Abstract

This article reports on an experiment investigating the differences between computer-mediated and face-to-face negotiations and between negotiators being deceptive about hidden agendas and negotiators without hidden agendas. Our results supported the hypotheses that individuals negotiating via instant messaging are more likely to use forcing negotiating, experience more tension, and have lower deception detection accuracy than individuals negotiating face-to-face. Unexpectedly, it was found that individuals negotiating via instant messaging were more satisfied with the negotiation process than were face-to-face negotiators. Finally, results supported the hypothesis that those being deceptive about hidden agendas experienced higher tension than those without hidden agendas. These findings have several implications for organizations: higher levels of tension from computer-mediated negotiations and from deception can affect the long-term effectiveness of employees, undetected deception in computer-mediated negotiations can have a negative impact on negotiations, and computer-mediation can lead to the use of a forcing negotiation style, which may improve the effectiveness of negotiators with individualistic goals.

Introduction

Negotiation is a form of decision-making in which two or more independent parties talk with one another in an effort to resolve their opposing interests and make joint decisions (Pruitt, 1981). Many popular press business articles (e.g., Walker, 2003) and books (e.g., Cohen, 1980; Fisher & Ury, 1981; Lax & Sebenius, 1986; Shell, 1999) have noted that every individual is a negotiator, and all individuals negotiate everyday in a variety of situations. In the workplace, negotiations are a regular part of the business day, as multiple employees work together to achieve tasks. Individuals also often negotiate when they have to rely on someone else to achieve their objectives (e.g., a co-worker who requires statistics from another worker in order to complete a project) (Thompson, 2005). Since negotiations are an integral part of work life, it is important for individuals and organizations to understand the impact of different negotiation settings.

Although negotiations are traditionally conducted face-to-face, with the increasing use of technology in the workplace, negotiations are more likely to be conducted via the computer. The use of computer-mediated communication (e.g., email, instant messaging) in negotiations is evident from the fact that there are now numerous electronic negotiation support systems with integrated email and instant messaging technologies (Neumann, Kurtzberg, Thompson, & Morris, 2003) as well as online resolution services that support negotiations using instant messaging (Yuan, Head, & Du, 2003). Furthermore, with increasing globalization, work groups increasingly have become dispersed geographically and therefore must rely on computer-mediated communication to accomplish their tasks.

Computer-mediated communication is an important influence on negotiations, because individuals' communication processes are affected by feedback, communication cues, language variety, and personal focus, which are different in computer-mediated communication than they are in face-to-face communication (Daft, Lengel, & Trevino, 1987). Lean communication media, such as computer-mediated communication, have low levels of these characteristics. This can be problematic because communication may be less efficient. For instance, communication partners can be difficult to understand because many visual cues, such as facial expressions, are missing. Research has begun to examine the effects of computer-mediated communication during negotiations; however, there have been mixed findings about the effectiveness of these negotiations (Arunachalam & Dilla, 1995; Rangaswamy & Shell, 1997) leaving researchers with an incomplete understanding in this area.

Deception, another important influence on negotiations, is also affected by the communication medium (Zmud, 1990). The prevalence of deception in business settings is evident from the fact that 25% to 67% of applicants falsify their resumes and attempt to justify those falsifications in job interviews (Prater & Kiser, 2002). Individuals are often deceptive when they have goals different from the individuals with whom they are working (Grover, 1993). Given that negotiations are defined as efforts at solving opposing interests, it is likely that deception could play a prominent role. Deception can be hard to detect in computer-mediated settings, because many cues to deception, such as posture shifts, are not transmitted in most types of computer-mediated communication (Rao & Lim, 2000). Unfortunately, deception in computer-mediated negotiations has not been thoroughly investigated. Because the use of computer-mediated communication is likely to increase in the workplace as computers become more prevalent and work teams become more geographically dispersed, it is of great importance to understand negotiation influences, such as deception, that are present in these settings.

This article investigates several outcomes related to deception and computer-mediated communication. We explore whether negotiators' tension, deception detection, satisfaction, and forcing communication style differ based on the communication medium (face-to-face vs. computer-mediated). Further, we examine differences in tension based on whether negotiators are being deceptive about hidden agendas.

First, the literature on negotiations is reviewed with primary emphasis placed on the differences between computer-mediated versus face-to-face communication. Next, the literature on deception and deception detection is reviewed and five research hypotheses are developed. The experimental procedures are then explained, along with the measures and statistical analyses used. Finally, the results are discussed, and the contributions of the study and future research ideas are identified.

Literature

Computer-Mediated vs. Face-to-Face Negotiations

Even though computer mediation is a relatively new area in negotiation research, many classic studies have investigated negotiations that took place in different communication settings. These studies looked at differences in negotiations conducted in settings where subjects could hear and see each other, could only see each other, or could only hear each other. Researchers found that negotiators find it difficult to identify (Pruitt & Carnevale, 1993) and reach (Carnevale & Isen, 1986; Carnevale, Pruitt, & Seilheimer, 1981) integrative solutions in face-to-face settings. Furthermore, negotiators without visual access to each other have a higher level of cooperation than do those with visual access (Lewis & Fry, 1977). In addition to a higher level of cooperation, researchers found that negotiators without visual access have more integrative outcomes than do negotiators with visual access to each other (Carnevale & Isen, 1986; Carnevale, et al., 1981). However, researchers have found that when both audio and visual access are eliminated, leaving only text-based communication, the level of cooperation is lower (Wichman, 1970), and bargaining effectiveness is likely to suffer (Rubin & Brown, 1975). Therefore, text-based computer-mediated communications, such as email and instant messaging, may hinder cooperative negotiations.

Several recent studies have expanded on this classic research and have examined the influence of computer-mediated communication on negotiation outcomes. Unfortunately, the findings have been mixed. Researchers predicted that lean, computer-mediated communication would not be a good fit for tasks in which individuals' motives differ (i.e., mixed motive) because of the lack of communication cues (McGrath & Hollingshead, 1993). In empirical studies, researchers found that computer-mediation led to less cooperative outcomes (Arunachalam & Dilla, 1995), and that that richer media (e.g., face to face) led negotiators to perceive their partners as using more of a collaborative bargaining approach and less of a competitive bargaining approach than did users of leaner media (Purdy, Nye, & Balakrishnan, 2000).

