creating alternate measures of CMC competence that are adapted to the specific medium the participant will be using, especially when comparing CMCs that are drastically different (e.g. e-mail vs. video conferencing). The sample was also predominantly female and relatively young, raising potential concerns regarding the generalizability of the findings and whether a gender effect was masked because of the low number of male participants. Furthermore, my analyses revealed potential issues with range restriction that may further hamper generalizability. Wh ile it would definitely help to conduct subsequent follow-up studies using samples of workers to confirm generalizability, this study still provides valuable insight into a young, technologically competent generation that should generalize well for the next generation of workers. Finally, in this study I was primarily intere sted in showing the difference in media richness between CMCs (i.e. instant messagi ng, vs. video-conferencing) so each team was restricted to a specific medium, but in practi ce teams may communicate utilizing multiple CMCs or use a mixture of FTF and CMCs. There has be en some promising research on the concept of fivirtuality,fl or the degree to which a team co llaborates and communicates virtually as mentioned in the work by (Mesmer-Magnus et al., 2011), but th at is beyond the scope of this investigation. 63 APPENDICES 64 Appendix A: Initial Measures CMC Competence Measure Instructions: People differ quite a bit in terms of how skilled they are at using computer media (including instant messaging, e-mail, Instagram, Twitter, Facebook, etc.) in communicating and conversing with others. For the following statements, we would like you to estimate, compared to typical people you encounter, how skilled you are in using computer-mediated communication (i.e., CMC). CMCs include things such as Facebook, Skype, e-mail, Twitter, Instagram, Google Hangout and so forth, basi cally whenever you are communicating using a computer or smartphone rather than face to face you are using a CMC to communicate. In the following questions, use the scale to select the response that best describes you. 1 = Not at all true of me 2 = Mostly not true of me 3 = Somewh at not true of me 4 = Neither true nor untrue of me; und ecided 5 = Somewhat true of me 6 = Mostly true of me 7 = Very true of me Select the response that best describes you. Reminder: CMCs are communication mediums such as Facebook, e-mail, Twitter, Skype, etc. 1. I enjoy communicating using CMCs. 2. I am nervous about using CMCs to communicate with others. [R] 3. I am very motivated to use CMCs to communicate with others. 4. I look forward to using CMCs to communicate with others. 5. Communicating through CMCs makes me anxious. [R]\ 6. I am very knowledgeable about how to communicate using CMCs. 7. I am never at a loss for something to say using CMCs. 8. I am very familiar with how to communicate using CMCs. 9. I always seem to know how to say things the way I mean them using CMCs. 10. When communicating with someone through CM Cs, I know how to adapt my messages to the medium. 11. I know when and how to close down a topic of conversation when using CMCs. 12. I manage the give and take of CMC interactions skillfully. 65 13. I am skilled at timing when I send my re sponses to people who contact me through CMCs. 14. I ask questions of the other person in CMC conversations so I know exactly what they mean and/or show them I™m paying attention. 15. I show concern for and interest in people I™m conversing with through CMCs. 16. I make sure my objectives are em phasized in my CMC messages. 17. My CMC messages are written in a confident style. 18. I am skillful at revealing composure and self-confidence in my CMC interactions. I choose which medium (i.e., e-mail, Facebook, Tw itter, Skype, etc.) to communicate based on... 19. –how quickly I need to get a message out to people. 20. –how lively the interaction needs to be. 21. –how much access the person I need to comm unicate with has to the CMC medium. 22. –how much information is involved in the message I need to communicate. 23. –how much access I have to the CMC medium. [R] 24. –how much personal or intimate the information in the message is. 66 Virtual Decision-Making Self-Efficacy Measure 1 = Strongly disagree 2 = Disagree 3 = Neither agree nor disagree 4 = Agree 5 = Strongly agree When I am in a virtual decision-making team using CMC, I am confident that... 1. ...I can always manage to overcome issues with gathering information and/or making decisions virtually if I try hard enough. 2. ...if I encounter difficulties, I can find the means and ways to get what I want. 3. ...it will be easy for me to stick to my aims and get my point across, or get information from others using CMC. 4. ...I can deal with unexpected issues with gathering information and making decisions using CMC. 5. ...I can solve most problems if I invest the necessary effort. 6. ...I can remain calm when facing difficulties with CMC and/or making decisions virtually because I can rely on my abilities. 7. ...when I have to make a decision using CMC, I can usually find several solutions. 8. ...I can usually handle whatever problem that comes my way. 9. ...setbacks and failures I encounter while working in the team will only make me try harder. 10. ...I do not need assistance from others to utilize the CMC medium to its full potential. 11. ...there are few decisions I would be uncomfortable making using CMC. 12. ...I can persist and solve most any problem using CMC. 67 User Engagement Measure Pleaseselecttheresponsethatbestdescribes yourexperience intoday™sexperiment. 1 = Strongly disagree 2 = Disagree 3 = Neither agree nor disagree 4 = Agree 5 = Strongly agree 1. I felt interested in the decision-making task. 2. The content of the company pr ofiles incited my curiosity. 3. The discussion with my team was fun. 4. I felt involved in the decision-making process. 5. My overall experience was rewarding. 6. I would recommend this experiment to my friends and classmates. 7. I was really drawn into the discussion. 8. I consider our performance successful. 9. Talking to each other through CMC was worthwhile. 