Instant Messaging Usage and Interruptions in the Workplace

2.1. Media richness

Daft and other scholars (Daft & Lengel, 1984, 1986; Daft & Weick, 1984) considered an organization an information processing system. The organization’s employees select an appropriate communication medium based on the level of uncertainty and ambiguity of their work; thus, media richness was developed as a key concept to understand the information capacity of a medium. A rich media type can deliver information through expressions, eye contact, posture, dress, and tone and consider the distance and location of the users during the interaction. These communication details can provide feedback and screening and help determine the attitudes of the interacting partners to maintain their attention in the interaction or terminate the dialogue.

From the perspective of the symbolic interactionist, Trevino, Daft, & Lengel (1990) discussed the relationship between an organization’s media choice and media richness. An organization is a dynamic and interactive web, in which the members establish a shared value system through their interactions. The organization members collectively develop basic assumptions and understandings regarding their words, behaviors, and actions. When seeking consensus through consultation and feedback, selecting an appropriate medium to address uncertainty and ambiguity in the messages becomes crucial. Therefore, when the level of uncertainty and ambiguity is high, using a rich, more personal communication medium (e.g., face‐to‐face communication) is more efficient than using a less rich, impersonal communication medium (such as a written statistical chart) (Trevino, Lengel, & Daft, 1987).

Each media type has a different degree of richness depending on the following four criteria that are based on individual perception: first, whether the medium provides instant feedback to facilitate rapid consensus; second, whether the medium delivers information with multiple cues, such as body language and tone, to facilitate the interpretation of the message; third, whether the medium uses natural language to facilitate the transmission of delicate and subtle messages, rather than a form such as a digital chart; and fourth, whether the medium creates a personal focus that describes needs, emotions, and feelings to the recipient (Trevino, Daft, & Lengel, 1990, p. 75).

Many previous studies describing the characteristics of IM are available. For example, IM allows users to instantly and synchronously communicate with individuals on their friend list. In addition to one‐on‐one communication, multiple users can dialogue simultaneously, and when a user cannot interact with individuals online, he/she can send an offline message (e.g., Jacobson, 2008). IM is often considered a hybrid media because it can be used for not only informal communication, which is similar to spoken language, but also for formal statements of record or file transfers (Cho, Treir, & Kim, 2005).

However, few studies are available on IM characteristics from the perspective of media richness. One study was conducted by Waldeck, Seibold, and Flanagin (2004). They found that, in the process of socialization in organizational settings, employees tend to obtain information using a medium with significant richness. The other study by Cameron and Webster (2005) used the media richness theory to investigate why employees use IM at work. However, the results showed that the majority of respondents believed that IM was not a rich media. Of the four criteria for the media richness assessment, only the instant feedback result was rated relatively high.

2.2. Social influence

Fulk, Schmitz, and Steinfield (1990) strongly questioned the assumptions of media richness. First, they indicated that many empirical studies do not support the concept of media richness because e‐mail, which is a medium with relatively low richness, is often used in communications containing highly interpersonal content (such as conflict settlements and negotiations) (Rice & Love, 1987; Steinfield, 1985; Markus, 1988). Second, these researchers’ own study observed that the attitudes and behaviors of one’s colleagues toward media use is a main factor that affects media selection in the workplace (Steinfield, 1986). Thus, based on the social information processing theory (Salancik & Pfeffer, 1978), Fulk, Schmitz, and Steinfield (1990) proposed that social influence is the main factor in media selection. The social information processing theory of Salancik and Pfeffer (1978) was originally proposed to define overall job attitudes and task designs of an organization. The basic assumption of the theory is that individuals will adjust their own attitudes, behaviors, and beliefs to adapt to their social environment; namely, to understand the cause of an individual’s behavior, one must examine information from the social environment in which the individual is situated. Individuals express their attitudes and needs based on the available information in their social environment because the social context provides clues to interpret an event and the necessary information for individuals to form an appropriate attitude and opinion.