However, others found different results. One study found no differences in the integrative nature of outcomes across media (Rangaswamy & Shell, 1997). The differences in this study's findings may be due to the type of communication technology used in the experiment. The study used an email system, which may have been more inefficient than the instant messaging systems used in the studies that found differences in the integrative and collaborative nature of the negotiations (Arunachalam & Dilla, 1995; Purdy, et al., 2000). A communication system with a significant amount of lag between the sending and receiving of messages (such as many email systems) could cause negotiators to give up on trying to get the best deal for themselves and simply try to reach agreement, since they are unable to respond efficiently to the other party and argue their points.

Yet another study found that individuals negotiating via a computer are more likely to use avoiding behaviors and less likely to use forcing behaviors than individuals negotiating face-to-face (Dorado, Medina, Munduate, Cisneros, & Euwema, 2002). The differences in this study's findings may have been due to the fact that one of the negotiators in the experiment was a confederate who used conflict escalating tactics, including personal attacks. In a situation where a negotiator attacks the other party personally, it would make sense that the other party would become agitated in a face-to-face setting and would want to defend his or her point of view forcefully. The same person would also be less likely to become agitated in a computer-mediated setting, since this type of communication is less personal. The mixed nature of these findings should become clearer as more empirical studies are conducted.

Deception and Deception Detection

A prevalent and potentially harmful influence in business negotiations is deception. Deception is a message knowingly transmitted by an individual to foster a false belief or conclusion in others (Buller & Burgoon, 1996). Deception in business settings often comes from individuals with hidden agendas. These individuals foster a false belief in others that they are indeed working toward a common goal, when in actuality, they have their own goals and agendas. When individuals have an agenda that conflicts with the goals of those with whom they work, they often use deception to rectify this divergence (Grover, 1993). Unfortunately, the communication literature on deception shows that the average person is not very good at detecting deception. Experiments on deception detection have reported either chance or lower than chance rates of deception detection accuracy (Feeley, DeTurck, & Young, 1995; Zuckerman & Driver, 1985). The truth bias, which is individuals' natural tendency to believe communication is truthful, can help explain these results. Individuals who initially believe others are truthful will often not recognize deception (McCornack & Levine, 1990; McCornack & Parks, 1986; Miller & Stiff, 1993).

According to the interpersonal deception theory, deceptive communication is an ongoing, interactive process between a deceiver and the person being deceived (i.e., the receiver) (Buller & Burgoon, 1996). Producing and defending deceptive messages is a cognitively intense mental process (Miller & Stiff, 1993), and as a result deceivers often leak cues to deception (Ekman, 1992). These cues can be verbal and nonverbal. Examples of verbal cues of deception include the number of statements of personal responsibility, other responsibility, and mutual responsibility and the number of factual and opinion statements. Examples of non-verbal cues of deception include number of eye blinks, smile duration, posture shifts, pauses, and response length (Miller & Stiff, 1993). In the deception process, deceivers make strategic changes to their behavior to hide cues to deception, such as eye movements and posture, and receivers, if they become suspicious, question deceivers in an attempt to better understand cues to deception (Buller & Burgoon, 1996). This process continues through the communicative event.

The communication medium can influence the outcomes of the deceptive communication process, because different communication media lead to varying degrees of cue leakage (Miller & Stiff, 1993). Media richness theory explains that different communication media have varying degrees of richness because of differences in feedback, cues, language variety, and personal focus (Daft & Lengel, 1986). Face-to-face is the richest media, whereas unaddressed written communication is the least rich media. Communication over lean computer-mediated channels such as email and instant messaging eliminates many communicative cues, such as tone of voice, facial expression, and posturing. Because of the limited number of cues available, deception may be much more difficult to detect in these settings. Recent research has recognized the importance of understanding deception in computer-mediated technologies such as instant messaging. Recent studies found that deceivers in computer-mediated communication tend to have more initiation, non-immediacy, and spontaneous correction, and tend to give more details than truth tellers (Qin, Burgoon, Blair, & Nunamaker, 2005; Zhou, 2005).

Based on this literature, we hypothesize that the communication medium will influence deception detection during negotiations. Because of the relatively limited number of cues to deception available in instant messaging (Miller & Stiff, 1993), individuals communicating via instant messaging will have a harder time realizing when their negotiation partners are deceptive about hidden agendas than will individuals communicating face-to-face. Further, individuals in cognitively taxing situations are more likely to leak cues (Miller & Stiff, 1993), and a face-to-face negotiation setting is more taxing than an instant messaging negotiation setting. Negotiations, particularly when they are taking place in a limited timeframe, require constant back and forth communication. In a face-to-face setting, this communication process is fast because the transmission process is fast. Further, the communication process is more taxing during face-to-face interactions because parties encounter uncomfortable dead time when not communicating. Conversely, text-based computer-mediated settings are less taxing since negotiators can read and transmit messages at their own pace without direct pressure from the other party. The cognitive demands in a face-to-face negotiation setting will cause deceivers to leak more deception cues than in a computer-mediated setting. The negotiation partners of deceivers have to deal with more background noise (non-deceptive cues) in face-to-face settings. However, in face-to-face, if the negotiators become suspicious about a single one of the many deception cues that are leaked and transmitted, they will be more likely to scrutinize the other cues. We therefore hypothesize that individuals negotiating over lean media will have lower levels of deception detection than will individuals using rich media.

Negotiation Style

Many negotiation researchers categorize individuals' negotiation tactics into two categories: integrative tactics and distributive tactics. Integrative tactics are concerned with tradeoffs and fulfilling the interests of all the negotiators (Pruitt, 1981). Distributive tactics are individualistic, used to get concessions from the other party, and concerned with getting resources on only one side of the negotiation (Pruitt, 1981).