68 Appendix B: Exploratory Factor Analysis Re sults Œ Principal Axis Factor Extraction, Direct Oblimin Rotation Items Factor 1 2 3 I enjoy communicating using CMCs. .86 [Reverse]I am nervous about using CMCs to communicate with others. .33 I am very motivated to use CMCs to communicate with others. .74 I look forward to using CMCs to communicate with others. .80 [Reverse]Communicating through CMCs makes me anxious. I am very knowledgeable about how to communicate using CMCs. .61 I am never at a loss for something to say using CMCs. .45 I am very familiar with how to communicate using CMCs. .53 I always seem to know how to say things the way I mean them when using CMCs. .55 When communicating with someone through CMCs, I know how to adapt my messages to the medium. .54 I know when and how to close down a topic of conversation when using CMCs. .53 I manage the give and take of CMC interactions skillfully. .62 I am skilled at timing when I send my responses to people who contact me through CMCs. .63 I ask questions of the other person in CMC conversations so I know exactly what they mean and/or to show them I'm paying attention. .50 I show concern for and interest in people I'm conversing with through CMCs. .53 I can show compassion and empathy with others through CMCs. .41 I make sure my objectives are emphasized in my CMC messages. .56 69 My CMC messages are written in a confident style. .52 I am skillful at revealing composure and self-confidence in my CMC interactions. .48 I choose which medium (i.e., e-mail, Facebook, Twitter, Skype, etc.) to communicate based on...- ...how quickly I need to get a message out to people. -.75 I choose which medium (i.e., e-mail, Facebook, Twitter, Skype, etc.) to communicate based on...-... how lively the interaction needs to be. -.66 I choose which medium (i.e., e-mail, Facebook, Twitter, Skype, etc.) to communicate based on...- ...how much access the person I need to communicate with has to the CMC medium. -.77 I choose which medium (i.e., e-mail, Facebook, Twitter, Skype, etc.) to communicate based on.. .-...how much information is involved in the message I need to communicate. -.74 [Reverse] I choose which medium (i.e., e-mail, Facebook, Twitter, Skype, etc.) to communicate based on...-...how much access I have to the channel or medium. .69 I choose which medium (i.e., e-mail, Facebook, Twitter, Skype, etc.) to communicate based on...- ...how personal or intimate the information in the message is. -.67 When I am in a virtual decision-making team using CMC, I am confident that...-...I can always manage to overcome issues with gathering information and/or making decisions virtually if I try hard enough. .49 When I am in a virtual decision-making team using CMC, I am confident that...-...if I encounter difficulties, I can find the means and ways to get what I want. .58 When I am in a virtual decision-making team using CMC, I am confident that...-...it will be easy for me to stick to my aims and get my point across, or get information from others using CMC. .50 When I am in a virtual decision-making team using CMC, I am confident that...-...I can deal with unexpected issues with gathering .61 70 information and making decisions using CMC. When I am in a virtual decision-making team using CMC, I am confident that...-...I can solve most problems if I invest the necessary effort. .65 When I am in a virtual decision-making team using CMC, I am confident that...-...I can remain calm when facing difficulties with CMC and/or making decisions virtually because I can rely on my abilities. .63 When I am in a virtual decision-making team using CMC, I am confident that...-...when I have to make a decision using CMC, I can usually find several solutions. .60 When I am in a virtual decision-making team using CMC, I am confident that...-...I can usually handle whatever problem that comes my way. .78 When I am in a virtual decision-making team using CMC, I am confident that...-...setbacks and failures I encounter while working in the team will only make me try harder. .40 When I am in a virtual decision-making team using CMC, I am confident that...-...I do not need assistance from others to utilize the CMC medium to its full potential. .46 When I am in a virtual decision-making team using CMC, I am confident that...-...there are few decisions I would be uncomfortable making using CMC. When I am in a virtual decision-making team using CMC, I am confident that...-...I can persist and solve most any problem using CMC. .61 71 Appendix C: Revised Measures CMC Competence Measure Instructions: People differ quite a bit in terms of how skilled they are at using computer media (including instant messaging, e-mail, Instagram, Twitter, Facebook, etc.) in communicating and conversing with others. For the following statements, we would like you to estimate, compared to typical people you encounter, how skilled you are in using computer-mediated communication (i.e., CMC). CMCs include things such as Facebook, Skype, e-mail, Twitter, Instagram, Google Hangout and so forth, basi cally whenever you are communicating using a computer or smartphone rather than face to face you are using a CMC to communicate. In the following questions, use the scale to select the response that best describes you. 1 = Not at all true of me 2 = Mostly not true of me 3 = Somewh at not true of me 4 = Neither true nor untrue of me; und ecided 5 = Somewhat true of me 6 = Mostly true of me 7 = Very true of me Select the response that best describes you. Reminder: CMCs are communication mediums such as Facebook, e-mail, Twitter, Skype, etc. 1. I enjoy communicating using CMCs. 2. I am nervous about using CMCs to communicate with others. [R] 3. I am very motivated to use CMCs to communicate with others. 4. I look forward to using CMCs to communicate with others. 5. I am very knowledgeable about how to communicate using CMCs. 6. I am never at a loss for something to say using CMCs. 7. I am very familiar with how to communicate using CMCs. 8. I always seem to know how to say things the way I mean them using CMCs. 9. When communicating with someone through CM Cs, I know how to adapt my messages to the medium. 10. I know when and how to close down a topic of conversation when using CMCs. 11. I manage the give and take of CMC interactions skillfully. 12. I am skilled at timing when I send my re sponses to people who contact me through CMCs. 72 13. I ask questions of the other person in CMC conversations so I know exactly what they mean and/or show them I™m paying attention. 14. I show concern for and interest in people I™m conversing with through CMCs. 15. I make sure my objectives are em phasized in my CMC messages. 16. My CMC messages are written in a confident style. 17. I am skillful at revealing composure and self-confidence in my CMC interactions. 73 Virtual Decision-Making Self-Efficacy Measure 1 = Strongly disagree 2 = Disagree 3 = Neither agree nor disagree 4 = Agree 5 = Strongly agree When I am in a virtual decision-making team using CMC, I am confident that... 1. ...I can always manage to overcome issues with gathering information and/or making decisions virtually if I try hard enough. 2. ...if I encounter difficulties, I can find the means and ways to get what I want. 3. ...it will be easy for me to stick to my aims and get my point across, or get information from others using CMC. 4. ...I can deal with unexpected issues with gathering information and making decisions using CMC. 5. ...I can solve most problems if I invest the necessary effort. 6. ...I can remain calm when facing difficulties with CMC and/or making decisions virtually because I can rely on my abilities. 7. ...when I have to make a decision using CMC, I can usually find several solutions. 