........................................................................................................................................................................................................................................................

new media & society

Copyright 2000 SAGE Publications London, Thousand Oaks, CA and New Delhi Vol2(2):157179 [14614448(200006)2:2;157179;012223]

ARTICLE

Dening interactivity
............................................................................................................................................................................................................................................

A qualitative identication of key dimensions

............................................................................................................................................................................................................................................

EDWARD J. DOWNES Boston University

............................................................................................................................................................................................................................................

SALLY J. McMILLAN University of Tennessee

............................................................................................................................................................................................................................................

Abstract
The literature on interactivity includes many assumptions and some denitions but few tools for operationalizing the concept of interactivity in computer-mediated environments. This article takes an early step in lling that gap. In-depth interviews with 10 individuals who work and teach in the eld of interactive communication led to a conceptual denition of interactivity based on six dimensions: direction of communication, time exibility, sense of place, level of control, responsiveness, and perceived purpose of communication. Suggestions are made for applying these dimensions to multiple forms of computer-mediated communication. Future research should empirically test the existence and application of these dimensions.

Key words
computer-mediated communication interactivity

The word interactivity has been used to describe products ranging from snoring dolls and web-based brochures to video games and online transactions. Scholars have employed the term to refer to everything from face-to-face exchanges to computer-mediated communication. However, much of the literature, both popular and scholarly, uses the term
157

New Media & Society 2(2)

interactivity with few or no attempts to dene it. Even when denitions are found they are often contradictory. Such imprecision invites research, such as that reported by this article, which adds shape and form to the concept of interactivity and suggests necessary foci for a conceptual denition. This article begins with a synopsis of the literature and explanations therein of the term interactivity. Based on concepts found in that literature, the authors asked individuals who are involved with emerging communication technologies to provide their insights about interactivity in computer-mediated environments. A qualitative analysis of their responses to interview questions revealed multiple themes that seem to underlie the concept of interactivity. After careful consideration of both the concepts found in the literature and the themes that emerged from the interviews, the authors propose a conceptual denition of interactivity based on six dimensions. Finally, suggestions are made for how this emerging denition can be used in future research. LITERATURE REVIEW An extensive body of literature exists about interaction in human communication. Much of this literature grows out of a sociological tradition. For example, the simultaneous transaction model addresses realtime, interpersonal exchange between individuals and is often framed in terms of interaction (DeFleur et al., 1997). Jensen (1998: 188) suggested that, from the sociological perspective, interactivity is the relationship between two or more people who, in a given situation, mutually adapt their behavior and actions to each other. Another body of research examines interaction of human beings with computers. This literature grows out of the computer science tradition and focuses on improving the interface of computer hardware and software. Thomas (1995: 2) described the study of humancomputer interaction as a situated, particular, detailed, gritty, errorprone, and largely nondetermined reality. Jensen (1998) reported that researchers in the eld of humancomputer interfaces have identied the style of control that exists between the human and the computer as the key determinant of interactivity. Only in the past 10 to 15 years have scholars in the mass communication tradition begun to examine the nature of interactivity in computer-mediated communication. Rafaeli was one of the early investigators of interactivity in the mass media context; in a 1990 study, he examined interactivity in the context of traditional media. However, much of his work has focused on computer-mediated environments. Rafaelis earliest denition of interactivity (1988: 111) was: An expression of the extent that, in a given series of communication exchanges, any third (or later) transmission (or message) is related to the degree to which previous exchanges referred to even earlier
158

Downes & McMillan: Dening interactivity

transmissions. In a 1997 study conducted with Sudweeks, Rafaeli revised that denition to: The extent to which messages in a sequence relate to each other, and especially the extent to which later messages recount the relatedness of earlier messages. Rafaeli and Sudweeks (1997) identied interactivity as a theoretical construct that grapples with the origins of captivation, fascination, and allure that can be inherent in computermediated groups. We chose to center our search for the denition of interactivity on this relatively new body of literature that focuses on computer-mediated communication. Examination of technologies such as the internet and its graphical subset, the world wide web, has led researchers to the conclusion that interactivity is a key advantage of such media (Morris and Ogan, 1997; Pavlik, 1996). But what is that interactivity? Answering this question is not a simple task. Braman (1989), who developed an approach to dening information, noted that denitions of core communication concepts (e.g. information and interactivity) are critical for policy makers and may dramatically impact society. Nevertheless, attempts to settle upon a single denition are problematic because these concepts are multi-faceted and because multiple denitions apply concurrently. Like information, interactivity is not a monolithic concept. But, because it is an emerging eld, examination of interactivity must be narrowed. We have chosen to narrow our examination of interactivity by focusing primarily on how individuals perceive interactivity in the context of computer-mediated communication. The literature of interactivity includes multiple concepts that help to explain how individuals perceive interactivity in the context of computerbased communication. For example, Heeter (1989) suggested users exert more effort when they attend to interactive media than to traditional media forms. However, McMillan (1998) found it was difcult to operationalize this concept of effort in an analysis of web sites. Therefore, the relationship of effort to interactivity requires further examination. Role taking and feedback are two additional concepts that have appeared in several studies of interactivity. Rice (1984: 35) suggested fully interactive media imply that the sender and receiver roles are interchangeable. More than a decade later, Rogers (1995: 314) echoed the same theme when he dened interactivity as the degree to which participants in a communication process can exchange roles in and have control over their mutual discourse. Steur (1992: 84) dened interactivity as the extent to which users can participate in modifying the form and content of a mediated environment in real time. Rice and Williams (1984) also focused on real-time communication and suggested that media are interactive if they have the potential for immediate, two-way exchange. But not all observers agree
159