Thus, the social information processing theory emphasizes that the nature of work or a task is constructed instead of given. When the related information is not available, individuals must communicate with others and develop a stable system of consciousness through communication. Individuals evaluate their information sources based on their interpersonal relationships: the closer the relationship with an individual, the more their opinions are applied to understand one’s own situation. Therefore, individuals learn appropriate attitudes and behaviors from their social context, and the information with the most direct effect is from one’s colleagues (including supervisors) for the following two reasons: first, because many stimulations occur at work when individuals are unsure of how to react, and their colleagues’ opinions and attitudes provide clues on how to respond to complex situations; and second, an organization’s employees may initially exhibit a friendly attitude toward their colleagues to show that they have integrated into the working environment, which gradually becomes their own attitude as time passes. Accordingly, Fulk, Schmitz, and Steinfield (1990) believed that media usage in the workplace are mostly co‐constructed through an individual’s mutual interactions, in which the most important influences are the comments, attitudes, and behaviors of an individual’s colleagues and supervisors (Fulk et al., 1987, p. 537).

Only a few have investigated media usage from the social influence perspective. For example, Quan‐Hasse, Cothrel, and Wellman(2005) investigated the process of social influence through the theory of social translucence of technology based on employees’ IM usage at a high‐tech software and service company. The social translucent technology theory was designed to compare computer‐mediated and face‐to‐face environments, thus emphasizing that individuals use social information to guide their behavior in face‐to‐face interactions (Erickson & Kellogg, 2000). Quan‐Hasse, Cothrel, and Wellman(2005) observed that employees spontaneously activated IM immediately after arriving at the office. IM conveyed an individual’s “availability” because it was important for colleagues and friends to know that they were available online. One employee noted, “IM not only provides the opportunities to collaborate; it also plays a role in the collaborative process” (P. 9). The personal information displayed in IM allowed others to perceive the possibility of collaboration; therefore, IM became a standard channel to connect with others in the organization. Thus, the culture of a new type of communication and cooperation was constructed with many problems solved through the employees’ active participation.

From the social influence perspective, Stephens and Davis (2009) investigated the electronic multitasking behavior of employees at a conference, in which employees used personal digital assistants (PDAs), mobile phones, or other communication technology connected to the Internet. The results confirmed that when an organizational employee observed others electronically multi‐tasking at a meeting and realized that their behavior was acceptable in the social environment, the employee increased his/her electronic multitasking behavior in meetings.

2.3. Purposes for IM usage

The reason for using IM by organization members has been analyzed through media richness and social influence, whereas the purpose of IM is to communicate content to other organization members. Early studies on the purpose of IM usage were conducted by Handel and Herbsleb (2002) and Isaacs et al. (2002) tested an IM system called Hubbub, developed by AT&T, and analyzed the content of 139 conversations of 28 Hubbub users. The results showed that more than 90% of the conversations were job‐related (91.4%), of which 31% concerned the arrangement or coordination of work procedures. Similarly, Handel and Herbsleb (2002) examined an IM system called RVM, developed by a multinational telecommunication company, to observe IM usage in four working groups (each working group included 4–28 members) from different organizations in different countries (UK and Germany). Among the 4,242 analyzed conversations, approximately 70% were job‐related (69%), whereas approximately 15% concerned non‐job‐related topics such as greetings and jokes. Approximately 90% (87%) of the job‐related conversations were related to knowledge about the work, and less than 10% were about procedural planning (8%).

Based on the above studies, Cho, Treir, and Kim (2005) classified the purposes of IM usage in organizational settings into three categories: organization of work, activities of knowledge work, and socializing. The organization of work included administrative matters such as asking about availability and establishing and announcing meetings. Knowledge regarding work included matters related to problem solving, information exchange, and requesting suggestions to reach a consensus and enhance mutual knowledge to facilitate task completion. Socializing referred to social dialogue that was not related to the job. Using these three major categories, the authors analyzed the IM conversations of two employees at a Korean tire manufacturing company and observed that the employees mostly used IM for knowledge regarding work (50 and 67%), whereas IM usage for other purposes varied (33% for socializing and 17% for organization of work for one employee; 25% for socializing, and 8% for organization of work for the other employee).