8. ...I can usually handle whatever problem that comes my way. 9. ...setbacks and failures I encounter while working in the team will only make me try harder. 10. ...I do not need assistance from others to utilize the CMC medium to its full potential. 11. ...I can persist and solve most any problem using CMC. 74 Appendix D: Task Information Packets Hidden Profile Packet A/Individual Participant Task (Full Information Set): ACME Inc.: Group Decision-Making for Investments Instructions ~~~ Most companies make important investment deci sions using a team approach. Your group here today represents the top management t eam of ACME (fiAcquiring Companies Means Employment!fl), Inc. Your company has been presented with the opportunity to acquire three smaller firms. ACME has $100 million to invest, which will allow the acquisition of only one of these firms. The Chairperson of the Board has appointed you to research the three acquisition targets and to recommend which one of them would be best for ACME. There are a number of key f actors that you should consider carefully in evaluating these companies. First, ACME prefers to acquire firm s that will maximize wea lth, over the long term. Which of these companies has the most promisin g future? Therefore you should consider the potential return on your investment. A second consideration is the likelihood of you actually getting that return, in the long run. That is, how precise is the projection and what is the probability that your actual return will be significa ntly different than the best estimate? Third, you should also consider the growth potential of each company™s market. You would prefer to invest in a company that competes in a growing market. A fourth consideration is the quality of the company™s management team. ACME takes a fihands-offfl approach with its subsidiaries. Therefore, you prefer to invest in compan ies whose management team can achieve the profitability you desire. Fina lly, you should judge each company™s general strategy and business policies. Do they seem like policies that will lead the company to profitability in the future? In order to help you evaluate these companies, your in-house financial analyst has researched each company. Further, you have retained the co nsulting services of Smith, Barney & Howe, a highly respected and successful investment c onsulting firm, also to analyze these three companies. The results appear in the reports c ontained in your information packets. You should review all of this information, and based upon it, come to a conclusion about which of these three companies would be the right acquisition for ACME. The Chair of the Board wants each of you indi vidually to submit your personal recommendation, whether or not it agrees with the team recommendation. After you have studied the material and recorded your personal recomme ndation, you will decide as a team which of the three companies ACME should acquire. There must be consensus agreement on the top ranked company . 75 Company A fiWhiz-Bang Electronicsfl Industry: Industrial Electronics Products: Electronic manuf acturing control devices Location: Metropol, California Size: $50 million in sales; 200 employees Age: Established 5 years ago I. Financial Your internal financial analyst estimates that the internal rate of return (i.e., the return on your investment) will be 15% annually over the next 10 years. This analyst believes the chances of you actually getting this return is 70 percent. Further, the analyst estimates that there is a 15 percent chance that ACME will either double this return (thereby providing a 30% return) or will have a zero return. The Smith, Barney & Howe consultants concur with the conclusions of your in-house analyst. In fact, SBH believes that there is an 80 percent chance of your obtaining the projected return. Both your internal financial analyst and the SBH consultants agree, however, that there is a near certain pr obability that ACME will suffer a loss during the first year, and that you would not achieve any return until after that time. This compa ny™s growth in sales has been positive, hovering around 5% annually from the beginning, but early projections indicate an increase to 8% for the next fiscal year. Further, this market is expected to grow in the foreseeable future. II. Strategic Whiz-Bang Electronics is young, and was founded by a group of bright and talented entrepreneurs whose management experience was limited, at the start. The company has an innovative and promising product line. The inexpe rience of the management team led to some early mistakes in marketing and di stribution such that customer awareness of the products is low. As a result the company has only a 6% market sh are and low customer perceptions of service. Furthermore, Whiz-Bang Electronics ™ pricing structure is not suitable for its target customers. The company leadership team has been actively de veloping their professional managerial skills through workshops and close work with experienced consultants. Industry watchers have noted that this group seems to be making more effec tive decisions, which are probably responsible for the recent sales growth. III. Labor Whiz-Bang Electronics has very high labor costs. It spends a lot of money on employee development. They offer training in a va riety of business-related skills ranging from communication to accounting principles. The company™s recruiting processes are drawn-out, but very thorough and careful. Recruiting expenses represent a very large chunk of the company™s operating budget. They provide fitness facilities and on-site child care for all employees. 76 Company B fiPower Energyfl Industry: Energy Products: Power for heavy manufacturing Location: Bigtown, Texas Size: $50.5 million in sales; 225 employees Age: Established 25 years ago I. Financial Your internal financial analyst estimates that the internal rate of return (i.e., the return on your investment) will be 25% annually over the next 10 years. This analyst believes the chances of you actually getting this return is 70 percent. Further, the analyst estimates that there is a 15 percent chance that ACME will either double this return (thereby providing a 50% return) or have a zero return. The Smith, Barney & Howe consultants disagree with the conclusions of your in-house analyst, however. They believe that the rate of return will be lower. In fact, SBH estimates the rate of return will only be 5%, a nd that the chance of you ge tting that return will be 40 percent. Further, SBH expects a 30% chance e ither way that the return could double (thereby providing a 10% return) or that it could be zero. Power Energy historically has experienced growth in sales averaging 10% annually. It experienced record growth of 15% five years ago. The growth figures since then have been 12%, 10%, 9.3%, and 8%. The best estimates indicate flat growth in the overall market over the near future. II. Strategic Power Energy has