New Media & Society 2(2)

about the importance of real time. For example, Rheingold (1993) suggested that the asynchronous characteristics of tools such as email, newsgroups, and listservs is one of the key benets of these interactive media. The importance of time in interactive communication needs to be explored in more depth. Chen (1984) suggested passivity and interactivity are qualities of individuals making use of media, not qualities of the media per se. Other researchers also suggested that individual uses are more important than media features in determining interactivity (Kayany et al., 1996; Walther, 1994). Some researchers, however, have argued that certain technologies permit more interactivity than others. For example, Snyder (1996) focused on ways in which the non-linear nature of hypertext enhances interactivity. Changes in patterns of control are another issue researchers and observers are beginning to link to interactivity. As noted above, Rogers (1995) identied control over mutual discourse as a key element of interactivity. OKeefe (1995) noted the web provides a two-edged sword: on the one hand, organizations have a robust media environment in which to tell their own story in their own words; on the other hand, they cannot control exactly what route individuals will take after they arrive at a web site. Finn (1998) suggested that the sender/receiver ratio of control in content creation, presentation, and preservation is a key dimension of computerbased communication information systems. Ha (1998) identied information collection as a key dimension of interactivity. Blattberg and Deighton (1991) also identied audience tracking as a key advantage that computer-mediated systems offer marketing communicators. Dreze and Zufryden (1997) noted the importance that marketers attach to tracking users in interactive environments and the concerns of consumers about losing their privacy in the process. However, the potential benet for users is that, as their activities are tracked, messages can be customized to match their interest. Other studies have focused on advantages and disadvantages of interactive media. For example, Ang and Cummings (1994) found that computermediated environments enhance information-seeking. Schaffer and Hannan (1986) found that recall was signicantly enhanced by increased interactivity. Sproull and Kiesler (1991) reported that, within organizations, electronic work groups can be as efcient or more efcient than face-to-face work groups. Finally, some researchers have associated disadvantages with computermediated communication. For example, Markus (1994) reported that electronic mail can lter out personal and social cues which may result in aming, public rebuke of senders, and depersonalization. Other commentators have suggested computer-mediated communication goes
160

Downes & McMillan: Dening interactivity

beyond simple disadvantage to real or perceived threat. For example, Stolz (1995) criticized computer-based communication as a threat to real world interaction among people in public places. In summary, the literature suggested that examination of interactivity should include attention to: user effort, sender and receiver roles, timeliness, characteristics of both the medium and the communicator, control, activity tracking, advantages, disadvantages, and potential threats. METHOD Based on the literature, we developed a structured interview instrument (Appendix 1) for use with a panel of experts (Appendix 2). This structured format, rather than one that was unstructured or more loosely structured, met our objectives. We wished to limit the interview to exploring the concept of interactivity while opening the door for serendipitous insights into related issues which came from the initial responses to the questions (Singleton, 1993). We believed it was necessary to qualify interactivity before we quantied it. Thus, our purpose in conducting the interviews was not so much to help us to know what interactivity is, but rather to help us to understand and to explain it. We sought breadth of informed opinion and shared intuition. Thus, we chose to explore the concept of interactivity in depth with a few people rather than examining broad-based conceptions of interactivity among a larger sample. Data were gathered from the 10 respondents in face-to-face interviews conducted in January and February 1998. These respondents all meet Marshall and Rossmans denition of elites experts who are considered to be inuential, prominent, and/or well informed people in an organization . . . (1989: 113). All had been working with new technologies for several years; all had taught for several years. Descriptions of each respondent are provided in Appendix 2. Real names are not used because respondents were promised anonymity. The tradition of interviewing a relatively small number of experts such as these as a means to uncover a social, cultural, economic, political, technical or other phenomenon is fully recognized in the literature on qualitative methods (Denzin and Lincoln, 1994; Lee, 1999; Mason, 1996; Patton, 1987; Strauss, 1996; Marshall and Rossman, 1989). The advantages to conducting interviews with elites is that such interviews allow for the discovery of complex interconnections in social relationships; facilitate analysis, triangulation and validity checks; generate working hypotheses; and provide great utility for uncovering the subjective side of a phenomenon (here, the concept of interactivity) under investigation (Marshall and Rossman, 1989).
161

