Abstract

This article examines how Instant Messaging (IM) systems help employees of a Korean organization improve their relationships with their coworkers within and across organizational boundaries—within departments, between departments, and outside the organization. We briefly review literature about IM in developing working relationships and build our research questions. We then provide data analysis results based on a survey and structured interviews. Subsequently, in an exploratory case study of two individuals, we extend the analysis of departmental boundaries by including hierarchical levels, job profiles, and different communication purposes. Quantitative Social Network Analysis and visualization are used to analyze the communication pattern of the two individuals.

Introduction

A working relationship is defined as "an interpersonal relationship that is task-based, non-trivial, and of continuing duration" (Gabarro, 1990, p. 81). Although working relationships can be understood in the context of interpersonal relationships, involving exchanges of social contexts and norms between partners, they are clearly distinguished from ordinary interpersonal relationships, in that they are based upon the shared goal of task achievement.

In building an interpersonal relationship, people start from first impressions. Likewise, building working relationships entails a high level of uncertainty in the process of negotiating work expectations and understanding social behaviors and contexts among the people involved. Initial norms and expectations about work partners can be influenced by information from formal and informal sources, such as role definition, job manuals, or informally-obtained information about a partner's personality and behaviors. However, a great deal of uncertainty still exists when a person attempts to develop a working relationship with a new work partner.

This article examines how Instant Messaging (IM) systems help employees of a Korean tire manufacturing company improve their working relationship with their coworkers at organizational boundaries—within departments, across departments, and outside the organization. We briefly review literature about IM in developing working relationships and build our research questions. We then provide data analysis results based on a survey and structured interviews. Subsequently, in an exploratory case study of two individuals, we extend the analysis of departmental boundaries by including hierarchical levels, job profiles, and different communication purposes.

Instant Messaging and Working Relationships

Many traditional computer-mediated communication (CMC) studies, grounded in Media Richness Theory and Social Presence Theory, have focused on the bandwidth of media in determining their capability to convey the meanings or intentions of communication partners (Daft & Lengel, 1986; Short, Williams, & Christie, 1976; Siegel, Dubrovsky, Kiesler, & McGuire, 1986; Sproull & Kiesler, 1986). Due to its low bandwidth, text-based online communication has been considered ineffective for developing interpersonal relationships, since there are greater chances of misunderstanding, and incorrect impressions are often generated (Preece, 2000).

However, some researchers argue that most traditional theories about interpersonal relationship development may not apply to online communication. For example, Parks and Floyd (1996) report that an online communicator can develop an interpersonal relationship with someone met via an Internet newsgroup despite a lack of synchronous or spontaneous interactions. Walther (1992) also suggests that even though it may take longer than in face-to-face settings, online communicators can develop interpersonal relationships by devising their own relationship strategies in CMC to deliver social and emotional cues.

Parks and Floyd (1996) and Walther (1992) focused mainly on friendship development in online environments. Working relationships share some, but not all, characteristics with such relationships. Above all, working relationships are task-based; i.e., task accomplishment underlies the factors—such as trust, norms and values development, and liking and attractions among partners—that influence the development of working relationships. Therefore, some dimensions of interpersonal relationship development may not be applicable to the context in which the accomplishment of tasks is put forward. For example, in developing effective working relationships, openness concerning task-related issues is more important than unconditional self-disclosure.

One of the major factors affecting the development of working relationship is effective communication. This may not necessarily entail formal interactions. Gabarro (1990) points out that spontaneous and informal interactions are more effective in developing working relationships than are scheduled and formal interactions. Haythornthwaite, Wellman, and Mantei (1995) also provide empirical evidence that members of a research group used more unscheduled face-to-face encounters or email than formally scheduled meetings to accomplish their work.

Against this backdrop, synchronicity of communications between work partners is desirable, because delayed feedbacks may hinder working partners from exchanging and interpreting intended meanings accurately. In a similar vein, physical proximity between partners has been considered essential in developing working relationships, because co-located partners are more likely to have frequent encounters to discuss emerging issues (Kraut, Egido, & Galegher, 1990).

In this study, we focus on the role of Instant Messaging as a virtual collaboration tool that can mitigate communication barriers that might be caused by distributed environments. Instant Messaging (IM) is quasi-synchronous, computer-based, one-on-one communication (Nardi, Whittaker, & Bradner, 2000). IM is "quasi-synchronous" because it involves a small amount of asynchronicity, e.g., typing time. Due to its overlapping characteristics incorporating synchronous and asynchronous communications, and transient and persistent conversations, IM is often considered a hybrid communication medium (Voida, Newstetter, & Mynatt, 2002). In the context of working relationship development and improvement, this results in a combination of synchronous conversation and asynchronous use of transcripts or document transfer. IM not only supports turn-taking and rapid, informal conversations that are similar to oral conversation, but it also supports more carefully crafted formal sentences recorded in a transcript file or the transfer of formal documents.

Cramton (2001) found that collaborators were not able to maintain proper work relationships when they failed to build up shared knowledge about their work context. By using IM, collaborators can build up common knowledge about their work by quickly exchanging and instantly correcting their intended meanings or by maintaining their working context with the use of recorded transcripts or transferred documents. Furthermore, as new needs arise, the work context can be shared easily via IM because users can check whether relevant people are logged in and available for communication.

Despite the potential for IM to be used to maintain work relationships in an organization, there is a general perception that IM is used mainly for socializing, not for task-related purposes. Because of this perception, some work organizations even prohibit their employees from using IM in the work place. However, Muller, Raven, Kogan, Millen, and Carey (2003) found that employees in a business organization used IM for substantive business purposes, showing the medium's appropriateness for task-based communication, which is key to the development of working relationships. Isaacs, Walendowski, Whittaker, Schiano, and Kamm (2002) also found that employees in a business organization used IM more for business purposes than for socializing. They found two styles of IM use: Heavy users used IM for a wide range of collaborative activities, including complex tasks, while light users used IM mainly for simple coordination. However, Herbsleb, Atkins, Boyer, Handel, and Finholt (2002) found that some employees resisted using IM for their work because they felt that informal communication through IM was superfluous and not useful. Thus, the findings about the effects of IM on work effectiveness appear to be inconclusive, calling for more in-depth study of the role of IM in corporate environments.