a 30% share of the market. The company also enjoys strong name recognition among the public. The current management team is responsible for moving this company to the top of its market, 20 years ago. Their manage ment style has evolved to a fimaintenancefl strategy, and some in the industry view them as be ing out of touch with current trends in their markets. Growing concern for the environment, especially related to energy consumption, have started to mandate changes in th e way that energy companies delive r product to their customers. Companies able to offer innovations that re duce negative environmental impact will almost certainly merge to the market forefront soon. The company has been involved in off-shore oil drilling and exploration, and has made significant profits. A recent fine and responsibility for some clean-up costs, however has resulted in a 6% reduction in bottom line profits ov er the next 2 years. One concern is that a number of foreign companies, whose off-shore e xplorations are subsidized by their governments, are poised to enter Power Energy™s market. 77 III. Labor Power Energy™s labor force consists primarily of semi-skilled workers and engineers who think of this company as offering them lifetime employment. The company is also known for its generous compensation and benefits packages. 78 Company C fiQuality Tool & Diefl Industry: Industrial Products Products: Tool & Die for heavy manufacturing Location: Midville, Indiana Size: $50.2 million in sales; 175 employees Age: Established 17 years ago I. Financial Your internal financial analyst estimates that the internal rate of return (i.e., the return on your investment) will be 8% annually over the next 10 y ears. This analyst believes the chance of you actually getting this return is 60 percent. Further, the analyst estimates that there is a 20 percent chance either way that ACME will double this re turn (thereby providing a 16% return) or will have a zero return. The analysis indicates further that there is a near certain probability that you will suffer a loss during the first year, and that yo u would not achieve any return until after that time. The Smith, Barney & Howe consultants agr ee with your analyst™s conclusions. Growth in sales has been averaging around 6% annually. II. Strategic Quality Tool & Die is in a mature industry with very little change forecasted for the foreseeable future. They have managed to maintain their 12% market share in an environment which is expected to remain in a competitive equilibrium in the near future. Their management team is solid and respectable. They have not been know n to make any major mistakes, nor have they contributed major innovati ons to their industry. III. Labor Their labor force is unionized, composed mostly of unskilled workers employed in assembly line jobs who receive their training on-the-job. The company has managed to keep the relationship with the unions relatively trouble-free, but a the newly elected union leadership is known to have an aggressive and confrontational attitude toward management. The company™s labor turnover has been low. 79 Hidden Profile Packet B: ACME Inc.: Group Decision-Making for Investments Instructions **** Most companies make important investment deci sions using a team approach. Your group here today represents the top management t eam of ACME (fiAcquiring Companies Means Employment!fl), Inc. Your company has been presented with the opportunity to acquire three smaller firms. ACME has $100 million to invest, which will allow the acquisition of only one of these firms. The Chairperson of the Board has appointed you to research the three acquisition targets and to recommend which one of them would be best for ACME. There are a number of key f actors that you should consider carefully in evaluating these companies. First, ACME prefers to acquire firm s that will maximize wea lth, over the long term. Which of these companies has the most promisin g future? Therefore you should consider the potential return on your investment. A second consideration is the likelihood of you actually getting that return, in the long run. That is, how precise is the projection and what is the probability that your actual return will be significa ntly different than the best estimate? Third, you should also consider the growth potential of each company™s market. You would prefer to invest in a company that competes in a growing market. A fourth consideration is the quality of the company™s management team. ACME takes a fihands-offfl approach with its subsidiaries. Therefore, you prefer to invest in compan ies whose management team can achieve the profitability you desire. Fina lly, you should judge each company™s general strategy and business policies. Do they seem like policies that will lead the company to profitability in the future? In order to help you evaluate these companies, your in-house financial analyst has researched each company. Further, you have retained the co nsulting services of Smith, Barney & Howe, a highly respected and successful investment c onsulting firm, also to analyze these three companies. The results appear in the reports c ontained in your information packets. You should review all of this information, and based upon it, come to a conclusion about which of these three companies would be the right acquisition for ACME. The Chair of the Board wants you to decide, as a team, which of the three companies ACME should acquire. There must be consensus agreement on which company to acquire . 80 Company A fiWhiz-Bang Electronicsfl Industry: Industrial Electronics Products: Electronic manuf acturing control devices Location: Metropol, California Size: $50 million in sales; 200 employees Age: Established 5 years ago I. Financial Your internal financial analyst estimates that the internal rate of return (i.e., the return on your investment) will be 15% annually over the next 10 years. Further, the analyst estimates that there is a 15 percent chance that ACME will ha ve a zero return. The Smith, Barney & Howe consultants concur with the conclusions of your in-house analyst. Both analyses agree that there is a near certain probability that ACME will suffer a loss during the first year, and that you would not achieve any return until after that time. This company™s growth in sales has been halting, hovering around 5% annually from the beginning. II. Strategic Whiz-Bang Electronics is young, and was founded by a group whose management experience was limited. The inexperience of the management team led to some early mistakes in marketing and distribution such that customer awareness of the products is low, and so are perceptions of service. Furthermore, the prici ng structure is not suitable for their target customers. As a result the company has been a market laggard, averag ing only a 6% market share. The company leadership has been trying to address these issues head-on. III. Labor Whiz-Bang Electronics has very high labor costs. It spends a lot of money on employee development, such as providing on-site fitness fa cilities. Their recru iting processes are drawn-out. These expenditures represent a very large chunk of the company™s operating budget. 