New Media & Society 2(2)

We chose to interview these experts for two additional reasons. First, our ontological position suggested that the respondents knowledge, views, understandings, interpretations, experience and interactions . . . [were] meaningful properties of the social reality under exploration (Mason, 1996: 39) that could be best teased out via the structured interview. And second, that our epistemological position suggested that a legitimate way to generate data on those ontological properties (was) to interact with people, to talk to them, to listen to them, and to gain access to their accounts and articulations (Mason, 1996: 40). A script of 10 questions (see Appendix 1) was written before interviews began, and each question was asked in the order presented. Occasionally a follow-up inquiry was made (Tell me more about that . . . Can you expand on that thought?). Each interview lasted between 20 and 90 minutes with an average time of 40 minutes. We had several qualitative methodological options. We chose to utilize structured interviews because, on balance, they appeared the best method by which to realize our goals of: (1) uncovering the subjective side of how the notion of interactivity was viewed; (2) providing background context and hypotheses for a future, quantitatively driven research phase; (3) documenting the evolving nature of the concept of interactivity; and (4) quickly obtaining large amounts of contextual data (Marshall and Rossman, 1989; Denzin and Lincoln, 1994). The collection and analysis of these data is rooted in a phenomenological paradigm which holds that reality (here, the reality of those who both teach and practice computer-mediated communications) is socially constructed (Firestone, 1987; Strauss, 1996). Our respondents were in an ideal position to identify the concepts and parameters of interactivity and to tell what the notion, in its present state, is. Further, the afrmatives represented in the themes provided rich insights into feelings, personal experiences, empathies, emotions, intuitions, subjective judgments, imaginations, and the diverse forms of creativity held by respondents (Rosoneau, 1992). After interviews were completed, we read and re-read transcripts looking for responses that would provide shape and form to the concept of interactivity. Each of the primary researchers, as well as two graduate students who are developing expertise in the impact of new technologies on communication management, read the transcripts. Each sought pervasive themes. Each read and re-read sections of the data, sometimes as much as a dozen times. We then reconvened, read the transcripts aloud, and determined themes on which there was wide agreement. The ndings section below discusses only those themes on which at least three of the four coders agreed. Naturally, every effort was made when writing results to allow the words of respondents to dene the parameters related to the concept under investigation.
162

Downes & McMillan: Dening interactivity

FINDINGS The interview data are reported below in three major categories: impacts, messages, and participants. Impacts Respondents comments about the impacts of computer-mediated interactive communication centered on three concepts: (1) revolution; (2) consequences; and (3) uncertainty. Revolution. Without exception, respondents talked about the impacts of interactive computer-mediated communication on individuals, organizations, and society. But they were not always consistent in their evaluations of the impacts of cyber-interactivity. Is it something revolutionary or just an adaptation of very old communication models? For example, Pam suggested that interaction has long been studied by sociologists and that computermediated interaction is really nothing new. Dennis suggested that the literature on interpersonal and organizational communication holds the key to understanding interactivity. Both Jake and Will noted that computer scientists have been studying interactivity for years in the context of humancomputer interfaces. And Jeff suggested that interactivity is the traditional communication model with a robust feedback loop. Thus, while parts of interactivity may be revolutionary, other parts seem to be nothing new. Some respondents lauded the revolutionary nature of the computer as a communication medium. For example, Joseph said the computer claims to do it all video, sound, music, good quality, text, and images and voice. This medium was also portrayed as breaking down walls between advertising and selling, marketing and fulllment, and organizations and their publics. However, some respondents identied unique aws in the medium. For example, two respondents noted specic problems with email. Pam said: Because it [email] is so casual, so fast, you dash it off, forgetting about the impression on your colleague. On a similar note, Jeff shared his impression that, few receivers seem to read email carefully and thoughtfully. Despite the highlighting of both positive and negative revolutions, most respondents talked about interactivity in relatively traditional terms. Most referred regularly to senders and receivers, creators and audiences, and developers and users. However, several suggested that new relationships were developing among those who create and those who consume messages. For example, Joseph suggested that the computer-mediated environment allows for three-party interactivity involving two people and a computer. While the terms sender and receiver were used frequently to suggest a kind of dichotomous relationship, many also expressed the belief that in interactive communication the roles become interchangeable.
163

New Media & Society 2(2)

In sum, respondents viewed interactivity as something old, something new, something borrowed from other disciplines, and something both people and computers do. While no clear consensus emerged about the revolutionary nature of the medium, most respondents used terminology, such as sender and receiver, that suggests a traditional communication model. Consequences. Regardless of their opinions about the revolutionary nature of computer-mediated communication, respondents were fairly unanimous in their belief that this new communication form has, and will continue to have, far-reaching consequences. Joseph illustrated this consensus when he said: It threatens whole industries, it threatens whole professions. Dennis expanded this discussion primarily to communication professionals:
For those people who have traditionally distributed information in a one-tomany model, interactivity is changing the way you can do or should do business because the opportunity now exists to put control in the hands of the user. The implications of that, I think, are truly interesting things about interactivity.