Can the spontaneous and informal conversations of IM be useful in more formal work environments that require more careful planning and thinking before exchanging messages? There are two different viewpoints on how people choose different communication media. One view is that people choose communication media rationally, i.e., they choose a medium that can best convey their intended meaning, and their choice of medium is restricted by work-related contextual factors such as distance between communication partners. The other view is that people choose communication media in a social context, i.e., that social norms and culture, in addition to rational reasons, affect their choice (Fulk, 1993; Trevino, Lengel, & Daft, 1987; Trevino, Webster, & Stein, 2000).

These two views complement with each other in explaining media choice. Although people may choose a medium depending on their purpose, they may use the medium differently depending on the social context in which they are situated. This in turn relates to the underlying social meaning of the selected medium. The theory of media symbolism (Trevino et al., 1987, 2000) suggests that some symbolic meanings are attached to media, e.g., letters are regarded as more formal than e-mail. Symbolic meanings play an important role when people choose media, resulting in different choices of media. For example, Lee and Lee (2004) found that a symbolic meaning was attached to e-mail so that this medium was dispreferred by employees for communication with their supervisors. Given that IM is currently used as a very informal and casual communication channel, we posit that IM might show distinctive communication patterns, especially depending on with whom IM communication occurs. Specifically, we posit that there is a different media symbolism associated with IM, depending on the formality of its use in a specific working relationship.

To summarize, IM is a communication medium that can be used in a business context for simple coordination but also for collaboration among heavy users. Its hybrid nature, between synchronous and asynchronous communication, renders IM a suitable medium for both ad-hoc informal and more formal communication. This makes it useful for establishing social as well as task-based aspects of a working relationship.

Research Questions

Building a good working relationship is important for members of an organization, in order to achieve their tasks in a timely manner. In today's highly competitive business environments, collaboration with other members may also improve work effectiveness and efficiency. Because of its nature as described above, we believe that IM can play a crucial role in fostering working relationships. In particular, we posit that the semi-synchronous and casual characteristics of IM influence the dynamics of relationship-building and thereby enhance collaboration among organizational members.

In order to understand the influence of IM on the development of new working relationships and the enhancement of existing relationships, it is important to consider with whom the individual is communicating. To develop effective working relationships, workers should be able to interact with different work partners (within the department, across different departments, or external collaborators or customers). These work partners are different in terms of social distance (i.e., their perception of being "close" to each other), as well as their actual physical distance (i.e., between two organizational locations). For example, a working relationship with colleagues at the same department is very different from relationships with customers, because different social norms apply. In order for a communication medium to become an effective means of fostering working relationships in an organization, the medium must reduce barriers between communication partners, both in terms of physical and social distance.

Thus, we ask the following research questions:

RQ1. Is IM used differently with collaborators who are within departments, across departments, and outside the organization?

RQ2. Does IM help to improve working relationships with employees who are within departments, across departments, and outside the organization? What characteristics of IM are helpful or detrimental to working relationships with employees within departments, across departments, and outside the organization?

Research Methodology

Computer-mediated interactions take place through the mediation of communication technologies, and are affected by the reactions of people to that mediation. The focus of this study is on how people adapt their communications to the IM environment in order to maximize its advantages, rather than how IM determines interactions among people. In other words, this study views the process of building working relationships from the perspective of users, instead of examining the relationship from the perspective of the IM system.

To reflect this perspective we combine several methods. Our data collection includes quantitative surveys collected from 137 employees, qualitative interviews with 13 employees of two work teams, and two employees' IM transcripts that are subjected to social network analysis.

Organizational Setting

The study was conducted in a Korean tire manufacturing company. The company has been growing rapidly, and has become the leader of the industry in Korea. The company exports its products to many foreign countries, and has remote sale offices located in the U.S.A., Japan, Shanghai, Mexico, and Germany. Approximately 800 employees work at the headquarters and 4,200 employees work at remote offices. In recent years, management has been trying to shift its corporate image from a stable and traditional company to a young and dynamic company. At the same time, the company has also focused on applying Information Technology and Systems to corporate management.

The organizational structure of Korean companies is clearly distinguished vertically; each employee is assigned to one of five hierarchical levels, although the work responsibilities of the five hierarchical levels may not be the same in all departments. In general, Level 1 employees are assistants to managers or clerks, usually high school graduates, who fulfill some operational functions, or employees with college diplomas who provide general services for other team members. Level 2 employees also provide general services but the level of responsibility is higher than Level 1. Level 1 and 2 employees are non-managers. Level 3 employees are middle managers who are either in charge of specific work areas such as marketing of one local region, or evaluation of one specific work area. Level 4 employees are also middle managers, but they play the role of key communication liaisons with other departments or work teams. A Level 5 employee is the head of a work team.

Collaboration within departments is usually done through team meetings or departmental meetings in which the higher level (usually level 3 or up) employees in each work team discuss important issues of common interest to different work teams. Collaboration across different departments or across partner companies is mostly done through formal documents and occasional face-to-face meetings.

Survey

First, a survey was distributed which consisted of three sections (See Appendix 1). The questions of the survey were written in Korean and later translated into English (see Appendix 1). The first section included questions about demographic information such as age, gender, department, hierarchical level in the company, and self efficacy of computer use. The second section measured perceived frequencies of IM use in communicating on a daily or weekly basis with 1) employees within the same department, 2) employees in different departments, and 3) employees at other partnership companies. In order to investigate the nature of IM as a communication medium further, the questions from the second section were repeated for other media: scheduled face-to-face meetings, unscheduled face-to-face meetings, phone, and email. This categorization of communication media is adopted from Haythornthwaite et al. (1995). The third section of the survey asked about perceived changes in the use of other communication media at work due to IM, and the last section consisted of questions about perceived changes in working relationship after IM use with employees in the same departments, at other departments, and at other companies. The last section asked about the perceived frequencies of IM use in top-down (from a superior to a subordinate) and bottom-up (from a subordinate to a superior) communication.