81 Company B fiPower Energyfl Industry: Energy Products: Power for heavy manufacturing Location: Bigtown, Texas Size: $50.5 million in sales; 225 employees Age: Established 25 years ago I. Financial Your internal financial analyst estimates that the internal rate of return (i.e., the return on your investment) will be 25% annually over the next 10 years. This analyst believes the chances of you actually getting this return is 70 percent. Further, the analyst estimates that there is a 15 percent chance that ACME will double this return (thereby providing a 50% return). The Smith, Barney & Howe consultants estimated a lower rate of return than did your internal analyst, and they believed there would be a 30 percent chance of doubling their estimated return. Power Energy historically has experienced growth in sales averaging 10% annually. It experienced record growth of 15% five years ago. Last year™s growth was 8%. II. Strategic Power Energy has been the market leader for over two decades. It dominates the market with 30% share. The company enjoys strong na me recognition among the public. The current management team is responsible for moving this company to the top of its market 15-20 years ago. Growing concern for the environment, esp ecially related to energy consumption, have started to mandate changes in th e way that energy companies deliver product to their customers. The company has been involved in the risky fiel d of off-shore oil drilling and exploration, and has made significant profits. A recent problem, however, resulted in the company receiving a fine and being responsible for some clean-up costs. III. Labor Power Energy™s labor force consists primarily of semi-skilled workers and engineers. The company has had the reputation of offering job se curity and generous compensation and benefit packages. 82 Company C fiQuality Tool & Diefl Industry: Industrial Products Products: Tool & Die for heavy manufacturing Location: Midville, Indiana Size: $50.2 million in sales; 175 employees Age: Established 17 years ago I. Financial Your internal financial analyst estimates that the internal rate of return (i.e., the return on your investment) will be 8% annually over the next 10 y ears. This analyst believes the chance of you actually getting this return is 60 percent. Further, the analyst estimates that there is a 20 percent chance either way that ACME will double this re turn (thereby providing a 16% return) or will have a zero return. The analysis indicates further that there is a near certain probability that you will suffer a loss during the first year, and that yo u would not achieve any return until after that time. The Smith, Barney & Howe consultants agr ee with your analyst™s conclusions. Growth in sales has been averaging around 6% annually. II. Strategic Quality Tool & Die is in a mature industry with very little change forecasted for the foreseeable future. They have managed to maintain their 12% market share in an environment which is expected to remain in a competitive equilibrium in the near future. Their management team is solid and respectable. They have not been know n to make any major mistakes, nor have they contributed major innovati ons to their industry. III. Labor Their labor force is unionized, composed mostly of unskilled workers employed in assembly line jobs who receive their training on-the-job. The company has managed to keep the relationship with the unions relatively trouble-free, but the newly elected union leadership is known to have an aggressive and confrontational attitude toward management. The company™s labor turnover has been low. 83 Hidden Profile Packet C: ACME Inc.: Group Decision-Making for Investments Instructions ****** Most companies make important investment deci sions using a team approach. Your group here today represents the top management t eam of ACME (fiAcquiring Companies Means Employment!fl), Inc. Your company has been presented with the opportunity to acquire three smaller firms. ACME has $100 million to invest, which will allow the acquisition of only one of these firms. The Chairperson of the Board has appointed you to research the three acquisition targets and to recommend which one of them would be best for ACME. There are a number of key f actors that you should consider carefully in evaluating these companies. First, ACME prefers to acquire firm s that will maximize wea lth, over the long term. Which of these companies has the most promisin g future? Therefore you should consider the potential return on your investment. A second consideration is the likelihood of you actually getting that return, in the long run. That is, how precise is the projection and what is the probability that your actual return will be significa ntly different than the best estimate? Third, you should also consider the growth potential of each company™s market. You would prefer to invest in a company that competes in a growing market. A fourth consideration is the quality of the company™s management team. ACME takes a fihands-offfl approach with its subsidiaries. Therefore, you prefer to invest in compan ies whose management team can achieve the profitability you desire. Fina lly, you should judge each company™ s general strategy and business policies. Do they seem like policies that will lead the company to profitability in the future? In order to help you evaluate these companies, your in-house financial analyst has researched each company. Further, you have retained the co nsulting services of Smith, Barney & Howe, a highly respected and successful investment c onsulting firm, also to analyze these three companies. The results appear in the reports c ontained in your information packets. You should review all of this information, and based upon it, come to a conclusion about which of these three companies would be the right acquisition for ACME. The Chair of the Board wants you to decide, as a team, which of the three companies ACME should acquire. There must be consensus agreement on which company to acquire . 