Amy discussed consequences in terms of the organizational communication environment:

The old paradigm of if knowledge is power, he who owns the knowledge owns the power means that sharing knowledge can undermine ones security in corporate America. That doesnt mean its doomed to failure, just that many cultural issues must be addressed before you can reasonably expect people to share.

Bob brought the concept of consequences to the personal level: One cannot anticipate the consequence of getting out on the web and providing information thats important to you. He said interactivity can be empowering to the individual operating outside the corporate structure:
There is quite a bit of anarchy out there which is encouraging. The fact that I can become a spokesman for solar power, not because I spent the money to get a solar powered house built, but because of a site on the world wide web is, I think, totally threatening to the power company. Because I have a voice now. I think the threat is to existing institutions and the old ways of doing things.

Interactivity may represent a threat to institutions and professional communicators at the same time that it creates new opportunities for individuals participating in a collaborative and interactive environment. But the balance of power is not yet clear. Rita poetically summed up the potentials: The misrepresentation that you can do with multimedia can be
164

Downes & McMillan: Dening interactivity

tremendously bad. Also, the other side is that you let other people create their dreams. Dreams and nightmares. In sum, respondents were in general agreement that computer-mediated interaction has far-reaching consequences with the potential to change entire industries and forge new paradigms. Their comments suggest that if the power shifts they envision unfold, communication management may become an oxymoron. Uncertainty. The above discussion of both the potentially revolutionary nature and consequences of computer-mediated interactive communication reveals an underlying reality: even experts such as the subjects of our interviews are still uncertain about the exact impacts of this new communication form. Respondents disagreed with each other about key concepts such as the importance of real time in interactive communication, the amount of effort interactive communication requires, and even whether human beings are an integral part of interactive communication. Internal disagreements were also found. For example, Joseph began by asserting that interactivity is typied by conversation but later indicated that transactions are one of the most intensely interactive processes. Furthermore, phrases such as I am not sure yet, and I cant dene it, but I know it when I see it were used by many of the respondents. Discussions of interactivity were characterized by a confusing mix of terms. For example, Rita used the terms interactivity and multimedia almost interchangeably. When asked whether the terms were synonymous she replied: Yes and no. All things in life are multimedia. The entire spectrum is also interactive. The new-technology multimedia works on the same concept with different tools. Multimedia for me is ancient. Interactivity is ancient. The tools are new. The concept of hypertextual links was also often mixed with the concept of interactivity. For example, Will said: I can rearrange, sometimes, the content where I am no longer bound to a linear fashion because of hypertext. So that would be an interactive multimedia experience. In sum, the impacts of interactive, computer-mediated communication technologies are still evolving. Even the experts are not yet certain exactly what the concept means. However, all of the respondents seem to suggest that interactivity is a multi-faceted concept. And despite occasional disagreements, it seems that two key components of interactivity are the messages themselves and the people who participate in interactive communication. Messages Three message-based concepts emerged from the interviews: (1) the nature and direction of messages; (2) the importance of time to message structure
165

New Media & Society 2(2)

and retrieval; and (3) the creation of a sense of place in computer-mediated environments. Nature and direction. Most respondents said the computer is closely tied to their concepts of interactivity. Some believe the computer provides an ideal medium for facilitating communication. For example, Joseph said: The internet was designed with interactivity in mind. All of its technical makeup and its design from the beginning were interactive. He also indicated that one of the key advantages of the computer as a medium is that it can integrate functions that used to be split between active and passive media. He illustrated with a comment on interactive advertising:
Right now advertising has little to do with buying and selling. Its done through a different medium, at a different time, in a different place, with a different purpose. Right? We dont advertise to sell something. We advertise to change someones attitude or image or proclivity (mental state). And we do it through a passive mass media. Buying and selling, we do in a whole different atmosphere, whole different system that involves transactions. Now we have a medium that can do both.