In order to gain qualitative insight into IM use, two work teams—an eight-member Marketing Strategy (MS) Team and a five-member Computing System Development and Support (CSDS) Team—were selected from the company. The first author met each member and conducted a structured interview in Korean. The responses of the subjects were recorded and translated later into English. Interviewees were asked: Do you think IM is used as a formal communication channel in your organization? Why or why not? With whom do you communicate using IM at your workplace? With whom do you NOT communicate using IM at your workplace and why? For what specific applications or purposes do you think IM is useful? Do you think the use of IM enhances (or hurts) your working relationship with your coworkers or your collaborators in partnership companies? If so, what are the salient characteristics of the medium that cause such enhancement (or detriment)?

Demographics

The survey was distributed to 150 employees (all the employees in the Marketing Department and Information Systems Department; 137 responded to the survey (91%). Among the 29 female employees (19%) and 121 male employees (81%) in the Marketing Department and Information Systems Department, responses came from 28 females (20%) and 105 males (77%), with four (3%) providing no gender identification. Of the 137 respondents, 111 said that they used IM; in subsequent analysis, only their responses were included. For those 111 people, the average number of months of IM use in their workplace is 27 months (SD = 16.1).

Since IM is a relatively new technology, it was expected that younger employees would use IM more heavily. Therefore, in terms of demographics, age and organizational hierarchy were predicted to affect the use of IM most significantly. In response to the question item "Do you use Instant Messaging systems?," the distribution of the respondents' age (see Figure1) shows that IM is more frequently used by younger employees (younger than 35 years old) than by older employees (35 years or older) (Χ² = 19.5, df = 5; p = .002).

The distribution of the respondents' hierarchical level in the company (see Figure 2) shows that IM is more frequently used by lower level employees (Levels 1 and 2) than by higher level employees (Levels 3 & 4) (Χ² = 14.9, df = 5; p = .011). Interestingly, all of the Level 5 employees reported that they used IM.

Data Analysis and Results

Research Question 1: Differences in IM Use Within and Across Organizational Boundaries

To investigate the first research question, three question items were used: 1) For how long in a day (or in a week) do you use IM for communication with employees within your department?, 2) For how long in a day (or in a week) do you use IM for communication with employees in other departments?, and 3) For how long in a day (or in a week) do you use IM for communication with employees at other companies? Table 1 shows the descriptive statistics for answers to these questions. On average, employees spent 50 minutes a week in IM interacting with other employees within their department, 45 minutes with other employees in other departments, and 39 minutes with employees at other companies.

To test if there are significant differences in the amount of IM use at different organizational boundaries, a non-parametric equivalent of the paired t-test, the Wilcoxon Signed Ranks Test, was used (see Table 2). A significant difference was found only in the condition of Within Departments vs. Between Departments; employees spent significantly more time in IM talking with employees within the same department than with employees at other departments (p=.016). There was no significant difference in the amount of IM use Between Companies vs. Between Departments or Between Companies vs. Within Departments. These findings suggest that in terms of reported use of IM, the employees used IM more within the same department than between departments or companies.

Research Question 2: Differences in Improvement of Working Relationships Within and Across Organizational Boundaries

To answer the second research question, three question items were used: After using IM in my work, my working relationship with 1) other employees within my department, 2) other employees at different departments, and 3) employees at other companies has worsened or has improved (on a scale of 1 to 7, where 1=worsened and 7=improved). Table 3 presents the descriptive statistics for the answers to these question items; it shows that employees perceived that their working relationships with employees in the same department, in different departments and at other companies has improved after IM use.

In order to test if there are significant differences in terms of the effects of IM use on relationship improvement at different organizational boundaries, the Wilcoxon Signed Ranks Test was conducted (see Table 4). It was found that employees perceived that IM use improved their working relationship with other employees at different departments (Between Departments) more significantly than their working relationship with employees at other companies (Between Companies) (p=.004). However, no significant difference was found in other conditions (Within Departments vs. Between Departments and Between Companies vs. Within Departments). These findings imply that the employees perceived that IM improved their working relationship with their internal coworkers equally, regardless of their physical proximity. However, they perceived that IM did not improve working relationship with their external collaborators as much as it did with their internal coworkers in other departments.

Improvement of Working Relationships Within vs. Between Departments

We now further investigate and hypothesize possible explanations for our findings, based on the analyses of other, supplementary question items and qualitative comments from the interviewees. Based on the results for RQ1 and RQ2, it is concluded that although IM was used significantly more for communication within departments than between departments, respondents felt that the use of IM improved their working relationships between departments to the same extent as it did within departments. This leads to the following additional question:

Why do employees perceive the benefit of IM in improving working relationships between departments as equal to those within departments, even though they use IM more in communicating within departments?

To answer this question, it is worth comparing the usage patterns of IM with the usage patterns of other communication media. Unscheduled face-to-face meetings occurred more frequently within departments than between departments (z=-3.5, p=.000), between departments than between companies (z=-3.5, p=.000), and within departments than between companies (z=-4.6, p=.000). This result is in contrast to the findings for the frequency of scheduled face-to-face meetings, which show no significant difference in the three situations. (See Appendix 2.)