Because many negotiations involve conflict, another approach to understanding negotiation tactics is to examine how individuals typically behave during conflict situations. Researchers developed a scale to measure the various ways in which individuals handle conflict in workplace situations (Van de Vliert, 1997). Van de Vliert (1997) contends that there are five ways that individuals handle conflict at work: problem solving, forcing, yielding, avoiding, and compromising. The first strategy, forcing, is concerned with imposing one's will on others, and it involves threats and bluffs, persuasive arguments, and positional commitments. The second strategy, yielding, involves accepting the will of others, concessions, unconditional promises, and offering help. The third strategy, avoiding, is concerned with lowering the importance of the issues, and it involves attempts to suppress thinking about the issues. The fourth strategy, problem solving, entails satisfying both parties as much as possible, and it involves an exchange of information about priorities and preferences and making trade-offs between issues that are important and unimportant. The fifth strategy, compromising, involves making conditional promises and threats and an active search for middle ground.

Forcing negotiation behavior is particularly important in situations where negotiators have individualistic goals (i.e., hidden agendas). Researchers have found that forcing behavior is negatively related to integrative outcomes in negotiations (Beersma & De Dreu, 1999); therefore, individuals who use a forcing style will be more likely to satisfy their goals in the negotiation than those with a non-forcing style. Empirical research has explored the differences in conflict-handling communication styles based on the type of communication medium. Researchers have found that computer-mediated negotiators use less of a collaborative negotiating approach and more of a competitive negotiating approach than do face-to-face negotiators (Arunachalam & Dilla, 1995; Purdy, et al., 2000; Rhee, Hasan, Jacob, & Barhki, 1995).

Even though others have reported conflicting findings regarding individuals' negotiation behavior in computer-mediated and face-to-face communication settings (Dorado, et al., 2002), we feel that there is enough evidence to hypothesize that negotiators in computer-mediated settings should be more forcing. As previously explained, the mixed nature of the findings may have been due to the inefficiency of the communication technologies and extreme negotiation behavior that were used in those studies. Computer-mediated communication already has been linked to depersonalization and self-focused communication (Kiesler, Siegel, & McGuire, 1984). Further, classic negotiation research found that when audio access and visual access are eliminated, individuals tend to cooperate less (Wichman, 1970). Last, researchers have found that computer-mediation leads to less integrative and collaborative behavior (Arunachalam & Dilla, 1995; Purdy, et al., 2000; Rhee, et al., 1995), and forcing behavior has been found to be related negatively to integrative outcomes in negotiations (Beersma & De Dreu, 1999). Thus, we posit that a forcing style of communication will be more evident in computer-mediated negotiators than in face-to-face negotiators.

Satisfaction

In this article, we refer to satisfaction as individuals' evaluation of whether or not their goals were met by the negotiation outcome (Oliver, Balakrishnan, & Barry, 1994). Researchers have suggested that negotiator satisfaction, measured with a post-negotiation questionnaire, can be an important negotiation outcome variable (Dwyer & Walker Jr., 1981). Unfortunately, few studies have investigated negotiation satisfaction in computer-mediated settings. Researchers (Purdy, et al., 2000) found that individuals who negotiated using rich communication media had greater desire for future negotiation interactions, which indicates that they were satisfied. Further, individuals communicating over rich media can communicate more efficiently, and they tend to have more integrative negotiation outcomes (Arunachalam & Dilla, 1995; Purdy, et al., 2000; Rhee, et al., 1995), which may lead to more satisfaction with the negotiation. Thus, we propose that individuals negotiating face-to-face (i.e., a rich communication channel) will have higher satisfaction with the negotiation process than individuals negotiating via a computer (i.e., a lean communication channel).

Tension

Tension refers to the internal feelings of discomfort one experiences during a stressful situation. Prolonged tension experienced by individuals can have devastating effects. For instance, Dana and Griffin (1999) note that subjective well-being (of which tension is one indicator) has bottom line implications for organizations. Employees, for example, can cost organizations money by having higher health care costs manifested from prolonged tension. Additionally, the indirect effects of workplace tension such as accidents, absenteeism, and lost productivity also cost organizations a great deal of money (Cooper & Cartwright, 1994). Furthermore, organizations are increasingly being held financially responsible for tension-related claims such as cardiovascular disease, resulting in a considerable increase in workers compensation claims (Dana & Griffin, 1999).

Pertaining to tension levels during negotiations, little research has examined the effect of communication medium. However, using a text-based communication medium is not as efficient for a complex task such as negotiation (McGrath & Hollingshead, 1993). It takes less time to listen to a person than it does to read a person's message, and furthermore, it takes longer to respond via writing or typing than it takes to speak a response to a person face-to-face. Last, it takes longer to receive clarification if the message is not understood. In order to clarify a written message via text messaging, a person would have to first read the message, type a response to the message, wait for the partner to respond to the message, and finally read the response. With face-to face communication, the person can immediately receive clarification by asking the other person on the spot.

Moreover, visual and audio communication cues are lacking in text-based communication (Daft & Lengel, 1986). Negotiators in text-based communication settings should have a more difficult time ascertaining the emotional content of messages, since they cannot look at and listen to their partners. This could lead to more frequent misunderstandings than would be found in face-to-face communication. The lack of visual cues and extra time it takes to communicate via text-based messaging is likely to lead to increased tension. Therefore, we posit that negotiators communicating using a computer will experience more tension than negotiators communicating face-to-face.

We lastly hypothesize that individuals with hidden agendas will experience more tension during negotiations than will individuals without hidden agendas. The demands from being deceptive about an agenda that is opposite to the agenda of the other party will cause deceptive individuals to have significantly more tension than individuals without a hidden agenda. Producing and defending deceptive messages is a cognitively intense mental process and requires more effort than producing truthful messages (Miller & Stiff, 1993). We therefore predict that deception stemming from a hidden agenda will lead to tension.