Jake sought balance in describing the medium: The important thing still . . . is the humans behind the medium. Having said that, it is important to have a good medium so you can interact. So if you have email or the internet you have a good medium for interaction. Others, however, expressed concern that the computer is less than a perfect medium for some types of communication. Kent shared a story of cyber-irting. He met a woman online through a matchmaking service. They exchanged email for several weeks. I was convinced I was in love with this person, he said. They shared many things in common, enjoyed the same humor, shared a depth of knowledge about popular culture. He saved all her email messages. I could go back and laugh at it later days later, hours later. I could go back and think about that funny thing which you cant do as well with phone conversations. Within a few weeks they talked on the phone and he knew instantly that the magic that existed in email was missing from their voice conversation. They met, and were further convinced that the relationship they had built online just didnt exist in the real world. It was not that either was particularly unattractive, Kent said, it was just that in the unedited environment of real-time exchange by voice or face-to-face communication they could not experience the same excitement and fun that they found in carefully crafting messages by email. Kent summarized the experience:
I dont know what to take out of there but there is denitely something interesting there. Interaction via email. Real time interaction versus chat. And how that is different from real time interaction via voice and meeting face-to166

Downes & McMillan: Dening interactivity

face. Timing makes a difference and meeting makes a difference. Interesting experience, but the medium can alter your opinion of someone.

It may be that all interactive communication must be two-way communication. However, some forms of computer-mediated communication are much more like traditional one-way communication than others. For example, Pam noted Just because you get at a computer and you start pushing buttons, or you get to jump from this site to that site, youve got some choice. She saw this as a form of interactivity. Web sites that provide online brochures for an organization might require user activity; the viewer may select options from a menu. But this is not much different from turning the pages of a physical brochure. In both the web site and the brochure, the communication is primarily one-way: from the sender to the receiver. Most respondents assumed that interactive communication must be twoway. They talked about how the technologies inherent in the internet allow for a robust feedback loop. And they also talked about how individuals can shift between sender and receiver roles. The literature also seems to assume two-way communication. The Rafaeli and Sudweeks (1997) description of the recursive nature of interactivity assumes a two-way ow of communication; and both Rice (1984) and Rogers (1995) suggest that interactive communication requires that sender and receiver roles be interchangeable. While it might be tempting to discount one-way communication as something less than interactivity, to do so would fail to recognize some unique forms of communication that are evolving on the internet. For example, Jake talked about how he uses searchable databases to facilitate group work. These databases provide a resource for remote education and are valuable forms of communication that do not have exact parallels outside of the environment of media such as the internet. These content-rich sites typify a form of one-way communication in which the receiver exerts primary control over the message. In sum, while many respondents lauded the benets of this new medium, several also indicated concern about the limitations inherent in computerbased interactivity. In particular, they experienced a sense of falseness when such communication became personalized and called for a deeper, shared meaning. Most respondents described interactivity as two-way communication, but some also recognized interactive communications that are more one-way in nature. Time. As some respondents talked about interactivity they suggested that the closer to real time an interchange, the more interactive it becomes. Kent said: This is probably because most of the interaction in the past was in real time. Dennis also stressed the importance of timeliness: I think it [real-time
167

New Media & Society 2(2)

communication] is incredibly important. Almost by denition, interaction requires immediate response. Will echoed similar ideas and placed them in an interpersonal context: The closer you can get to the human exchange the more valuable it becomes. Other respondents did not believe timing of messages is central to interactivity. Jake said: Real-time is not at all important. I can send mail to you and if you are there, you can give me the immediate response. But if you are not, you can log in a couple of hours later or another day. And still you will be able to respond to what I said. Which to me is a plus. Pam took a situational perspective. She said: Obviously, the closer it is to that immediate time, then, I think it ts . . . its more interactive. Joseph was even more situationally oriented in his discussion of the role of timing: Time is important, but you have to consider it in terms of the nature of the conversation. When do we expect and need immediacy, and when do we expect and need a delay? Timing, our data suggest, may be more important to some of the types of interactive communication than to others. For example, when an individual is communicating with a marketer about a potential purchase, immediate or quick response may be an important measure of interactivity. But, in other cases the ability to respond at a convenient time may be more important. As several interview respondents pointed out, tools such as email allow them to achieve interpersonal communication without co-presence. Thus, the importance of timing seems to be its level of exibility to the demands of the situation rather than its immediacy. In sum, while there was a consensus that media such as the internet allow for interaction to take place at different times, there was a wide range of opinion as to whether real time is necessary for such an interchange to be labeled interactive. A key issue seems to be that participants have some control over the timing of messages in computer-based interactive environments. Place. The nal message-based concept to emerge from interviews was sense of place. Several respondents indicated that interactive computermediated communication can transcend geography to create a kind of virtual place. For example, Kent said the online world seems to create public places again. In these public places people can meet on the basis of what they are thinking and writing rather than on the basis of physical characteristics such as appearance, status, etc. Many respondents referred to the place where computer-mediated communication occurs. For example, Joseph said interactivity goes on right there in that computer and person. Similarly, Bob referred to highly interactive web sites as those that invite you back time and again,
168