84 Company A fiWhiz-Bang Electronicsfl Industry: Industrial Electronics Products: Electronic manuf acturing control devices Location: Metropol, California Size: $50 million in sales; 200 employees Age: Established 5 years ago I. Financial Your internal financial analyst estimates that the internal rate of return (i.e., the return on your investment) will be 15% annually over the next 10 years. Further, the analyst estimates that there is a 15 percent chance that ACME will ha ve a zero return. The Smith, Barney & Howe consultants concur with the conclusions of your in-house analyst. Both analyses agree that there is a near certain probability that ACME will suffer a loss during the first year, and that you would not achieve any return until after that time. This company™s growth in sales has been halting, hovering around 5% annually from the beginning. II. Strategic Whiz-Bang Electronics is young, and was founded by a group whose management experience was limited. The inexperience of the management team led to some early mistakes in marketing and distribution such that customer awareness of the products is low, and so are perceptions of service. Furthermore, the prici ng structure is not suitable for their target customers. As a result the company has been a market laggard, averag ing only a 6% market share. The company leadership has been addressing these issues head-on. III. Labor Whiz-Bang Electronics has very high labor costs. It spends a lot of money on employee development, such as providing on-site fitness fa cilities. The company™s recruiting processes are drawn-out, and these expenditures represent a very large chunk of the company™s operating budget. 85 Company B fiPower Energyfl Industry: Energy Products: Power for heavy manufacturing Location: Bigtown, Texas Size: $50.5 million in sales; 225 employees Age: Established 25 years ago I. Financial Your internal financial analyst estimates that the internal rate of return (i.e., the return on your investment) will be 25% annually over the next 10 years. This analyst believes the chances of you actually getting this return is 70 percent. Further, the analyst estimates that there is a 15 percent chance that ACME will double this return (thereby providing a 50% return) or will have a zero return. The Smith, Barney & Howe consultants estimated a lower rate of return than did your internal analyst, and they believed there would be a 30 percent chance of doubling their estimated return. Power Energy historically ha s experienced growth in sales averaging 10% annually. It experienced record growth of 15% five years ago. The best estimates indicate flat growth in the overall market over the near future. II. Strategic Power Energy has been the market leader for over two decades. It dominates the market with 30% share. The company enjoys strong na me recognition among the public. The current management team is responsible for moving this company to the top of its market. The company has been involved in off-shore oil drilling and exploration, and has made significant profits, despite recent problems. One c oncern is that a number of foreign companies, whose off-shore explorations are subsidized by their governments, are poised to enter Power Energy™s market. III. Labor Power Energy™s labor force consists primarily of semi-skilled workers and engineers. The company has had the reputation of offering job se curity and generous compensation and benefit packages. 86 Company C fiQuality Tool & Diefl Industry: Industrial Products Products: Tool & Die for heavy manufacturing Location: Midville, Indiana Size: $50.2 million in sales; 175 employees Age: Established 17 years ago I. Financial Your internal financial analyst estimates that the internal rate of return (i.e., the return on your investment) will be 8% annually over the next 10 y ears. This analyst believes the chance of you actually getting this return is 60 percent. Further, the analyst estimates that there is a 20 percent chance either way that ACME will double this re turn (thereby providing a 16% return) or will have a zero return. The analysis indicates further that there is a near certain probability that you will suffer a loss during the first year, and that yo u would not achieve any return until after that time. The Smith, Barney & Howe consults agre e with your analyst™s conclusions. Growth in sales has been averaging around 6% annually. II. Strategic Quality Tool & Die is in a mature industry with very little change forecasted for the foreseeable future. They have managed to maintain their 12% market share in an environment which is expected to remain in a competitive equilibrium in the near future. Their management team is solid and respectable. They have not been know n to make any major mistakes, nor have they contributed major innovati ons to their industry. III. Labor Their labor force is unionized, composed mostly of unskilled workers employed in assembly line jobs who receive their training on-the-job. The company has managed to keep the relationship with the unions relatively trouble-free, but the newly elected union leadership is known to have an aggressive and confrontational attitude toward management. The company™s labor turnover has been low. 87 Appendix E: Checklist Form Checklist Form Here are some pieces of information that were mentioned in some of your information packets. We wanted to see which pieces of informa tion you shared with the group. Please put a checkmark next to each information piece that was mentioned during your conversation with the group, regardless of whether you considered it an fiimportantfl piece or not. Remember, this is only what you specifically said or wrote, not what you remembered seeing from the study materials. Also, please provide us with your preference prior to disc ussing with the team. Member A Choice: Member B Choice: Member C Choice: Team Choice: Independent Choice: Company A -- fiWhiz Bang Electronicsfl Part 1 ____________ 1. Internal analyst expects 15 percent chance of 30% IRR (i.e., double estimated return) ____________ 2. Near certain probability of first year loss. ____________ 3. Halting sales growth ____________ 4. Low customer perceptions of service ____________ 5. Pricing structure may not be suitable ____________ 6. Company has been market laggard ____________ 7. Company leadership addressing problems head on ____________ 8. Employee expenditures take a large chunk of company budget ____________ 9. Internal analyst expects 70 percent chance of 15% IRR ____________ 10. Internal analyst expects fift een percent chance of 0 IRR fi ____________ 11. SBH expects 80 per cent chance of 15% IRR ____________ 12. Early projection indicate 8% increase in sales growth for next year (i.e., positive sales growth) 88 ____________ 13. Market expected to grow in near future ____________ 14. Founded by bright & ta lented entrepreneurs ____________ 15. Innovative and promising product line ____________ 16. Management team activ ely developing professional managerial skills (i.e., participating in workshops and working with consultants) ____________ 17. Industry watchers note the group is making more effective decisions/more effective decisions probably responsible for recent sales increase ____________ 18. Company offers employees tr aining in business-related skills ____________ 19. Company as thorough and careful recruiting process ____________ 20. Company provides on-site child care Company B -- fiPower Energyfl ___________ 21. Internal analyst estimates 70 percent chance of return 25% IRR ___________ 22. Internal analyst estimates fifteen percent chance of 50% IRR ___________ 23. Company has made significan t profits