Methods

Research Design

We used a 2x2 factorial design for this study, with the independent variables of role (deceiver vs. receiver) and communication medium (face-to-face vs. computer). For role, we randomly assigned half of the participants to the role of the deceiver. For communication medium, half of the participants were assigned randomly to communicate using Microsoft Instant Messenger. The other half communicated face-to-face. Participants were assigned randomly to one of the four treatments, and there were 28 participants per treatment cell.

Participants and Experimental Procedure

We ran the experiment testing 132 undergraduate business students at a large Southeastern university in a simulated negotiation previously used in another study (Thompson, 1990). The average age of the participants was 21.52 years. With respect to demographics, 56 (50%) of the participants were male, and 56 (50%) were female. All participants were given class credit for participating. In order to motivate the participants to perform well, they were told that they would have a chance to win a $25.00, $50.00, or $100.00 prize based on their individual performance in the negotiation. Approximately 30 subjects (15 dyads) were tested in each experimental session. Each deceiver was assigned the role of an employer, and the receiver was assigned the role of a new employee at the employer's company.

At the beginning of each session, receivers (employees) and deceivers (employers) were moved into separate rooms and were briefed about their task. All participants were told that they would be negotiating the starting salary, vacation time, raise schedule, starting date, and medical benefits for an employment contract (Thompson, 1990). We picked this task because an employment contract is a relevant negotiation topic for undergraduate business students and because it is a situation where they would likely not be expecting deception.

Each employment contract item had an assigned point value for each possible agreement level. Both deceivers and receivers were given point schedules that listed the points that they would receive for each agreement value (see Tables 1 and 2). As Tables 1 and 2 show, the task contained two purely fixed-sum issues (vacation, annual raise), two issues that could be traded off (salary, medical coverage), and one compatible issue (starting date) for which bargainers had identical preferences.

The employers were told that they would have 20 minutes to reach agreement on the five points of the negotiation. Additionally, they were informed that an agreement must be reached on one value listed in their payoff schedule for each of the negotiation points. In addition to their own payoff schedules, employers (i.e., the deceivers) were given the payoff schedule of their negotiation partner (the employee). Employers were told that their goal was to maximize their own scores and to minimize the scores of the employees. Their overall scores in the negotiation would be calculated as their point total minus the point total of their partner. It was stated that the employers that received the three best scores would win $25.00, $50.00, and $100.00. The employers were instructed to be deceptive by getting their negotiation partners to think they (the deceivers) only had their own payoff schedules and that they were only attempting to maximize their own scores.

Employees were given similar instructions, but they were not given the payoff schedules for their partners. They were told that their goal was to maximize their scores in the negotiation by getting their partners to agree on outcomes that benefited themselves the most. They were also told that they could not show their payoff tables to their partners.

After the subjects were given instructions, half of the employers and half of the employees were moved to a room in which a number of computers were set up. Each participant was at a separate workstation and was paired with someone else in the room. Participants were unable to see the other participants' screens, and they did not know with whom they were communicating. Participants had a copy of their payoff schedules at their workstation, and they communicated using Microsoft Instant Messenger.

The other half of the participants performed the negotiation face-to-face in a different room. To control for relational closeness influences, which can cause individuals to believe that others are honest (Miller & Stiff, 1993), participants were paired with an individuals with whom they were not familiar. Participants had a copy of their payoff schedule(s), which they were not allowed to show to their partner. We reduced the risk of interference between the face-to-face groups by separating the groups as much as possible and asking the groups to keep their conversation at a low volume level. We also asked participants to ignore the other groups as much as possible. We conducted a pilot study to test this procedure, and no groups reported interference problems with other groups.

Prior to the negotiation task, the participants were given a survey containing the informed consent form, the political skill inventory (Ferris, Treadway, Kolodinsky, Hockwarter, Kacmar, Douglas, et al., 2005), and experience with electronic messaging questions (Carlson & Zmud, 1999). After the negotiation task was concluded, the participants were given a follow-up survey containing the perceived deception (only for the employees), forcing communication style, satisfaction, tension, and demographics questions.

Measures

Independent Variables

The independent variables for this study are the experimental manipulations of the role (i.e., employer-deceiver and employee-receiver) and the communication medium (face-to-face and computer-mediated).

Dependent Variables

The dependent variables for this study are the use of a forcing communication style, negotiation satisfaction, tension, and deception detection accuracy. Individuals' perceived level of forcing communication was measured with a section of the Dutch Test for Conflict Handling (DUTCH) (Van de Vliert, 1997). The scale measures individual's perceived conflict management strategies. Individuals who use forcing communication style attempt to impose their will on others through the use of threats and bluffs, persuasive arguments, and positional commitments (Van de Vliert, 1997).

We used the four forcing items from the DUTCH scale that measures individuals' perception of their typical conflict management strategies (De Dreu, Evers, Beersma, Kluwer, & Nauta, 2001) to ask how they managed the conflict in the previous negotiation (α=.80). The items included "I pushed my own point of view," "I searched for gains," "I fought for a good outcome for myself," and "I did everything to win." The items were measured on a 7-point Likert scale ranging from 1 (strongly disagree) to 7 (strongly agree).

Negotiation satisfaction was measured using a Graham, Mintu, and Rodgers' (1994) 4-item scale (α=.91). The items included "How satisfied were you with the agreement you reached?" "How satisfied were you with the agreement relative to your pre-negotiation expectations?" "How satisfied were you with your individual score in the negotiation?" and "How satisfied were you with your performance during the negotiation?" These items were measured on a 7-point Likert scale ranging from 1 (satisfied) to 7 (dissatisfied).

Tension was measured by a scale adapted from the Stress Response Inventory (α=.92) that was developed previously and validated (Koh, Park, Kim, & Cho, 2001). The scale was designed to measure several dimensions of stress responses such as tension, aggression, anger, and frustration. Ten items were selected strategically and adapted to reflect tension experienced during negotiations. The items included "During the negotiation exercise, I was under great deal of tension," "I felt fidgety or nervous during the negotiation exercise," "During the negotiation, I felt like screaming," "During this negotiation, I was tense," "During the negotiation exercise, my face became flushed or felt hot," "I felt myself becoming angry while participating in the negotiation exercise," "My body was trembling during the negotiation exercise," "I did not feel like communicating during this negotiation exercise," "I often sighed during this negotiation," and "I was very bothered during this negotiation exercise." These items were measured on a 7-point Likert scale ranging from 1 (strongly disagree) to 7 (strongly agree).