Downes & McMillan: Dening interactivity

suggesting that the site visitor is not only viewing but being transported to a specic place. Two respondents focused on ways the multimedia nature of web sites can create a sense of place. Will described a future in which web sites could have the ability to let him pre-tour his vacation location by immersing him in media experiences ranging from text to contextual video. But he noted a potential downside to such a sensory-rich experience: If I can get the full experience in the synthesized form, do I need the place? Similarly, Dennis envisioned the total immersion of virtual reality as an advanced form of cyber-interactivity. By utilizing multiple types of media and response mechanisms, the computer can be completely responsive to his actions. Some of the factors that may help to create a sense of place include greater use of multiple types of media and a greater opportunity for interchanges among the participants. For example, Joseph identied the Gap Kids web site as one of the best places on the web. He spoke of how the site enables the individual to physically participate in a shopping experience including trying on the clothes as one drags and drops specic clothing items onto virtual people. In an interesting twist on the concept of place, Jake rst noted that one of the primary benets of computer-mediated communication is that people dont have to be in the same place at the same time to be able to communicate effectively. But then he went on to note that tools such as email, listservs, and bulletin boards can create a kind of virtual place where a virtual community can come together to collaboratively solve problems. In sum, respondents suggested that the more interactive a computermediated communication environment becomes, the more likely that the individual will feel that he/she has been transported to a virtual place. In those places, they may encounter experiences ranging from a resurgence of public life to new forms of retail shopping. Participants Three concepts emerged from the interviews that relate to the individuals who participate in interactive media: (1) control; (2) responsiveness; and (3) perceived goals. Control. Some respondents focused on control from the senders perspective and some from the users point of view. Pams comments summarize useroriented perspectives about control:
I think control is a critical factor. Maybe its kind of the central factor. Its not just that I determine where I go or what content I view or what buttons I select. Its that I control the timing, I control the content, I control to whom Im going to address this. I control, in some respects, with whom I want to have this exchange and interaction.
169

New Media & Society 2(2)

Dennis also expressed the opinion that interactive media shift control to the users of the medium. He said that this shift can be very troubling to professionals who have become accustomed to controlling the content of traditional mass media:
The true meaning or the consequence of greatest import in interactivity is that we are transferring control to the user. The control of production is one issue, but that only matters to us as professionals if we perceive that the control of production somehow contributes to the control of perception and understood meaning. Every time you give the user more control it means you have even less control over the potential meaning than could be derived from your information.

Not all respondents, however, believed that interactive communication shifts control to the user. Bob acknowledged that such a shift might be an ideal to strive for, but it is not yet a reality:
At this time, the author has the control. The person who has put together the game, or the web site, or the information has the control. They set the parameters for the response on the part of the user. I think as we train students to become authors, to become producers of interactive material, the goal is for them to set the parameters as deep as possible and as wide as possible so that the user perceives that he/she is in control. Are they really in control? Not yet.

In sum, respondents were in general agreement that while the sender has control of the message, the receiver has control as to whether he/she will: (1) indicate to the sender that the message was received; (2) attend to the message; or (3) pretend that it was never sent. Thus, it would seem that the receiver has control over the feedback loop in a way more characteristic of mass media than of interpersonal communication. Responsiveness. As noted earlier, the literature reects some differing perspectives on the amount of effort required for interactive communication. In discussing effort with the experts who provided the insights for this article, we began to realize that effort may be a less meaningful concept than responsiveness. In some cases, the individual participants in interactive communication may need to exert more effort because they are responding to messages they see on the screen. But this may seem like less effort because the interchange with a responsive partner (human or computer) helps to facilitate communication. Kent provided an example of how a responsive system can require both more and less effort:
I think interactivity requires a lot of effort. But on the web it seems that viewers are very passive still. People dont want to work on the web. There is effort, but there is payoff because you have so much information. It is less effort in a way. I would rather make a decision [about purchasing an airline
170

Downes & McMillan: Dening interactivity

ticket] myself and see all the options than work through that person who is reading stuff off the computer. So, I see it as less effort and more empowering.

Among those respondents who believed that interactivity requires effort, most framed that effort in terms of responsiveness. For example, Joseph suggested interactivity requires twice as much effort as traditional mediated communication because communicators have to think about how to present information and also about how to respond to information they receive. Dennis suggested that effort is the denition of interactivity, because if you are giving control to me then the output requires my exerting that control. In other words, the participant must respond to the control options that are provided in the medium. Responsiveness is a key element of the Rafaeli and Sudweeks (1997) denition of interactivity. They wrote that interactive communication requires all messages in a sequence to relate to each other. This implies that interactivity is a series of related active and reactive communications. Respondents were in general agreement that computer-mediated communication is more active than other types of mediated communication. Joseph spoke of individuals communicating through computers: They act by typing words or selecting things or moving things, or talking or a variety of things. Reaction goes beyond simple selection and choice of options. Rather than just making choices, individuals can customize the content to suit their needs. Will said that the capability of customizing messages for individuals or groups was central to the notion of interactivity. Using the example of search engines, he pointed out that the individual should be able to drive the search engine and the search engine should respond. If the individual has conducted a particular kind of search in the past, the next time he returns to the search engine it will remember his searching patterns. As it learns my preferences, it can then make suggestions to me, he said. In sum, responsive communication may require more effort and/or activity than passive message reception. But respondents were fairly unanimous in their belief that responsive communication has so many benets that the participant may think that less effort has been exerted. Perceived goals. The actual goal of communication may be determined by the creator of the message. However, discussions of goals among our respondents seem to focus on perceptions about goals rather than the goals themselves. Because this article focuses on how individuals perceive interactivity, it is the individuals perceptions rather than the actual goals that are the focus. Interview respondents talked about ways in which organizations are using computer-mediated communication to advance their causes. The same medium can be used to communicate a sales message and to transact the
171