in off-shore drilling ___________ 24. Company dominates mark et/ leader for 2 decades ___________ 25. Company has strong name recognition ___________ 26. Management team™s reputation well respected ___________ 27. SBH expects 30 percent chance of getting 10% IRR (i.e., double estimated return) ___________ 28. Reputation for job security ___________ 29. Generous compensation benefits 89 ___________ 30. Last year™s growth was 8% ___________ 31. Off-shore drilling an d exploration are risky ___________ 32. Company™s recent problem has re sulted in fines & clean-up costs. ___________ 33. Estimates indicate flat growth in the market ___________ 34. Foreign competition poised to enter market ___________ 35. Foreign competitors have government subsidy ___________ 36. Company has had recent problem ___________ 37. Management team moved company to top of market 15-20 years ago. ___________ 38. Growing concern for envi ronment mandating changes in energy companies ___________ 39. Internal analyst estimates fifteen percent chance of 0 IRR ___________ 40. SBH disagrees with the internal analyst ___________ 41. SBH estimates a 5% IRR ___________ 42. SBH estimates 40 percent chance of 5% IRR ___________ 43. SBH expects 30 percen t chance of getting 0 IRR. ___________ 44. Growth figures for last few years have been 12%, 10%, 9%, 3% (i.e., declining sales growth) ___________ 45. Fine & clean-up costs result in 6% reduction in bottom line profits ___________ 46. Management team style has evolved to fimaintenancefl strategy/ some view mana gement team as out of touch ___________ 47. Companies with reduced negative environment impact will emerge as leaders ___________ 48. Employees see company as offering lifetime employment 90 REFERENCES 91 REFERENCES Adams, S. J., Roch, S. G., & Ayman, R. (2005). Communication medium and member familiarity: The effects on decisi on time, accuracy, and satisfaction. Small Group Research, 36(3), 321-353. doi: 10.1177/1046496405275232 Adrianson, L., & Hjelmquist, E. (1991). Gr oup Processes in Face-to_face and Computer- Mediated Communication. Behaviour & Information Technology, 10(4), 281-296. Baltes, B. B., Dickson, M. W., Sherman, M. P ., Bauer, C. C., & LaGanke, J. S. (2002). Computer-Mediated Communication and Gr oup Decision Making: A Meta-Analysis. Organizational Behavior and Human Decision Processes, 87(1), 156-179. doi: http://dx.doi.org/10.1006/obhd.2001.2961 Bandura, A. (1977). Self-efficacy: toward a unifying theory of behavioral change. Psychological Review, 84(2), 191-215. doi: 10.1037//0033-295x.84.2.191 Bandura, A. (1982). Self-efficacy mechanism in human agency. American Psychologist, 37(2), 122-147. doi: 10.1037//0003-066x.37.2.122 Berger, J., Cohen, B. P., & Zelditch, M. (1972). Status Characteristics and Social Interaction. American Sociological Review, 37(3), 241. doi: 10.2307/2093465 Bunz, U., & Campbell, S. W. (2004). Politeness accommodation in electronic mail. Communication Research Reports, 21(1), 11-25. Castella, V. O., Abad, A. M. Z., Alonso, F. P., & Silla, J. M. P. (2000). The influence of familiarity among group members, group atmosp here and assertiveness on uninhibited behavior through three different communication media. Computers in Human Behavior, 16(2), 141-159. Copeland, M. V. (2006). The Mighty Micro Multi-National. Business 2.0, 7, 106-114. Crede, M., & Sniezek, J. A. (2003). Group judgment processes and outcomes in video- conferencing versus face-to-face groups. International Journal of Human-Computer Studies, 59(6), 875-897. doi: 10.1016/j.ijhcs.2003.07.001 Culnan, M., & Markus, M. L. (1987). Information Technologies. Newbury Park, CA: Sage. Daft, R. L., & Lengel, R. H. (1984). Informa tion Richness - A New Approach to Managerial Behavior and Organization Design. Research in Organizational Behavior, 6, 191-233. Daft, R. L., & Lengel, R. H. (1986). Organiza tionl Information Requirements, Media Richness and Structural Design. Management Science, 32(5), 554-571. 92 Daft, R. L., & Wiginton, J. C. (1979). Language and Organization. The Academy of Management Review, 4(2), 179-191. doi: 10.2307/257772 Dennis, A. R., & Kinney, S. T. (1998). Testing Media Richness Theory in the New Media: The Effects of Cues, Feedback, and Task Equivocality. Information Systems Research, 9(3), 256-274. doi: doi:10.1287/isre.9.3.256 Driskell, J. E., & Mullen, B. (1990). Status, Expectations, and Behavior: A Meta-Analytic Review and Test of the Theory. Personality and Social Psychology Bulletin, 16(3), 541- 553. doi: 10.1177/0146167290163012 Driskell, J. E., Radtke, P. H., & Salas, E. (2003). Virtual teams: E ffects of technological mediation on team performance. Group Dynamics-Theory Research and Practice, 7(4), 297-323. doi: 10.1037/1089-2699.7.4.297 Eastin, M. S., & LaRose, R. (2000). Internet Self-Efficacy and the Psychology of the Digital Divide. Journal of Computer-Mediated Communication, 6 (1), 0-0. doi: 10.1111/j.1083-6101.2000.tb00110.x ElShinnawy, M., & Markus, M. L. (1997). The poverty of media richness theory: Explaining people's choice of electronic mail vs. voice mail. International Journal of Human-Computer Studies, 46(4), 443-467. doi: 10.1006/ijhc.1996.0099 Graetz, K. A., Boyle, E. S., Kimble, C. E., Thompson, P., & Garloch, J. L. (1998). Information Sharing in Face-to-Face, Teleconferencing, and Electronic Chat Groups. Small Group Research, 29(6), 714-743. doi: 10.1177/1046496498296003 Hantula, D. A., Kock, N., D'Arcy, J. P., & DeRosa, D. M. (2011). Media Compensation Theory: A Darwinian Perspective on Adaptation to Electronic Communication and Collaboration. Berlin: Springer-Verlag Berlin. Harter, J. K., Schmidt, F. L., & Hayes, T. L. (2002). Business-unit-level relationship between employee satisfaction, employee engagement, and business outcomes: A meta-analysis. Journal of Applied Psychology, 87 (2), 268-279. doi: 10.1037//0021-9010.87.2.268 Hiltz, S. R., Johnson, K., & Turoff, M. (1986). Experiments in Group Decision Making: "Communication Process and Outcome in Face-to-Face Versus Computerized Conferences". Human Communication Research, 13(2), 225. Howard, M. C. (2014). Creation of a Computer Self-Efficacy Measure: Analysis of Internal Consistency, Psychometric Properties, and Validity. Cyberpsychology Behavior and Social Networking, 17 (10), 677-681. doi: 10.1089/cyber.2014.0255 Joinson, A. N. (2004). Self-esteem, interpersonal risk, and preference for e-mail to face-to-face communication. Cyberpsychology & Behavior, 7(4), 472-478. doi: 10.1089/cpb.2004.7.472 93 Kerr, D. S., & Murthy, U. S. (2009). The effectiveness of synchronous computer-mediated communication for solving hidden-profile problems: Further empirical evidence. Information & Management, 46 (2), 83-89. doi: 10.1016/j.im.2008.12.002 Kiesler, S., Siegel, J., & McGuire, T. W. ( 1984). Social Psychological - Aspects of Computer- Mediated Communication. American Psychologist, 39(10), 1123-1134. doi: 10.1037//0003-066x.39.10.1123 Kiesler, S., & Sproull, L. (1992). Group