We measured deception detection using a three item scale (α=.92) previously used in the e-commerce literature (Grazioli & Jarvanpaa, 2000). The participants without a hidden agenda were asked to evaluate on a 7-point scale the extent to which they felt that the information communicated from their partner was "accurate" versus "misleading," "truthful" versus "deceptive," and "factual" versus "distorted."

Control Variables

We collected data for political skill, experience with electronic messaging, and negotiation score to control for the potential individual difference influences on our outcome variables.

Political skill is a measure of social effectiveness in workplace situations that closely relates to negotiation skill. Political skill measures an individual's ability to identify with others and to obtain goals by presenting one's behavior in the best possible light. Furthermore, politically skilled individuals have the ability to adapt and calibrate their behaviors to situations in order to elicit desired responses from others. These individuals are able to develop and use diverse networks of people and appear as though they possess high levels of integrity, authenticity, sincerity, and genuineness (Ferris, et al., 2005).

Political skill was measured using Ferris, et al.'s (2005) 18-item political skill inventory (α=.91). Example items include, "I spend a lot of time and effort at work networking with others," and "I am able to make most people feel comfortable and at ease around me." These items were measured on a 7-point Likert scale ranging from 1 (strongly disagree) to 7 (strongly agree).

Researchers have suggested that media richness perceptions and communication abilities can change over time. As participants develop experience with each other, the channel, the message topic, and the communication context, they will perceive the channel as being better able to handle rich messages and will potentially be able to communicate richer messages (Carlson & Zmud, 1999). For example, communicators can learn to use slang and shorthand to communicate with those they know well and over media they have used many times.

Participants' experience with electronic messaging was measured using an established and validated multi-item scale (Carlson & Zmud, 1999) (α=.65). Participants were asked to rate their agreement with statements such as "I am very experienced using electronic messaging" and "I feel that electronic messaging is easy to use" on a 7-point Likert scale ranging from 1 (strongly disagree) to 7 (strongly agree). The term "electronic messaging" replaced the term "email" and the term "group members" replaced "communication partner" from the original scales. The term "electronic messaging" was explained to participants as consisting of email, instant messaging, and any other tools that they have used to communicate using a computer.

The negotiation score was calculated as the percentage of possible total points that the negotiator scored in the task.

Results

Correlation Analysis

We first conducted a correlation analysis to determine if political skill, experience with electronic messaging, or the negotiation score were related to our dependent variables (see Table 3). Negotiation score was significantly correlated with the two dependent variables, negotiation satisfaction (.371, p<0.01) and forceful negotiation style (.322, p<0.01); however, none of the other control variables tested were correlated with the dependent variables. Therefore, in the following analyses only the negotiation score was used as a covariate.

Hypothesis Tests

All but the first hypotheses were tested using a MANOVA with one covariate, the negotiator score, and three dependent variables: satisfaction, tension, and use of a forcing communication style during the negotiation. First the F-test (Wilks' Lambda) for our covariate was examined. The overall F-test for negotiation score was significant (F(3,105)=5.10, p<.003). Individual F-tests revealed that negotiation score was significantly related to negotiation satisfaction (F(1,107)=13.31, p<.001) and the use of a forcing communication style (F(1,107)=8.60, p<.005).

Next, the overall F-test for the independent variables was examined. The overall F-test for the communication medium was significant (F(3,105)=3.94, p<.011); however the overall F-test for negotiator role was not significant (F(3,105)=1.11, p<.352). Thus, hypothesis four was not supported.

Given that the overall F-test for the communication medium was significant, we examined the individual F-tests for satisfaction, tension, and the use of a forcing communication style to test our first three hypotheses. There was a statistically significant effect for tension (F(1,107)=4.22, p<.043), satisfaction (F(1,107)=4.18, p<.044), and the use of a forcing negotiation style (F(1,107)=6.75, p<.012). Negotiators communicating via the face-to-face channel had less tension (mean=20.57, s.d.=8.815), satisfaction (mean=18.59, s.d.=4.759), and forcing negotiating (mean=19.98, s.d.=3.859) than did negotiators communicating with instant messaging (tension mean=24.49, s.d.=10.904; satisfaction mean=20.36, s.d.=4.132; forcing mean=21.67, s.d.=2.936). These results supported hypothesis one and hypothesis three. However, the results for satisfaction were the opposite of what we hypothesized, so there was no support for hypothesis two.

To test hypothesis one, another analysis was conducted to examine differences between the deception detection accuracy of negotiators who did not have a hidden agenda in both communications settings. Only the deception detection of individuals playing the roles of employees in the experiment was examined because their communication partners were manipulated to be deceptive. We found that there was a significant difference in the hypothesized direction (t=1.960, p<.028). Specifically, receivers communicating face-to-face had higher deception detection accuracy (mean=15.93, s.d.=3.078) than did receivers communicating with instant messaging (mean=14.14, s.d.=3.709). Overall, 9% of negotiators correctly labeled their partners as deceivers.

Discussion

Our results supported four of our five hypotheses. We found that individuals negotiating using instant messaging used more of a forcing communication style and had more tension than did individuals negotiating face-to-face. We also found that individuals who were deceptive about a hidden agenda had more tension during the negotiation than did individuals without a hidden agenda. Furthermore, individuals communicating face-to-face had higher deception detection accuracy than did individuals communicating using instant messaging. In contrast to hypothesis three, we found that individuals negotiating using instant messaging had higher satisfaction than those negotiating face-to-face.