In addition, Cho, Treir, and Kim (2005) investigated whether IM enhanced the boundary spanning activities (BSAs) of employees and compared differences in IM usage across organizational boundaries, such as within a department, across departments, and across organizations. The results showed that employees communicated more via IM with employees in the same department, indicating IM does not promote cross‐organizational or cross‐departmental behaviors. Cho et al. argued that because the IM‐connecting network is a relatively close personal online community, only individuals who knew one another were included; therefore, the authors concluded that IM does not contribute to the development of external business partners or new working relationships. However, after individuals become acquainted with one another (after passing a minimum threshold), relationships can be enhanced through IM.

Studies by Quan‐Hasse, Cothrel, and Wellman(2005) also identified the above three categories for the purpose of IM usage in the workplace. In terms of BSAs, Quan‐Hasse, Cothrel, and Wellman(2005) compared the usage frequencies for three communication media, IM, face‐to‐face/telephone, and e‐mail, at three different working distances: within a working group, within an organization, and outside the organization. The results showed that within a working group and within an organization, the usage frequencies for e‐mail and IM were higher than that of face‐to‐face/telephone, whereas for communication outside the organization, the frequency of e‐mail usage was highest. Although the geographic distances within a working group and within an organization were similar, the employees chose to use computermediated e‐mail and IM, which demonstrated the “local virtualities” phenomenon suggested by Quan‐Hasse, Cothrel, and Wellman(2005); namely, a group of individuals in close geographic proximity communicate with computer‐mediated tools to generate a dense network of cooperation. This “local virtualities” phenomenon could be accounted for from the perspective of social influence. That is, within a working group or within an organization, communication media, such as e‐mail and IM, can provide individuals with sufficient opportunities for mutual cooperation or the opportunity to obtain job‐related information and feedback.

Based on the above discussion, this study investigated the purposes of IM usage within an analysis framework of four purposes: work of organization, activities of knowledge work, socializing, and BSAs.

2.4. IM interruptions

What is an interruption? O’Conail and Frohlich (1995) recorded work performed by two professionals over one week and defined an interruption as “a synchronous interaction which is not initiated by the recipient, is unscheduled, and results in the recipient discontinuing their current activity” (p. 262). This definition has been adopted by most scholars that study IM interruptions (Garrett & Danziger, 2008; Rennecker & Godwin, 2003, 2005). IM users typically receive IM messages without warning, and the presence of an IM is typically accompanied by a sound or picture to remind the recipient to respond to the message. Therefore, an IM interruption is often an unexpected communication event that is initiated by others that interrupts an originally established workflow; thus, an individual stops working on or loses focus on the current task (Jett & George, 2003).

According to the information processing and the symbolic interactionist perspectives, IM is a mechanism of information processing for organization members to establish a shared system of meaning in an interactive process of consultation and feedback. In addition, IM intervention appears to be a product that involves continuous consultation and feedback by organization members in the process of seeking consensus. However, will IM interruption be different because of the different degrees of perceived IM media richness? Previous studies on IM interruptions from the perspective of media characteristics showed two different results: first, IM does not interrupt work.

Nardi, Whittaker, and Bradner (2000) observed and interviewed 20 employees from three organizations regarding their IM usage, and they observed that the media features of IM provided better control for message recipients. They argued that the common face‐to‐face or telephone communication methods often caused a fundamental asymmetry in conversation. Namely, during phone or face‐to‐face communications, the message sender mostly decides the time and subject of the conversation and receives an instant response, whereas the recipient is forced to respond to the topic as directed by the sender; thus, it is easy to interrupt the message recipient (p. 5). Nardi, Whittaker, and Bradner (2000) suggesteded that IM can balance the asymmetry of this interaction because it provides the opportunity to “negotiate conversational availability” between the message recipient and sender. The recipient can decide whether to instantly respond to an IM message using the functions of a delayed response, by showing the status bar as busy, or by the “plausible deniability of presence,” which is also acceptable to IM users. These observations, that IM interruptions are less disruptive than telephone or face‐to‐face communications, were also supported by subsequent studies. For example, the above study by Quan‐Hasse, Cothrel, and Wellman(2005) analyzing high‐tech software and service companies observed that employees used face‐to‐face/telephone interactions less frequently because these two communication media were more likely to interrupt the ongoing tasks of colleagues than IM.