It seems that employees working in different departments or different companies have difficulty holding instant "on-the-fly" meetings whenever a new need emerges, thus they are more dependent on scheduled meetings. These results support the findings of previous studies that individuals who are located physically far from each other are less likely to choose face-to-face meetings (Trevino et al., 2000) and more likely to choose an online medium such as email (Haythornthwaite et al., 1995). However, to develop a solid working relationship with coworkers, people should have frequent, spontaneous contact with each other so that they can discuss emerging issues whenever necessary and maintain mutual understanding about their projects (Kraut et al., 1990). Therefore, unscheduled, intermittent encounters are more effective than scheduled, sparse meetings in developing working relationships among collaborators (Gabarro, 1990; Haythornthwaite et al., 1995). IM may provide more benefits for coworkers in different departments by helping them overcome physical distances and providing chances for unexpected and unscheduled "virtual corridor talks," than for coworkers within the same department who have plenty of chances for "physical corridor talks." As one of our interviewees commented: "Using IM, I can talk with my coworkers who are working on a different floor whenever we have a chance. This would not have been possible without IM."

Improvement of Working Relationships Within vs. Between Organizations

Our results show that employees perceived that the use of IM helped them to improve working relationships with internal colleagues (in other departments in the same company) more significantly than with external business partners at other companies. In terms of physical distances between coworkers, employees might have perceived the use of IM as equally beneficial for improving working relationships with external collaborators as with internal collaborators. Therefore, it would be interesting to investigate the following question:

Why do respondents perceive IM as more beneficial for improving working relationships inside the organization's boundaries than beyond the organization's boundaries?

It seems that one drawback of using IM in developing working relationships lies in the process of registering a new contact in the contact list. One of the interviewees said:

I seldom use IM communication with customers. We (I and the customer) don't have such a close relationship. How can I add someone who is not very close to me to my IM contact list? I barely know him/her. At least I have to know how he/she looks or [what] his/her voice is like.

This remark implies that other forms of interactions, such as face-to-face meetings or phone conversations, are necessary to initiate a new working relationship. Since IM forms a closed personal online community, the very action of adding a new contact to IM seems to be based on an assumption that at least a minimum level of relationship has already been established with the contact. Only after the relationship moves beyond the threshold of the initial minimum level can the relationship be further enhanced by IM. In other words, IM appears to be less effective in developing a new working relationship with external business partners than in maintaining existing relationships with internal coworkers.

In-Depth Case Studies

In order to generate further insights into our two primary research questions, we analyzed the role of job characteristics and communication purpose in relation to IM usage. This was done through transcript coding and Social Network Analysis, and the findings were illustrated with network visualizations. Through these methods we were also able to extend the analysis by adding a preliminary and exploratory discussion of the vertical boundaries of organizational hierarchies. We expected these aspects to be important explanatory factors for the establishment and maintenance of working relationships across different organizational boundaries.

Methodology and Procedure

In order to examine how individuals use IM in their workplace, the IM transcripts of two individuals, which were collected over a period of six months, were analyzed in depth: One was a member of the MS team and the other a member of the CSDS team. The transcripts produced more than 80,000 messages in Korean, which were later translated later into English and analyzed to identify the purposes of IM use in different job-related situations. These two employees were selected because they were identified as the heaviest IM users in each work team.

The member of the MS team (hereafter called John) is a 26-year-old male at Level 1, and the member of the CSDS team (hereafter called Paul) is a 33-year-old male at Level 3 in the organizational hierarchy. John's main responsibilities include analyzing market trends, positioning the company in the market based on the analysis of market trends, projecting the profitability of the company based on sales records, and developing a Knowledge Management website for the company. Meanwhile, Paul has responsibility for managing an ERP (Enterprise Resource Planning) system that was recently completed in the company, and providing technical support to remote offices (including overseas offices in China, Shanghai, and the U.S.A.) for their ERP implementation projects. Some of these remote offices have already launched ERP systems while others are at the development or testing stage.

To investigate how these two employees who have two distinct job roles use IM for collaborating with other coworkers within and across different organizational boundaries and hierarchies, we categorized the purpose of each IM conversation. The transcripts provided by John and Paul were coded by the third author for conversation partners and purposes. The categorization of conversation purposes was based on categories adapted from Muller et al. (2003) and Isaacs et al. (2002). The coding process was "reflective and incremental," that is, whenever a new category emerged that needed to be added to our original categories, the coder went back to the previous coding results and adjusted them based on the new set of categories. For conversations incorporating multiple purposes, all purposes were coded for one unit of conversation.

The definitions of the 17 categories used in the transcript coding are given in Table 5.

To gain deeper insights into the structure of the employees' IM networks and their role in building, maintaining, and improving working relationships, the method of Social Networks Analysis (SNA) was utilized, as social networks are virtually present whenever a group of people interacts electronically (Wellman, 1997). Viewing the IM network as a community resting on an underlying social network enables a systematic examination of such computer-networked communities (Wellman, Salaff, Dimitrova, Garton, Gulia, & Haythornthwaite, 1996).

Social Network Analysis is a framework for the analysis of structured social relationships (Wasserman & Faust, 1994), which, in the organizational context, can reflect role-based authority relationships of formal organizational structures, informal structures based on communication, information exchange, or affection (Tichy, Tushman, & Fombrun, 1979). Informal networks coexist with the formal structure of a company and can help solve problems in more efficient ways (Huberman & Hogg, 1995). By collecting and measuring relationships among the individual participants in the Instant Messaging network, qualitative and quantitative descriptions of the group's relationships can be derived.

In this study, the method of ego-centric networks (or simply ego networks) is applied for quantitative analysis and visualization of the IM networks of John and Paul. An ego-centric network consists of a focal actor, termed ego, a set of alters who have ties with ego, and the measurement of the ties between ego and its alters. Ego networks are used frequently to study the social support of persons (Wasserman & Faust, 1994). Using this approach enables scientific measurements of the most important network properties derived from a very simple set of data stored in a matrix. Each alter in an ego network has its own personal network again, so that ultimately all ego networks interlock to form a social network. The social network analysis conducted for this study makes use of graphical methods for analyzing and visualizing social networks (Hage & Harary, 1983). Such visualization approaches originated from the early work of Moreno (1934), who introduced a notation wherein points represent actors and edges indicate the links between actors.