A first finding was that negotiators using instant messaging perceived themselves as having more of a forceful negotiation style than did face-to-face negotiators. We expected this result, since computer-mediated communication has been linked to depersonalization and self-focused communication (Kiesler, et al., 1984). Individuals using instant messaging were likely to be less inhibited and fight harder for their own goals since they did not feel the direct presence of their negotiation partner. This finding was also supported by the fact that forcing behavior has been found to be related negatively to integrative outcomes in negotiations (Beersma & De Dreu, 1999), and previous studies have found that the elimination of visual and audio access can lead to less integrative negotiation outcomes (Arunachalam & Dilla, 1995; Purdy, et al., 2000; Rhee, et al., 1995). We also found that the use of a forcing communication style was correlated positively with an individual's negotiation score in the experiment. This finding is important, because it could mean that individuals are more willing to fight for their goals in computer-mediated negotiations than they are in face-to-face negotiations, and that a forceful style might lead to better individual performance in the negotiation.

Unexpectedly, we found that computer-mediated negotiators were more satisfied than were face-to-face negotiators. We expected that individuals would be more satisfied with face-to-face negotiation since face-to-face negotiators have been found to have greater desire for future negotiation interactions (Purdy, et al., 2000). This finding could be explained by the fact that individuals were overwhelmed with cue management in face-to-face negotiations. Deceivers likely were frustrated as they attempted to minimize their cue leakage in the face-to-face setting, and their partners may have been confused by the deceptive cues they were receiving (especially if they did not realize their partner was being deceptive), and this could have lowered both parties' negotiation satisfaction. Furthermore, individuals using computer-mediated negotiation did not have to worry as much about visual cue leakage. That is, because the deceivers were not face-to-face, they did not have to worry about hiding the physical cues to deception.

However, this finding could also have stemmed from the fact that negotiators communicating using instant messaging thought they had a better outcome relative to their partner, because they had incorrect perceptions of their partners' goals. With the absence of many communication cues, it was likely to be much harder to read the reaction of the other party when decisions were reached. Therefore, negotiators using instant messaging might have thought that their partners were unhappy with their outcomes and thus would likely be happier with their own outcomes. We also found that satisfaction was correlated significantly with the use of a forcing negotiation style. This could mean that individuals' satisfaction could have also come from feelings that they put forth a good effort and adequately pushed for their preferred outcome.

Additionally, we found that negotiators using instant messaging had more tension than did face-to-face negotiators. We expected this outcome because text-based communication media is not the most efficient way to perform a complex task such as a negotiation (McGrath & Hollingshead, 1993). In instant messaging, there is a lag time between one's own communication and the partner's response. The partner has to read the communication and then type his or her answer. This takes significantly more time than a face-to-face encounter, thus it is inefficient. Furthermore, the negotiators using instant messaging lacked important visual cues from their partners. They could not see how their suggestions were taken because they were not face-to-face. This could leave the instant messaging negotiators feeling as if they did not have all of the information necessary to negotiate effectively with their partners. These influences likely caused tension in the negotiators using instant messaging. Furthermore, this frustration with the negotiation process likely led to negative perceptions of their negotiation partners (e.g., why are they taking so long to respond). It has been found that individuals using computer-mediated communication develop more negative perceptions of their partners than do individuals communicating face-to-face (Fischer-Lokou & Gueguen, 2001; Fischer-Lokou, Gueguen, & Lepy, 2004; Purdy, et al., 2000). These perceptions could have also led to more tension during the negotiation process.

We also found that individuals being deceptive about a hidden agenda had more tension than did individuals without a hidden agenda. This outcome was expected given deception is a cognitively intense process (Buller & Burgoon, 1996), and it is recognized that producing and defending deceptive messages requires more effort than producing truthful messages (Miller & Stiff, 1993). This effort, on top of the effort put into the negotiation process, likely led to more tension.

Last, we found that negotiators communicating face-to-face were more able to recognize deception than were negotiators communicating with instant messaging. This was expected because more cues to deception are available in face-to-face communication than in computer-mediated communication (Miller & Stiff, 1993); therefore, deception should be much easier to detect in face-to-face settings. Overall, only 9% of the negotiators correctly identified their partners as deceptive. The low deception rate was likely the result of the truth bias, which is individuals' natural tendency to believe that others are telling the truth (McCornack & Levine, 1990). The truth bias is especially powerful in situations where individuals are not warned about deception, such as the setting of this experiment.

Our results have several implications for negotiators and organizations. One implication stems from our findings regarding tension. We found that individuals negotiating with instant messaging had more tension than did individuals communicating face-to-face, and that deceptive individuals had more tension than did individuals without a hidden agenda. Organizations need to recognize that individuals negotiating using instant messaging are likely to experience tension and that this tension could affect other aspects of their workplace effectiveness. As noted earlier, tension can have a bottom line effect on organizations, causing increased absenteeism and accidents (Cooper & Cartwright, 1994). Organizations might want to encourage individuals to use multiple communication channels when negotiating in order to minimize the amount of tension coming from computer-mediated negotiations. Further, organizations should recognize that deceptive individuals are more likely to experience tension. Deception is difficult to detect, especially over computer-mediated communication channels, and tension might be a potential indicator of deception. However, tension can come from many sources, and so its reliability as an indicator of deception might be limited.

Furthermore, organizations need to realize that individuals conducting negotiations using instant messaging are more susceptible to deception than are individuals communicating face-to-face. Deceptive individuals who have goals that differ from those of the organization will likely have a negative impact on the organization. This is especially problematic, because we found that individuals communicating using instant messaging also tend to be more satisfied with their negotiations than are face-to-face communicators. Satisfied individuals may not worry about their weaknesses, and they may not attempt to review a negotiation where deception was present, or they may not attempt to overcome future deception.

Last, organizations should recognize that individuals using instant messaging might be more forceful in their negotiations than individuals negotiating face-to-face. This is important, because we found that a forceful negotiation style was correlated with individuals' score in the negotiation. Individuals who are trying to negotiate the best deal possible for themselves may have a better outcome when they are forceful, which may be more likely to happen in computer-mediated settings. Forceful negotiating can be problematic because integrative agreements are harder to reach when individuals are being forceful (Beersma & De Dreu, 1999). However, we found that individuals are able to reach agreement, even when being forceful, when they are given a time limit in which to reach agreement.