The second result from previous studies on IM interruptions from the perspective of media characteristics is: IM interrupts work. Rennecker and Godwin (2003) indicated that IM might lead to interruptions through its three media characteristics: status display, the pop‐up recipient notification, and polychronic communication. They argued that although IM offers users several methods to control the frequency and time of interruptions (for example, by displaying “busy” in the status bar or delaying a response), these IM characteristics extend the amount of time and geographic scope for users to be contacted and thus increase the chance of a “spontaneous interaction,” which constitutes an “interruption.”

However, from the perspective of social influence, organization members learn the proper methods of acquiring information from a social context; therefore, if an individual’s colleagues believe that using IM will cause an interruption, then IM is an interruption and vice versa. Prior empirical studies yielded the two likely results as well: first, IM is an extra communication tool and can interrupt work. Herbsleb et al. (2002) observed IM usage for 17 months in six work groups from different organizations and found that most employees considered IM a superfluous communication tool and that the use of IM was “further encroaching on their time to do real work” (p. 176). Cameron and Webster (2005) also argued that IM is a redundant communication medium for an organization. According to the authors’ results from 19 respondents at four organizations, more than one‐half of the respondents thought that the media characteristics of IM were interruptive, and the most common complaint regarding IM was “the tendency for IM messages to break one’s concentration while focused on anther task” (p. 97).

The second results demonstrated by prior empirical studies in terms of IM and interruptions is that IM is an integrated communication tool and does not interrupt work. Garrett and Danziger (2008) argued that IM is not a source of interruption at work; rather, IM is a good communication management tool for the workplace. For the message sender, IM provides a more unobtrusive method to determine someone’s availability compared with face‐to‐face interaction or telephone communication. For the message recipient, sending a message to announce his/her own status, ignoring a message, or not immediately processing an IM message are all acceptable social behaviors. In addition, IM also allows users to more easily integrate job‐related and job‐unrelated communications by passing on a simple greeting between friends or family to maintain communication with one another without spending a long time maintaining contact after work (Nardi, Whittaker, & Bradner, 2000)

4.1. Design and sample

In this study, the research participants were employees in Taiwan who “used a computer.” They were sampled from the major occupational categories used by the Directorate‐General of Budget, Accounting, and Statistics (DGBAS), Executive Yuan, Taiwan in 2010.¹⁾ Since employees in categories 5–9 of the 10 categories, including occupations in the fields of agriculture, equipment, and labor, had minimal opportunity to access a computer and the Internet at work, only five of the occupational categories were included in the study. These categories included elected representatives, executives and managers, professionals, technicians and associate professionals, service staff, and servicemen.

Because of limited resources, the present study employed the convenience and snowball sampling method to recruit research participants. This sampling method was also used in previous studies on media usage in organizations (Stephens & Davis, 2009; Timmerman, 2002). The questionnaire survey was administered for approximately nine weeks from late April to late in June 2010. Either an electronic or a paper version of the questionnaires was distributed and collected based on the respondents’ requests. Altogether, 283 valid questionnaires were collected, including 131 paper and 152 electronic versions. Among the 283 questionnaires, the highest proportion was from professionals (39.6%) followed by service support staff (27%), technicians and professional assistants (22.5%), elected representatives, supervisors, and managers (9.8%), and servicemen (0.4%). This pattern of distribution of professions was the same for both paper and electronic versions of the questionnaires. The overall results are shown in Table 1.²⁾

Table 1.