In order to generate the complex visualizations of the Instant Messaging activities automatically, the software application Commetrix was utilized (Trier, 2005). Originally developed by the second author to enable the analysis of asynchronous virtual discussions like NNTP/Usenet Newsgroups and PHP Bulletin Board Software, a special connection to MSN-Messenger was developed to automatically scan, extract, and analyze the network structures from over 80,000 communication acts stored in the sampled electronic IM archives of the two employees. The messages between authors or the references between postings (in the case of discussion networks) are then visualized as a force-directed two-dimensional Spring Embedder Algorithm (Fruchterman & Reingold, 1991). During multiple iterations, a clustered network graph emerges, visualizing structured patterns of strong relationships in a group. This procedure was followed to uncover important structural properties within the Instant Messaging network.

Introducing Communication Purposes and Job Profiles as Additional Factors

In order to better understand how employees use IM to establish and improve their working relationships within and across different organizational boundaries and hierarchical levels, we included the context of communication purposes and their relation to the respondents' job roles. Figures 3 and 4 show how the two employees used IM for different communication purposes.

We can now discuss use patterns based on the two employees' different job characteristics. Most importantly, John and Paul work in different areas: John works in the Marketing Strategy (MS) Team while Paul is employed in the Computing System Development and Support (CSDS) Team. Accordingly, the tasks of these two employees contrast with each other in terms of the flow of service/information, i.e., Paul is a service provider in the sense that he provides technical advice to other computing staff members who install and maintain ERP systems in remote offices, while John is an information collector in the sense that his major task is to collect and analyze information from inside and outside of the company.

Their job characteristics are reflected in their IM usage pattern. John used IM mostly for collecting information: 25% of his IM use was for document transfer. Meanwhile, Paul focused more on collaborative problem solving. Paul used IM very heavily in identifying system problems, and finding solutions collaboratively with his remote work partners. IM seems to provide an effective and efficient communication channel for problem solving by allowing people to stay in the same context. For example, Paul and his IM partners used IM intensively for technical problem solving by copying error messages or system logs and pasting them into the Messenger window, which facilitated discussion of the problems.

In order to gain further insight into the use of IM for different communication purposes, we integrated categories 1 to 9 of Table 5 to form the domain organization of work (prepare and announce meetings, etc.); similarly, categories 10 to 15 constitute activities of knowledge work (problem solving, information exchange, requesting advice). Based on this reduced set of domains, it becomes clear that both John and Paul used IM mostly for knowledge work (50% for John and 67% for Paul). They used IM to build a shared common ground (Clark, 1996; Ducheneaut & Bellotti, 2003; Whittaker, 2003) or mutual knowledge (Krauss & Fussell, 1990), but in different forms. John tried to build mutual knowledge by transferring documents, while Paul shared text messages with his partners in a more synchronous manner to ensure that they understood the same technical problem in the same context.

Socializing, which can be interpreted as a means of maintaining and improving relationships to foster knowledge work, also accounts for part of their communication (33% for John, 8% for Paul). However, consistent with the findings of Isaacs et al. (2002), task-based communication turned out to be the primary purpose of the two employees' communication. The different amount of socialization could be attributed to the fact that John is more a "Collector" who needs to establish good connections to draw resources from his network, whereas Paul is more a "Provider," who does not require much socializing to maintain his relations.

This content-oriented analysis can be extended by an analysis of the different social network patterns of the two employees. In 217 days, John sent 2,574 and received 3,185 messages, while Paul sent 3,9595 and received 38,217 messages in 244 days. Simple comparison between the inbound and outbound traffic does not provide clear evidence for John being either a passive collector or an active initiator of messages. This also applies to Paul. However, comparing the traffic volume of both persons shows a clear difference. On average, Paul sent 162 messages per day, whereas John only sent 12 messages daily, indicating that Paul is a heavier IM user than John. With this large volume of messages Paul maintains 139 contacts, which is more than three times as many as John's 45 contacts. The different scope of the two employees' networks can be visualized with an ego-centric social network visualization (see Figure 5). Paul's network intuitively appears much larger.

These differences in relationship strength (also compare the section on IM and hierarchy) and intensity of use may stem from their different job profiles. Paul, as a member of the IT department and a provider of technical services, used IM to actually conduct his tasks, whereas John seems to have used IM as a supplementary tool to further his work indirectly. Interestingly, however, Paul does not spend considerably more time using IM. Compared to John, Paul used IM very efficiently, with shorter messages and smaller gaps between communication sessions. This also supports the argument that Paul used IM as a productivity-enhancing instrument to improve the efficiency of his work, while John used IM as a supportive communication medium to enhance his social relationships in the work place.

Research Question 1: Differences in IM Use Within and Across Organizational Boundaries

The analysis of the quantitative structures of the communication networks of John and Paul has provided further detail about the patterns of working relationships within and across organizational boundaries and hierarchies. To refer to our first research question, the spread of the two individuals' contacts across departments and hierarchies was analyzed in more detail (see Figure 5, which includes the main information about the location and hierarchy of the contacts; see also Tables 6 & 7). In terms of message volume, John sent 69% of his messages to coworkers in other departments, whereas most of Paul's IM communications occurred within the department (91%).

This pattern fits with the communication purposes and job characteristics as described in the previous section: John's main responsibility is to collect information from outside and report it to his supervisors. It should be noted that he received 16% of messages (out of the total 18%) and sent 10% of messages (out of the total 13%) from and to his superiors within his department. Meanwhile, Paul's main responsibility is to solve technical problems and provide technical consulting for the ERP system installed in the company, and thus most of his conversations were very technical and shared only with other computer support staff members within his department. This job characteristic is also consistent with the result of our quantitative analysis that most of Paul's IM communications occurred within the department (91%).