decision making and communication technology. Organizational Behavior and Human Decision Processes, 52(1), 96-123. doi: http://dx.doi.org/10.1016/0749-5978(92)90047-B Kingston, D., Cindy-Lee, D., & Sword, W. ( 2007). Exploring Breast-feeding Self-efficacy. Journal of Perinatal & Neonatal Nursing, 21 (3), 207. Kirkman, B. L., & Mathieu, J. E. (2005). The di mensions and antecedents of team virtuality. Journal of Management, 31(5), 700-718. doi: 10.1177/0149206305279113 Kock, N. (2004). The psychobiological model: To wards a new theory of computer-mediated communication based on Darwinian evolution. Organization Science, 15(3), 327-348. doi: 10.1287/orsc.1040.0071 Kock, N. (2005). Media richness or media naturalness? The evolution of our biological communication apparatus and its influence on our behavior toward e-communication tools. Ieee Transactions on Professional Communication, 48 (2), 117-130. doi: 10.1109/tpc.2005.849649 Lam, S. S. K., & Schaubroeck, J. (2000). Improving Group Decisions by Better Pooling Information: A Comparative Advantag e of Group Decision Support Systems. Journal of Applied Psychology, 85(4), 565-573. doi: 10.1037//0021-9010.85.4.565 Lu, L., Yuan, Y. C., & McLeod, P. L. (2012). Twenty-Five Years of Hidden Profiles in Group Decision Making: A Meta-Analysis. Personality and Social Psychology Review, 16 (1), 54-75. doi: 10.1177/1088868311417243 Marcus, B. H., Selby, V. C., Niaura, R. S., & Rossi , J. S. (1992). Self-efficacy and the stages of exercise behavior changeResearch Quarterly for Exercise and Sport, 63(1), 60-66. Markus, M. L. (1994). Electronic Mail as the Medium of Managerial Choice. Organization Science, 5(4), 502-527. doi: 10.1287/orsc.5.4.502 Martins, L. L., Gilson, L. L., & Maynard, M. T. (2004). Virtual teams: What do we know and where do we go from here? Journal of Management, 30(6), 805-835. doi: 10.1016/j.jm.2004.05.002 Mesmer-Magnus, J. R., & DeChurch, L. A. (2009). Information Sharing and Team Performance: A Meta-Analysis. Journal of Applied Psychology, 94(2), 535-546. doi: 10.1037/a0013773 94 Mesmer-Magnus, J. R., DeChurch, L. A., Jimenez-Rodriguez, M., Wildman, J., & Shuffler, M. (2011). A meta-analytic investigation of virtuality and information sharing in teams. Organizational Behavior and Human Decision Processes, 115(2), 214-225. doi: 10.1016/j.obhdp.2011.03.002 O'Brien, H. L., & Toms, E. G. (2013). Examining the generalizability of the User Engagement Scale (UES) in exploratory search. Information Processing & Management, 49 (5), 1092-1107. doi: 10.1016/j.ipm.2012.08.005 Piaget, J., & Cook, M. T. (1952). The origins of intelligence in children. New York, NY: International University Press. Rice, R. E. (1993). Media Appropriateness - Using Social Presence Theory to Compare Traditional and New Organizational Media. Human Communication Research, 19(4), 451-484. doi: 10.1111/j.1468-2958.1993.tb00309.x Rice, R. E., & Shook, D. E. (1990). Relationships of Job Categories and Organizational Levels to Use of Communication Channels, Including Electronic Mail - A Metaanalysis and Extension. Journal of Management Studies, 27(2), 195-229. doi: 10.1111/j.1467-6486.1990.tb00760.x Ring, K., Braginsky, D., & Braginsky, B. (1966). Performance styles in interpersonal relations: a typology. Psychological Reports, 18(1), 203-220. doi: 10.2466/pr0.1966.18.1.203 Ring, K., Braginsky, D., Levine, L., & Braginsky, B. (1967). Performance styles in interpersonal behavior: an experimental validation of a typology. Journal of Experimental Social Psychology, 3(2), 140-159. doi: 10.1016/0022-1031(67)90018-2 Rosen, B., Furst, S., & Blackburn, R. (2007). Overcoming Barriers to Knowledge Sharing in Virtual Teams. Organizational Dynamics, 36(3), 259-273. doi: http://dx.doi.org/10.1016/j.orgdyn.2007.04.007 Siegel, J., Dubrovsky, V., Kiesler, S., & McGuir e, T. W. (1986). Group processes in computer- mediated communication. Organizational Behavior and Human Decision Processes, 37(2), 157-187. doi: http://dx.doi.org/10.1016/0749-5978(86)90050-6 Spitzberg, B. H. (2006). Preliminary development of a model and measure of computer-mediated communication (CMC) competence. Journal of Computer-Mediated Communication, 11(2), 41. Spitzberg, B. H. (2011). The interactive medi a package for assessment of communication and critical thinking (IMPACCT©): Testi ng a programmatic online communication competence assessment system. Communication Education, 60(2), 145-173. doi: http://dx.doi.org/10.1080/03634523.2010.518619 Spitzberg, B. H., & Cupach, W. R. (2002). Interpersonal Skills. Thousand Oaks, CA: Sage. 95 Sproull, L., & Kiesler, S. (1986). Reducing Social Context Cues: Electronic Mail in Organizational Communication. Management Science, 32(11), 1492-1512. Stahl, G. K., Maznevski, M. L., Voigt, A., & Jonsen, K. (2010). Unraveling the effects of cultural diversity in teams: A meta-analysis of research on multicultural work groups. Journal of International Business Studies, 41(4), 690-709. doi: 10.1057/jibs.2009.85 Stasser, G., & Titus, W. (1985). Pooling of unshared information in group decision making: Biased information sampling during discussion. Journal of Personality and Social Psychology, 48(6), 1467-1478. doi: 10.1037/0022-3514.48.6.1467 Stasser, G., & Titus, W. (1987). Effects of information load and percentage of shared information on the dissemination of uns hared information during group discussion. Journal of Personality and Social Psychology, 53 (1), 81-93. doi: http://dx.doi.org/10.1037/0022-3514.53.1.81 Straus, S. G., & McGrath, J. E. (1994). Does the Medium Matter - The Interaction of Task Type and Technology on Group-Performance and Member Reactions. Journal of Applied Psychology, 79(1), 87-97. doi: 10.1037/0021-9010.79.1.87 Tidwell, L. C., & Walther, J. B. (2002). Computer-mediated communication effects on disclosure, impressions, and interpersonal evaluations: Getting to know one another a bit at a time. Human Communication Research, 28(3), 317-348. doi: 10.1111/j.1468-2958.2002.tb00811.x Tosun, L. P., & Lajunen, T. (2010). Does Internet use reflect your personality? Relationship between Eysenck's personality dimensions and Internet use. Computers in Human Behavior, 26(2), 162-167. doi: 10.1016/j.chb.2009.10.010 Walther, J. B. (1992). Interpersonal Effects in Computer-Mediated Interaction - A Relational Perspective. Communication Research, 19(1), 52-90. doi: 10.1177/009365092019001003 Walther, J. B. (1996). Compute r-mediated communication: Impersonal, interpersonal, and hyperpersonal interaction. Communication Research, 23 (1), 3-43. doi: 10.1177/009365096023001001 Walther, J. B., & Bunz, U. (2005). The rules of virtual groups: Trust, liking, and performance in computer-mediated communication. Journal of Communication, 55 (4), 828-846. doi: 10.1111/j.1460-2466.2005.tb03025.x Walther, J. B., & Burgoon, J. K. (1992). Relational Communication in Computer-Mediated Interaction. Human Communication Research, 19(1), 50-88. doi: 10.1111/j.1468-2958.1992.tb00295.x