Demographics of IM users

4.2. Measurement

4.2.3. Social influence

The present study measured social influence with eight items (α=.84) based on the scales developed by previous media usage studies in Taiwan (Liu & Ku, 1997; Su, 2004). The items were selected and revised to fit the current context. The items are listed in Table 2.

Table 2.

Scale items and descriptive statistics for media richness, social influence and interruptions

4.2.4. Purposes for IM work

Sixteen items were used to measure three purposes for IM usage based on Cho, Treir, and Kim (2005). These include 9 items for organization of work, 6 items for activities of knowledge work, and 1 item for socializing. The respondents were asked to check all the appropriate answers for the purposes of their IM usage in the office based on 16 provided items (multiple choice). BSAs were measured in four categories: 1) colleagues in the same department, 2) colleagues in a different department, 3) work‐related personnel from a different company, and 4) friends and family. The respondents were asked to check the appropriate answers for their most frequent IM communication partners (single choice). The details of the items are listed in Table 3.

The results showed that the main purpose of IM usage was knowledge work (39.2%) followed by organization of work (34.6%) and socializing (9.6%). BSAs were mostly with friends and family (34.7%) followed by work‐related personnel across different companies (26.4%), colleagues in the same department (22.3%), and colleagues in different departments (16.5%) (Table 3).

Table 3.

Purposes of IM usage

6.1. The reasons for using IM

This study observed that the remarks, attitudes, and behaviors of employees’ supervisors and colleagues regarding IM usage were the most important predictive variables for IM interruptions at work. Therefore, in a social environment where supervisors and colleagues often make positive comments regarding IM, where employees’ superiors and colleagues use IM, and the experience of using IM is frequently discussed with supervisors and colleagues, such employees experience more IM interruptions. From the perspective of social influence, information that frequently appears in the social environment becomes the cognitive signals for the proper attitude of an organization’s employees. When supervisors and colleagues show positive attitudes regarding the use of IM, IM interruptions become a component of the common construct of social significance. The process of such interruptions can be considered a rationalization process of the organization’s employees (Salancik & Pfeffer, 1978). Through the support of the collective construction of the social environment, personal behavior is rationalized. Individuals develop an acceptable rationale for their behavior so that such behavior becomes meaningful. An important reason for making IM interruptions a component of one’s job is that all of the individual’s colleagues have demonstrated the same attitudes and needs. With organizational norms where supervisors and colleagues all interact using IM, employees activate IM first thing at work, connect to the standard contact network of the company, and show their status as “available” to set themselves in the state of “always‐on”, as described by Quan‐Hasse, Cothrel, and Wellman(2005). Such employees thereby experience more IM interruptions.

Additionally, the “local virtualites” phenomenon proposed by Hasse, Cothrel, and Wellman(2005) can also explain why social influence shows a greater influence on IM interruptions. If a group of individuals in close geographic proximity (in the same office) interact using a computer‐mediated tool, a dense network of cooperation is generated among them. Because of the interactive effect of this dense network of cooperation and its social influence, individuals learn about their colleagues’ needs, values, and demands when they interact, and IM usage is more likely to interrupt their work.

Compared to IM choice because of social influence among employees in an organization, the results showed that IM choice because of media richness associated with fewer IM interruptions. Among the four aspects of IM richness, only instant feedback and personal focus showed a higher correlation with IM interruptions. Notably, the descriptive statistics showed that the multiple communication cues provided by IM caused the highest score on the richness assessment (mean=3.81) but not instant feedback (the mean for these two questions were 3.59 and 3.62) or personal focus (the mean for these two questions were 3.30 and 3.43) (Table 2). Therefore, from the perspectives of information processing theory and symbolic interactionism, instant feedback that can instantly interrupt one’s workflow and personal messages that can enhance one’s personal sense of social presence increase the likelihood of disturbances for organization members and thus increase the degree of interruptions in personal feelings. However, although the media characteristics of IM can easily interrupt employees’ workflows, if an individual can control and manage the media characteristics of IM in the process of consultation, feedback, and the construction of meaning (Nardi, Whittaker, & Bradner, 2000; Quan‐Hasse, Cothrel, & Wellman, 2005), even if the individual perceives that IM has the most abundant multiple communication clues, he/she should not experience significant IM interruptions.