In relation to the social network analysis, one aspect of the survey data caught our attention: how IM is used in collaborating with coworkers at different levels of the organizational hierarchy. Employees use IM more in top-down communications (with their subordinates) than in bottom-up communications (with their superiors) (p=.000) (See Appendix 3). Combined with the finding of the use of more formal and less spontaneous media such as scheduled face-to-face meetings or email, which show no significant difference between bottom-up and top-down communications, we conclude that the informal and spontaneous nature of IM discourages employees from using IM for communicating with their superiors, perhaps because they wish to avoid possible misunderstandings.

Analyzing the communication patterns of IM use in terms of organizational hierarchy in the context of the communicator's intended purpose provides deeper insight into the above finding. In relation to superiors in the same department, John is a reactive collector and receiver with a ratio of 1:0.625 messages received:sent. This is an interesting result, as given his job, John would be expected to be the active reporting person and his superior the receiver and collector of the information. The inverse tendency was observed in John's conversation with subordinates in different departments: John received 1,038 and sent 1,234 messages (a ratio of 1:1.2). Hence, John is more of an initiator of informal messages to his subordinates than he is a collector. John's communication patterns are consistent with the survey findings that show that IM is used more in top-down communications than bottom-up communications.

Compared to John's ratio of 1:0.625, Paul's communication with superiors in the same department displays a different pattern. Paul received 280 messages and sent 503, a ratio of 1:1.8. Paul is hence an active sender, unlike John, who receives more IM messages from his superiors than he sends. Paul's usage contrasts sharply with the survey's finding that IM is used more in top-down communication than bottom-up communication. However, considering his job description—providing technical support to all hierarchy levels in the organization—and considering that Paul used IM less for socializing and more for knowledge work than John, we might reasonably assume that Paul displays less resistance to using IM for sending messages to his superiors because his purposes are more task-oriented.

The patterns of John's and Paul's IM communication have been automatically rendered and visualized as an ego centered network in Figure 5 using Commetrix (Trier, 2005). Two participants have a relation (edge) if they have exchanged Instant Messages. The color of the relation represents the number of messages exchanged. The more two people communicate, the darker is the color of their connecting edge and the closer they are positioned to each other in the graph. The star pattern, which was produced by the layout algorithm, is typical for an ego-centric data set.

Using these graphs, the relationship patterns of John and Paul can be analyzed in more detail by looking for strong ties. For this we defined a threshold that separates strong from weak connections: It is at half the strength of the strongest relation (Rmax/2 with Rmax = strongest relation in the egonet). This provides us with a relative view that eliminates the differences in usage intensities between John and Paul. Applying this method, John maintains 11% (5) strong ties (threshold Rmax/2 = 218 interactions) which are responsible for 45% of the overall message volume. The remaining 89% (40 contacts) are weak ties. The tight connections do not show any preference for a special hierarchical group. They include contacts at different hierarchical levels within and outside the department: one peer and two subordinates from external departments plus two internal superiors. Paul maintains 5% (7) strong relationships (threshold Rmax/2 = 1147 interactions). They generate only 30% of Paul's processed message volume. Still, comparing the average amount of interaction, Paul has considerably stronger communication ties with his contacts than does John. If John's threshold for a strong tie were applied to Paul, he would maintain 48 strong relationships. Figure 5 also shows that Paul has strong ties only to subordinates in the same department. Generally, he has very few dense relationships with employees in different departments and none with his superiors. This pattern is consistent with the difference in the job profiles of John and Paul: John collects information from inside and outside of the organization, while Paul collaborates extensively with intradepartmental coworkers who have similar technical knowledge to solve problems in the computer system.

Research Question 2: Differences in Improvement of Working Relationships Within and Across Organizational Boundaries

To illustrate the findings for RQ2, which explored the improvement of working relationships, the quantitative development of John's and Paul's relationships across hierarchies and across departments over the entire sampling period of six months was visualized in Figure 6. Over time, new contacts are added continuously. Core contacts move to the center.

Analyzing the hierarchies and locations of the contacts shows that John added 26 new contacts to his 17 existing contacts (250% growth): four external collaborators, four internal superiors, seven external peers, and 11 external superiors. Within the same time period, Paul added one internal superior, 48 internal subordinates, 15 external subordinates, 18 external peers, and two external collaborators. Starting with 35 contacts at the beginning of the sample period, he accumulated 84 new contacts for more than a 300% growth rate in only six months.

The continuous increase in working relationship strength of John and Paul to colleagues at different hierarchies and locations is depicted in Figure 7.

Conclusions

This study has shed light on the effectiveness of IM use in developing and improving working relationships. The results suggest that the employees of the Korean tire manufacturing company used IM to maintain their working relationships with coworkers within and across various organizational boundaries, and perceived that their relationships both inside and outside of the company had improved after adopting IM in their work. However, some discrepancies were noted between the employees' actual IM use and their perception of the effects of IM use.

Inside the company, they used IM more in collaborating with their coworkers in the same department than with coworkers in different departments, yet their perception of how much IM improved their working relationships did not differ between intradepartmental and interdepartmental communications. To investigate the reason further, we analyzed and compared the usage patterns of IM and other communication media; the results suggest that IM provides opportunities for instant and spontaneous conversations (virtual corridor talk), whereas employees at different locations have less opportunity for "on-the-fly" contacts and are more likely to hold face-to-face meetings and exchange e-mail.

However, a significant difference was found between the employees' perception about IM use for communication with their counterparts in other companies and communication inside the company. The employees perceived IM as more beneficial for working relationships inside the company than outside the company. Although the informal and spontaneous nature of IM is supposed to reduce communication barriers between remote collaborators, social distance still seems to play a role in determining employees' perception of effectiveness of IM in working relationship development, especially when they need to maintain some formality with their communication partners. Similarly, our survey data show that the employees used IM for top-down communications more than for bottom-up communications. Specifically, the data suggest that employees are reluctant to use IM for reporting to their superiors.