Limitations

When interpreting the results of this study, it is important to remember that we looked at dyadic negotiations where one party was deceptive about a hidden agenda that was the opposite of the agenda of the other party. Some of the tendencies identified in this study that affect the cognitive load of the negotiators might not be present in negotiations with more than two parties or in negotiations where one party is not being deceptive about a hidden agenda.

A second limitation of this study is the use of student subjects. Undergraduate students have a different level of experience and different levels of motivation than do organizational employees. Although we controlled for their motivation by offering performance incentives and collected data to control for the political skill and experience using electronic messaging of the subjects, other differences could still be a factor in the results of this study.

A third limitation is the use of subject perceptions of the tension and negotiation style in this study. We had subjects rate their tension and level of forcing communication after the completion of their negotiation. Their perceptions could have been affected by individual influences, and it should be noted that the results might be different if these variables were measured using another method.

Future Research

This study is an attempt to better understand how one type of computer-mediated communication and one form of deception affect negotiations outcomes. Many other types of computer-mediated communication and deception exist. Future studies need to examine negotiations over email and over the many web-based conferencing applications that are now being used in organizational settings. Moreover, these studies could move beyond dyadic pairs to larger groups of negotiators.

Future studies may also want to investigate other negotiation styles. We looked at forcing communication because it is related to many of the tendencies that individuals have when they communicate using computers. Other tendencies need to be identified and linked to computer-mediation and deception. These tendencies might include problem solving, yielding, avoiding, and compromising behavior, all of which might be influential in integrative negotiations.

Conclusions

References

Arunachalam, V., & Dilla, W. N. (1995). Judgment accuracy and outcomes in negotiation: A causal modeling analysis of decision-aiding effects. Organizational Behavior and Human Decision Processes, 61 (3), 289-304.

Beersma, B., & De Dreu, C. (1999). Negotiation processes and outcomes in prosocially and egotistically motivated groups. International Journal of Conflict Management, 10 (4), 385-402.

Buller, D., & Burgoon, J. (1996). Interpersonal deception theory. Communication Theory, 6 (3), 203-242.

Carlson, J., & Zmud, R. (1999). Channel expansion theory and the experimental nature of media richness perceptions. Academy of Management Journal, 42 (2), 153-170.

Carnevale, P. J., & Isen, A. M. (1986). The influence of positive affect and visual access on the discovery of integrative solutions in bilateral negotiation. Organizational Behavior and Human Decision Processes, 37 (1), 1-13.

Carnevale, P. J., Pruitt, D. G., & Seilheimer, S. D. (1981). Looking and competing—Accountability and visual access in integrative bargaining. Journal of Personality and Social Psychology, 40 (1), 111-120.

Cohen, H. (1980). You Can Negotiate Anything. Secausus, NJ: Lyle Stuart.

Cooper, C., & Cartwright, S. (1994). Healthy mind; healthy organizations—A proactive approach to occupational stress. Human Relations, 47 (4), 455-471.

Daft, R., & Lengel, R. (1986). Organizational information requirements, media richness, and structural design. Management Science, 32 (5), 554-570.

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

Dana, K., & Griffin, R. (1999). Health and well being in the workplace: A review and synthesis of the literature. Journal of Management, 25 (3), 357-384.

De Dreu, C., Evers, A., Beersma, B., Kluwer, E., & Nauta, A. (2001). A theory based measure of conflict management strategies in the workplace. Journal of Organizational Behavior, 22 (6), 645.

Dorado, M., Medina, F., Munduate, L., Cisneros, I., & Euwema, M. (2002). Computer-mediated negotiation of an escalated conflict. Small Group Research, 33, 509-524.

Dwyer, F. R., & Walker Jr., O. (1981). Bargaining in an asymmetrical power structure. Journal of Marketing, 45 (000001), 105-115.

Ekman, P. (1992). Telling Lies: Clues to Deceit in the Marketplace, Politics, and Marriage (Vol. 2). New York, NY: W.W. Norton.

Feeley, T., DeTurck, M., & Young, M. (1995). Baseline familiarity in lie detection. Communication Research, 12 (2), 160-169.

Ferris, G. R., Treadway, D. C., Kolodinsky, R. W., Hockwarter, W. A., Kacmar, C. J.,

Douglas, C., et al. (2005). Development and validation of the political skill inventory. Journal of Management, 31 (1), 126-152.

Fischer-Lokou, J., & Gueguen, N. (2001). Impact of mediator, mutual representation of the negotiators and decision making in a dyad: Evaluation in the case of computer-mediated communication. Studia Psychologica, 43 (1), 13-21.

Fischer-Lokou, J., Gueguen, N., & Lepy, N. (2004). Effects of communication on information networks versus face-to-face on the reciprocal representation of negotiators. Bulletin de psychologie, 57 (5), 525-533.

Fisher, R., & Ury, W. (1981). Getting to Yes. Boston, MA: Houghton Mifflin.

Graham, J., A. Mintu, & W. Rodgers, W. (1994). Explorations of negotiation behaviors in ten foreign cultures using a model developed in the United States. Management Science, 40, 72-95.

Grazioli, S., & Jarvanpaa, S. (2000). Perils of Internet fraud: An empirical investigation of deception and trust with experienced internet consumers. IEEE Transactions on Systems, Man, and Cybernetics, 30 (4), 395-410.

Grover, S. (1993). Why professionals lie: The impact of professional role conflict on reporting accuracy. Organizational Behavior and Human Decision Processes, 4 (3), 478-495.

Kiesler, S., Siegel, J., & McGuire, T. W. (1984). Social psychological aspects of computer mediated communication. American Psychologist, 39 (10), 1123-1134.

Koh, K., Park, J., Kim, C., & Cho, S. (2001). Development of stress response inventory and its application in clinical practice. Psychosomatic Medicine, 63, 668-678.