For practical implications, the present study suggests that instead of worrying about employees being lazy and idling because of IM media richness, the management team of an organization should try to utilize the social influence established by interactions among its employees and construct a new organizational culture of using IM for communication and problem solving. With the increasing richness and diversity of various communication media, such as social networking websites and smart phones, instead of tight control, organizations should take advantage of the media characteristics to establish a new organizational culture that encourages its positive use. Under social influence, even a consensus on media richness must be established through organizational interactions (Markus, 1994).

6.2. Purposes of IM usage

Overall, the IM purposes investigated in this study were not correlated with IM interruptions. Namely, the reasons for employees of organizations to choose to use IM can better predict the extent of IM interruptions than the communication purposes of using IM. Therefore, from the perspective of social information processing, the organizational attitude and behavior of IM usage constructed by all members of an organization determines the nature of the work. In this case, the individual purpose of IM usage was not important.

In terms of BSAs, previous studies have shown that IM usage does not promote BSAs (Cho, Treir, & Kim, 2005; Quan‐Hasse, Cothrel, & Wellman, 2005). However, the results of this study showed that in addition to family and friends, work‐related personnel from different organizations were the major communication partners followed by colleagues in the same department and colleagues in a different department (Table 3). It seems that office workers in Taiwan often participate in BSAs with different organizations using IM and have fewer BSAs within their department. However, what is the content of such BSAs?

This study further conducted a chi‐square analysis of the communication partners of BSAs in combination with the other three purposes of IM usage, and the results showed that the communication partners were not significantly correlated with organization of work (x2=4.36, df=3, p>.05); yet, they were significant with activities of knowledge work (x2=10.65, df=3, p<.05) and socializing (x2=43.82, df=3, p<.001). That is, for IM usage with the purpose of activities of knowledge work, the communication partners were mostly job‐related personnel, regardless of whether the partners were from the same department, different departments, or different companies. For the purpose of social activities, the communication partners were mostly friends and family. Therefore, although face‐to‐face communication was important to coordinate complex cases of cooperation and problemsolving in the workplace, long‐distance teamwork using new technology has become a widely accepted mode of operation. IM users have various different types of communication channels and can develop a specific channel that best suits their needs.

In addition, although the results of the study showed that the frequency distribution for the purposes of using IM was identical to previous studies (Cho, Treir, & Kim, 2005; Isaacs et al., 2002; Handel & Herbsleb, 2002), the previous studies were all based on the analysis of actual collected dialogue from IM users. The present study was based on a self‐reported questionnaire to investigate the purpose of IM usage. Therefore, to play the role of “good subjects” and meet the “demand characteristics” (Wimmer & Dominick, 2002), the respondents might have been reluctant to check “socializing” as the IM usage purpose, thus resulting in a higher probability of activities of knowledge work being identified. Simultaneously, did this effect also influence the distribution of the interaction partners identified by the respondents so that the actual proportion of friends and family was higher? For future studies, the IM interaction content of the organization employees could be obtained to investigate the relationships between IM interruptions and the purposes of IM usage for BSAs, organization of work, activities of knowledge work, and socializing.

For practical implications, the results of this study should be a relief for management. Although both knowledge regarding work and socializing interrupts work, if one compares the results to the reasons for IM usage, the interruptions caused by IM for the purposes of IM usage were not significant. In other words, if the social influence of IM usage for the employees in an organizational setting is properly managed, it is not important whether the employees work or chat using IM; that is, the purpose of IM usage remains controlled.