There could be several reasons for reluctance to use IM in the workplace. First, the informality of the medium could convey inappropriate impressions about the content of the message. Second, employees may be reluctant to use such a spontaneous medium to convey important contents, especially in situations where they want to check and edit their messages before sending them. Third, these two factors could be accentuated when employees communicate with external collaborators or with superiors with whom they wish to maintain more formal relationships. Moreover, since IM is a relatively new technology, it may not yet have been adopted by many people, especially in older generations. The issue of IM non-use calls for more in-depth research in the future.

We also considered IM use in the context of job profiles of the employees and the purposes of use. We compared the IM use of two employees with different job profiles: John, a collector of information from inside and outside of the organization, and Paul, a service provider to technical staff. The social network analysis and transcript coding suggest that different job profiles affect the way employees use IM: John used it to improve his social connections when collecting information from diverse sources, while Paul used IM to improve his productivity in supporting coworkers with whom he shares a clear context in that they are focusing on solving technical problems.

Some limitations of this study must be noted. Since the study was conducted in one specific culture, Korea, the findings can not be generalized and extended to other cultures. Organizational hierarchy and vertical relationships are considered more important in working relationships in Korea than they are in western cultures; thus the effects of hierarchy on IM use need to be investigated in other cultures in the future. However, within the context studied, we believe that our findings suggest interesting perspectives on how members of an organization use IM with collaborators at different social positions and in different locations in relation to organizational boundaries.

References

Clark, H. (1996). Using Language. Cambridge: Cambridge University Press.

Cramton, D. C. (2001) The mutual knowledge problem and its consequences for dispersed collaboration. Organization Science, 12 (3), 346-371.

Daft, R. L., & Lengel, R. H. (1986). Organizational information requirements, media richness and structural design. Management Science, 32 (5), 554-572.

Ducheneaut, N., & Bellotti, V. (2003). Ceci n'est pas un object? Talking about objects in email. Human-Computer Interaction, 18, 85-110.

Fruchterman, T. M. J., & Reingold, E. M. (1991). Graph drawing by force-directed placement. Software - Practice & Experience, 21 (11), 1129-1164.

Fulk, J. (1993). Social construction of communication technology. Academy of Management Review, 36 (5), 921-950.

Gabarro, J. (1990). The development of working relationships. In J. Galegher, R. E. Kraut, & C. Egido (Eds.), Intellectual Teamwork: Social and Technological Foundations of Cooperative Work (pp. 79-110). Hillsdale, NJ: Lawrence Erlbaum Associates.

Hage, P., & Harary, F. (1983). Structural Models in Anthropology. Cambridge: Cambridge University Press.

Haythornthwaite, C., Wellman, B., & Mantei, M. (1995). Work relationships and media use: A social network analysis. Group Decision and Negotiation, 4 (3), 193-211.

Herbsleb, J., Atkins, D., Boyer, D., & Handel, M. (2002, April). Introducing Instant Messaging and Chat in the Workplace. Paper presented at the Conference on Human Factors in Computing Systems (CHI), Minneapolis, MN.

Huberman, B. A., & Hogg, T. (1995). Communities of practice: Performance and evolution. Computational and Mathematical Organization Theory, 1 (1), 73-92.

Isaacs, E., Walendowski, A., Whittaker, S., Schiano, D. J., & Kamm, C. (2002, November). The character, functions, and styles of Instant Messaging in the workplace. Proceedings of the 2002 ACM Conference on Computer Supported Cooperative Work. New York: ACM Press.

Krauss, R. M., & Fussell, S. R. (1990). Mutual knowledge and communicative effectiveness. In J. Galegher, R. E. Kraut & C. Egido (Eds.), Intellectual Teamwork: Social and Technological Foundations of Cooperative Work (pp. 111-146). Hillsdale, NJ: Lawrence Erlbaum Associates.

Kraut, R. E., Egido, C., & Galegher, J. (1990). Patterns of contact and communication in scientific research collaboration. In J. Galegher, R. E. Kraut & C. Egido (Eds.), Intellectual Teamwork: Social and Technological Foundations of Cooperative Work (pp. 149-171). Hillsdale, NJ: Lawrence Erlbaum Associates.

Lee, J., & Lee, Y. (2003, December). Cultural Implications of Electronic Communication Usage: A Theory-Based Empirical Analysis. Paper presented at the International Conference on Information Systems, Seattle, WA.

Moreno, J. L. (1934). Who Shall Survive?: Foundations of Sociometry, Group Psychotherapy, and Sociodrama. Washington, D.C.: Nervouse and Mental Desease Publishing Co. Reprinted in 1953 (Second Edition) and in 1978 (Third Edition) by Beacon House Inc., Beacon, New York.

Muller, M. J., Raven, M. E., Kogan, S., Millen, D. R., & Carey, K. (2003, November). Introducing Chat into Business Organizations: Toward an Instant Messaging Maturity Model. Paper presented at Conference on Supporting Group Work, Sanibel Island, FL.

Nardi, B., Whittaker, S., & Bradner, E. (2000, December). Interaction and outeraction: Instant Messaging in action. Proceedings of the 2000 ACM Conference on Computer Supported Cooperative Work. New York: ACM Press.

Parks, M. R., & Floyd, K. (1996). Making friends in cyberspace. Journal of Communication, 46 (1), 80-97.

Preece, J. (2000). Online Communities: Designing Usability, Supporting Sociability. Chichester and New York: John Wiley & Sons.

Short, J., Williams, E., & Christie, B. (1976). The Social Psychology of Telecommunications. London: Wiley.

Siegel, J., Dubrovsky, V., Kiesler, S., & McGuire, T. W. (1986). Group process in computer-mediated communication. Organizational Behavior and Human Decision Processes, 37 (2), 157-187.

Sproull, L., & Kiesler, S. (1986). Reducing social context cues: Electronic mail in organizational communication. Management Science, 32 (11), 1492-1512.

Tichy, N. M., Tushman, M. L., & Fombrun, C. (1979). Social network analysis for organizations. Academy of Management Review, 4 (4), 507-519.