New Media & Society 2(2)

sale. But not all computer-mediated communication is designed to persuade. Rather, respondents seemed to view persuading and informing as two points on a continuum of communication goals. In some cases, communication goals are more oriented toward exchanging information. For example, Jake shared the experience of online collaboration that enabled a group of programmers to coordinate their work for a specic client. Public relations practitioners have long recognized the fact that communication may be oriented to persuading and/or informing. In fact, goals are one of the key variables Grunig and Grunig (1989) use in dening four models of public relations. But we believe study of interactivity needs to move beyond how organizations dene their communication goals to how communication participants perceive the goals of those with whom they communicate. In conclusion, individual perceptions of communication goals are an important aspect of interactivity. Individuals are willing to accept messages based on both persuasive and informational goals in the interactive communication environment. But the ways in which they respond to those messages may be more consistent with their own perceptions than with the actual goals of the content creator. DEFINITION Joseph told us that any attempts to dene interactivity are futile at this time. Nobody knows, because the eld of interactive communication is in its infancy, what the possibilities are. You have to base your denition of interactivity on whats out there. And it changes every day. So your denition is going to have to be a bit loose in the beginning. Bob conrmed the evolutionary nature of computer-mediated interactivity:
We dont know what its going to be. It has its own momentum, in part set by the communication industry, and the software industry, and the users settings. We dont know where its going, except that its going real fast. And a slight shift in its heading will head it off in another hemisphere, if were not careful.

Several respondents told us interactivity is not a simple concept. Joseph said: Its not a single dimension. You are not trying to argue this is interactivity, this is not. Youre just trying to say that it includes these things. Pam said: Interactivity is going to be a continuum, I think. Despite the evolving nature of computer-based interactivity, we believe that a denition does emerge from the concepts found in the literature and in our in-depth interviews. However, it is clear that interactivity is a multidimensional construct and each of those dimensions seems to be represented by a continuum. Concepts related to the impact of interactive communication (revolution, consequences, and uncertainty) help to provide a framework for
172

Downes & McMillan: Dening interactivity

understanding why interactivity is such an important concept in the evolving eld of computer-mediated communication. However, the three message-based dimensions and the three participant-based dimensions seem to offer the most promise for helping us understand and dene the actual concept of interactivity. With respect to the message-based dimensions, our research suggests that interactivity increases as: Two-way communication enables all participants to actively communicate. Timing of communication is exible to meet the time demands of participants. The communication environment creates a sense of place. With respect to the participant-based dimensions, our research suggests that interactivity increases as: Participants perceive that they have greater control of the communication environment. Participants nd the communication to be responsive. Individuals perceive that the goal of communication is more oriented to exchanging information than to attempting to persuade. DISCUSSION Most communication in computer-mediated environments may be interactive. However, we believe varying levels of interactivity exist. Each of the message dimensions (direction, time, and place) and the participant dimensions (control, responsiveness, and perceived goals) seem to be continua. But even when these dimensions have values on the low end of the continuum, individuals may still perceive that they are participating in interactive communication. Something about the computer-mediated communication environment suggests that interactivity occurs even when the direction of communication is more one way than two way, when the participant has relatively little control, and so on. For example, many organizations believe that they have created an interactive communication tool when they create a corporate web site (McMillan, 1999). Many of these corporate web sites are little more than an online brochure. But sites such as these do have some characteristics (albeit at low levels) of interactivity. Site visitors believe that by pushing buttons and navigating through the site they are exerting some control and experiencing some machine-based responsiveness. While turning the page of a paper-based brochure is also exerting control and receiving a mailed
173