Lax, D., & Sebenius, J. (1986). The Manager as Negotiator. New York: Free Press.

Lewis, S. A., & Fry, W. R. (1977). Effects of visual access and orientation on discovery of integrative bargaining alternatives. Organizational Behavior and Human Performance, 20 (1), 75-92.

McCornack, S., & Levine, T. (1990). When lovers become leery: The relationship between suspicion and accuracy in detecting deception. Communication Monographs, 57 (3), 219-230.

McCornack, S., & Parks, M. (1986). Deception detection and relationship development: The other side of trust. In M. J. McLaughlin (Ed.), Communications Yearbook 9 (pp. 377-389). Beverly Hills, CA: Sage.

McGrath, J. E., & Hollingshead, A. B. (1993). Putting the 'group' back in group support systems: Some theoretical issues about dynamic processes in groups with technological enhancements. In L. Jessup & J. Valacich (Eds.), Group Support Systems: New Perspectives (pp. 78-96). New York: MacMillan.

Miller, G., & Stiff, J. (1993). Deceptive Communication. Newbury Park, CA: Sage.

Neumann, D., Kurtzberg, T., Thompson, L., & Morris, M. (2003). Applying the Montreal taxonomy to state of the art e-negotiation systems. Group Decision and Negotiation, 12 (4), 287-310.

Oliver, R., Balakrishnan, P., & Barry, B. (1994). Outcome satisfaction in negotiation: A test of expectancy disconfirmation. Organizational Behavior and Human Decision Processes, 60 (2), 252-275.

Prater, T., & Kiser, S. B. (2002). Lies, lies, and more lies. SAM Advanced Management Journal, 67 (2), 9-36.

Pruitt, D. G. (1981). Negotiation Behavior. New York: Academic Press.

Pruitt, D. G., & Carnevale, P. J. (1993). Negotiation in Social Conflict. Pacific Grove, CA: Brooks Cole.

Purdy, J., Nye, P., & Balakrishnan, P. (2000). The impact of communication media on negotiation outcomes. International Journal of Conflict Management, 11 (2), 162-187.

Qin, T., Burgoon, J., Blair, J., & Nunamaker, J. (2005). Modality effects in deception detection and applications in automatic-deception-detection. Proceedings of the Thirty-Eighth Hawaii International Conference on System Sciences. Los Alamitos, CA: IEEE Press.

Rangaswamy, A., & Shell, G. R. (1997). Using computers to realize joint gains in negotiations: Toward an electric bargaining table. Management Science, 43 (8), 1147-1163.

Rao, V., & Lim, J. (2000). The impact of involuntary cues on media effects. Proceedings of the Thirty-Third Hawaii International Conference on System Sciences. Los Alamitos, CA: IEEE Press.

Rhee, H., Hasan, P., Jacob, V., & Barhki, R. (1995). Effects of computer mediated communication on group negotiation: An empirical study. Proceedings of the Twenty-Eighth Hawaii International Conference on System Sciences. Los Alamitos, CA: IEEE Press.

Rubin, J. Z., & Brown, B. R. (1975). The Social Psychology of Bargaining and Negotiation. New York: Academic Press.

Shell, G. R. (1999). Bargaining for Advantage: Negotiation Strategies for Reasonable People. New York: Viking.

Thompson, L. (1990). An examination of naive and experienced negotiators. Journal of Personality and Social Psychology, 59 (1), 82-90.

Thompson, L. (2005). The Mind and Heart of the Negotiator (3rd ed.). Upper Saddle River, NJ: Pearson-Prentice Hall.

Van de Vliert, E. (1997). Complex Interpersonal Conflict Behavior. London: Psychology Press.

Walker, R. (2003). Take it or leave it: The only guide to negotiating you will ever need. Inc., 25 (8), 75-82.

Wichman, H. (1970). Effects of isolation and communication on cooperation in a two- person game. Journal of Personality and Social Psychology, 16 (1), 114-120.

Yuan, Y., Head, M., & Du, M. (2003). The effects of multimedia communication on web-based negotiation. Group Decision and Negotiation, 12 (2), 89-109.

Zhou, L. (2005). An empirical investigation of deception behavior in instant messaging. IEEE Transactions on Professional Communication, 48 (2), 147-160.

Zmud, R. (1990). Opportunities for strategic information manipulation through new information technology. In J. Fulk & C. Steinfeld (Eds.), Organizations and Communication Technology (pp. 95-116). Newbury Park, CA: Sage.

Zuckerman, M., & Driver, R. (1985). Telling lies: Verbal and nonverbal correlates of deception. In W. Siegman & S. Feldstein (Eds.), Nonverbal Communication: An Integrated Perspective (pp. 129-147). Hillsdale, NJ: Lawrence Erlbaum.

About the Authors

Gabriel A. Giordano is a MIS doctoral candidate in the College of Business at Florida State University. His research interests include deception and deception detection in computer-mediated negotiations, interviews, and collaborative groups.
Address: MIS Department, College of Business, Florida State University, Tallahassee, Florida 32306-1110, USA

Jason S. Stoner is a Ph.D. candidate in organizational behavior and human resource management at Florida State University. His research interests are in the areas of stress and identity.
Address: Management Department, College of Business, Florida State University, Tallahassee, Florida 32306-1110, USA

Robyn L. Brouer is a Ph.D. student in organizational behavior and human resource management at Florida State University. She has research interests in the areas of leadership, social effectiveness, the multi-dimensions of person-environment fit, social influence processes including impression management and politics, leader-member exchange, and work stress.
Address:  Management Department, College of Business, Florida State University, Tallahassee, Florida 32306-1110, USA

Joey F. George is Professor of Information Systems and the Thomas L. Williams Jr. Eminent Scholar in Information Systems in the Management Information Systems Department in the College of Business at Florida State University. His research interests include the detection of deceptive computer-mediated communication, computer-based monitoring, group support systems, and information systems development.
Address: MIS Department, College of Business, Florida State University, Tallahassee, Florida 32306-1110, USA