Trevino, L. K., Lengel, R. H., & Daft, R. L. (1987). Media symbolism, media richness, and media choice in organizations. Communication Research, 14 (5), 553-574.

Trevino, L. K., Webster, J., & Stein, E. (2000). Making connections: Complementary influences on communication media choices, attitudes, and use. Organization Science, 11 (2), 163-182.

Trier, M. (2005, January). IT-supported visualization of knowledge community structures. Proceedings of the 38th Hawaii International Conference on System Sciences. Los Alamitos: IEEE Press.

Voida, A., Newstetter, W., & Mynatt, E. (2002, April). When conventions collide: The tensions of Instant Messaging attributed. Proceedings of the ACM Conference on Human Factors in Computing Systems (pp. 187-194). New York: ACM Press.

Walther, J. B. (1992). Interpersonal effects in computer-mediated interaction: A relational perspective. Communication Research, 19 (1), 52-90.

Wasserman, S., & Faust, K. (1994). Social Network Analysis. Cambridge, MA: Cambridge University Press.

Wellman, B. (1997). An electric group is virtually a social network. In S. Kiesler (Ed.), Culture of the Internet (pp. 179-205). Mahwah, NJ: Lawrence Erlbaum.

Wellman, B., Salaff, J., Dimitrova, D., Garton, L., Gulia, M., & Haythornthwaite, C. (1996). Computer networks as social networks: Collaborative work, telework, and virtual community. Annual Review of Sociology, 22, 213-238.

Whittaker, S. (2003). Things to talk about when talking about things. Human-Computer Interaction, 18, 149-170.

Appendix 1: Questionnaire Items

Perceived Frequencies of the Communication Medium Use by Organizational Boundary

How frequently do you use the following communication media for communication with your coworkers in your department?

Scheduled Face-to-Face Meeting: _____________ time(s) / day OR _____________ time(s) / week
Unscheduled Face-to-Face Meeting: _____________ time(s) / day OR _____________ time(s) / week
Phone: _____________ time(s) / day OR _____________ time(s) / week
Email: _____________ time(s) / day OR _____________ time(s) / week
IM: _____________ minute(s) / day OR _____________ minute(s) / week

How frequently do you use the following communication media for communication with your coworkers at different departments?

Scheduled Face-to-Face Meeting: _____________ time(s) / day OR _____________ time(s) / week
Unscheduled Face-to-Face Meeting: _____________ time(s) / day OR _____________ time(s) / week
Phone: _____________ time(s) / day OR _____________ time(s) / week
Email: _____________ time(s) / day OR _____________ time(s) / week
IM: _____________ minute(s) / day OR _____________ minute(s) / week

How frequently do you use the following communication media for communication with your collaborators at different companies (e.g., Customers or business partners)?

Scheduled Face-to-Face Meeting: _____________ time(s) / day OR _____________ time(s) / week
Unscheduled Face-to-Face Meeting: _____________ time(s) / day OR _____________ time(s) / week
Phone: _____________ time(s) / day OR _____________ time(s) / week
Email: _____________ time(s) / day OR _____________ time(s) / week
IM: _____________ minute(s) / day OR _____________ minute(s) / week

Perceived Working Relationship Changes after IM Adoption:

After adopting IM in my work, my working relationship with my coworkers at the same department has:

After adopting IM in my work, my working relationship with my coworkers at different departments has:

After adopting IM in my work, my working relationship with my collaborators at different companies (eg. customers or business partners):

Perceived Frequencies of Communication Medium for Top-down and Bottom-up communications

How frequently do you use the following communication media for communication with your subordinates?

Scheduled Face-to-Face Meeting: _____________ time(s) / day OR _____________ time(s) / week
Unscheduled Face-to-Face Meeting: _____________ time(s) / day OR _____________ time(s) / week
Phone: _____________ time(s) / day OR _____________ time(s) / week
Email: _____________ time(s) / day OR _____________ time(s) / week
IM: _____________ minute(s) / day OR _____________ minute(s) / week

How frequently do you use the following communication media for communication with your supervisors?

Scheduled Face-to-Face Meeting: _____________ time(s) / day OR _____________ time(s) / week
Unscheduled Face-to-Face Meeting: _____________ time(s) / day OR _____________ time(s) / week
Phone: _____________ time(s) / day OR _____________ time(s) / week
Email: _____________ time(s) / day OR _____________ time(s) / week
IM: _____________ minute(s) / day OR _____________ minute(s) / week

Appendix 2

Wilcoxon Signed Ranks Test for Scheduled and Unscheduled Face-to-Face Meetings

Appendix 3

IM usage in bottom-up and top-down communications (minutes/week)

Wilcoxon Signed Ranks Test for Comparing Bottom-up vs. Top-down communications using different communication media

About the Authors

Hee-Kyung Cho is an assistant professor of the Department of Information Systems and Decision Sciences at Silberman College of Business of Fairleigh Dickinson University. Her research interests include computer-mediated communication (CMC), decision support systems (DSS), technology-supported learning, and human computer interaction (HCI).
Address: Fairleigh Dickinson University Metropolitan Campus, 1000 River Road, Teaneck, NJ 07666 USA

Matthias Trier is a research affiliate and member of the research group Integrated Knowledge Management (IKM) of the Institute for Business Informatics of the Technical University Berlin. He is working in the field of Knowledge Management with a special focus on the computer-supported visualization of Knowledge Communities, also called Communities of Practice. Simultaneously, he works on production systems for e-Learning contents in Virtual Universities.
Address: 28-29 Franklinstrasse, Berlin 10587 Germany

Eunhee Kim is a Ph.D. candidate at the Information Systems Department of New Jersey Institute of Technology. Her research interests include Asynchronous Learning Networks, technology supported learning, computer-mediated communication (CMC) and human computer interaction (HCI).
Address: Department of Information Systems, New Jersey Institute of Technology, University Heights, Newark, NJ 07102 USA