6.3. IM interruptions

Similar to many other studies of IM interruptions, this study also considered interruption an invasive behavior, although not enough attention was paid to both its positive and negative consequences (Jett & George, 2003). Rennecker and Godwin (2003) suggested that the determination of whether an IM interruption causes a positive or negative effect on an individual’s productivity depends on the individual work content at the time of the IM interruption and the content of the interruption itself. The authors provided six observations: first, the IM interruption is less likely to have a negative effect if a user is familiar with the work content at the time of the IM interruption; second, a longer and more frequent IM interruption has a greater chance of causing a negative effect; third, an IM message related to work content is more likely to have a positive effect; fourth, an IM interruption is less likely to have an adverse effect if an IM user is close to completing his/her task; fifth, the more complicated the individual’s work, the greater the negative effect of an IM interruption; and sixth, the longer an interruption is, the more negative the effect on individual productivity. Further studies should clarify the interactions that these six interruption variables have on social influence, media richness, organization of work, knowledge regarding work, socializing and BSAs.

Additionally, because this study considered interruptions as invasive behaviors, three progressive questions were developed to measure an IM interruption, from “I often receive messages from IM at work” (mean = 3.48) to “I am often interrupted by incoming IM messages at work” (mean = 3.22) to “I leave everything at hand and respond to the messages at once” (mean = 2.49). Although the overall mean for the scale was 3.06, a careful review of the score of each of the questions showed that the consistency among the employees regarding a state of progressive interruption gradually declined with mean scores from 3.48, 3.22 to 2.49. The distribution of the mean scores indicated that the organization’s employees might receive IM messages often, but the messages might not necessarily interrupt their ongoing work, or IM messages might interrupt their ongoing work, but the employees do not drop everything to interact on IM.

To understand the differences among these three types of IM interruptions, this study conducted a hierarchical regression analysis to investigate the correlation of each interruption question with the reasons and purposes of IM usage. The results showed that for the interruption question of “I often receive messages from IM at work,” the social influence (Beta=.35, p<.001) and media richness (Beta=.15, p<.05) and socializing (Beta=.14, p<.05) showed the highest correlations. For the interruption question of “I am often interrupted by incoming IM messages at work,” the social influence (Beta=.26, p<.001) and socializing (Beta=.06, p<.05) showed the highest correlations. Finally, for the interruption question of “I leave everything at hand and respond to the messages at once,” only social influence (Beta=.27, p<.001) predicted this interruption type.

Therefore, for each of the three different interruption cases, social influence was the strongest predictive variable of IM interruption. In addition to social influence, media richness and the purpose of social activities were also correlated with interruptions: if the employees in the organization used IM for social activities because of the media richness of IM usage, these employees felt interrupted when “I often receive messages from IM at work.” If the employees in organizational settings often used IM to socialize, they were more likely to feel interrupted when “I am often interrupted by incoming IM messages at work.” Thus, there were different types of interruptions, and future research can offer a more detailed discussion and analysis of interruption as a research concept, including methods for its measurement (Jett & George, 2003).

In short, the social influence from colleagues and supervisors allows IM to more easily become an interruption of one’s ongoing task. However, IM has sufficient immediacy and provides enough personal clues to allow the members of an organization to discuss job‐related knowledge so that the two sides of a conversation or a group can share needed text, images, or files. The results show that, although IM can interrupt ongoing tasks, in a workplace with the commonly used media of communication, IM may indeed provide users an opportunity to “negotiate availability,” and the ability to receive feedback and share information makes IM different from other types of communication. However, further study is required to determine whether such interruptions have different impacts on the overall operations of an organization. Therefore, more detailed studies based on the type of IM interruptions will provide a more precise understanding for management teams to facilitate an organization’s overall operations.

Although this is the first study that has attempted to systematically link the relationships between the reasons for IM usage, the purposes of IM usage, and the employees’ IM interruptions, the results can only provide suggestive observations instead of a typical conclusion. According to the results of this study, future research may continue with the observations of different subjects, environments, and times (Cook & Campbell, 1979) to better understand the use and interruptions caused by the increasingly rich communication media in organizations.