New Media & Society 2(2)

answer to a request for more information is also responsive, somehow those activities that take place in the real world are not seen as being as interactive as their cyberspace counterparts. Why is this? Is this a shortterm anomaly that reects societys tendency to be enamored of new technologies? Or is there some inherent characteristic of computer-mediated communication that makes it truly more interactive? Future research needs to explore these questions. At the other extreme, cyberspace does seem to offer opportunities for new forms of interactivity. Virtual communities described and promoted by writers such as Rheingold (1993) seem to represent such an opportunity. Such virtual communities often include user-created content, virtual meeting rooms, both synchronous and asynchronous two-way communication options, and an information-rich environment. In environments such as this, each of the dimensions identied above may have values on the high end of the continuum: communication is two way, timing of communication is exible, the communication environment creates a sense of place, participants have control over their communication experience, communication is responsive, and the purpose of the site seems to focus on information exchange. But is this kind of environment new, or merely an extension of both interpersonal and mediated communication? Does moving a conversation into a virtual community change that conversation in any way from what it would have been in a geographical community? Or, does the technology of computer-mediated communication simply enrich the long-distance conversations that telephone technology has facilitated for more than 100 years? Future research should explore these questions. Thus far, we have suggested two extremes: either low values or high values on each of the interactivity continua. But what if some values are high while others are low? For example, is there a form of cyberinteractivity in which the message dimensions of interactivity (direction, time, and place) are low while the participant dimensions (control, responsiveness, and goal) are high? Content-rich, searchable databases and online publications might t this model as individuals exert a high level of control over the information they receive but have relatively little two-way communication, control over timing of messages, or sense of place as they navigate through such databases and publications. But do individuals who access this type of information consider it to be interactive? What if message dimensions are high while participant dimensions are low? E-commerce might t this prole as online retailers create virtual marketplaces that lead the shopper through an experience that is sensitive to the shoppers time needs and allows for two-way communication that results in a purchase. But we must ask whether this virtual marketplace represents a new form of interactivity or whether it simply moves the shopping
174

Downes & McMillan: Dening interactivity

experience to a new space and time in much the same way that catalogs have long served as an alternative to the physical shopping experience. Future research needs to further explore the relative importance of both message dimensions and participant dimensions of interactivity. Much early attention seems to focus on the message dimensions. When asked to dene interactivity, our respondents often rst talked about characteristics of the interactive medium. But these message dimensions seem to be very closely linked to participant dimensions. In particular, the individuals control over the message seems to be a key determinant of interactivity. In addition to a broad exploration of the relative importance of messagebased and participant-based concepts, future research also needs to explore each of the six dimensions of interactivity identied in this research. How important is each to the individuals perceptions of interactivity in computer-mediated environments? Scales should be developed to measure each dimension and the relationship of each to the global concept of interactivity. Despite the uncertainty about interactivity and its potentially revolutionary characteristics, interactive communication may indeed have far-reaching consequences as suggested by our interviewees. It is important for those who study, develop, and participate in computer-mediated communication to understand the concepts and potential consequences of each of the six dimensions of interactivity identied in this article. APPENDIX 1 Interview instrument Im interested in your thoughts on issues related to the concept of interactivity and computers. The information Im gathering, over the next few minutes, from talking to experts (such as you) will help me and I hope will help others to better understand the impact of new technologies. Im tape recording our conversation so that, in the weeks ahead, I can refer to the tape, and, in turn, take some notes off it. Just answer the questions with whatever comes to mind. Feel free, too, to pass on any questions. The information gathered from this interview and others will be used in order to write an academic article; its not intended for any purpose other than that. At the conclusion of interviews such as these I always destroy the tapes and, therefore, can assure you your answers will remain condential. How does that sound? Any questions? 1. What does the concept of interactivity mean to you? 2. How much effort, relative to that expended for other types of communication, do you think interactive communication requires?
175

New Media & Society 2(2)

3.

4.

5.

6.

7. 8. 9. 10.

One of my colleagues recently suggested to me that fully interactive media create a situation in which the roles of the sender and receiver are interchangeable. What do you think of that idea? How important do you think it is for interactive communication to occur in real time? In other words, is it necessary for responses in order to be deemed interactive responses to be immediate? Someone recently said, and Ill paraphrase, that interactivity is a quality of individuals who are making use of a medium, rather than being a quality of a medium itself . How would you respond to this? Some kinds of communication give the communicator more control than others. For example, a public relations practitioner does not have complete control over how a reporter will cover a press conference, whereas an advertising professional has control over what an advertisement will look like. Tell me what you think about the issue of control as it applies to interactive communication? How important is it to track the activity of people who are participating in interactive media? What do you see as the advantages and disadvantages of different forms of interactive communication? A communications professional recently said to me that she sees interactivity as threatening. How would you respond to her? Finally, thank you for your willingness to share your ideas on these topics. Can you tell me if there are any major, related issues we didnt address above. What have we missed?

APPENDIX 2 Prole of respondents Joseph gained substantial experience as both an educator and executive for one of the nations leading computer rms. He serves on the full-time faculty at one of the countrys top communication programs. Pam recently published a book on the uses of new technologies and has an extensive background in media research, theory, and computer-assisted instruction and design. Dennis has extensive professional and academic experience, primarily in the advertising industry. He teaches a variety of courses on interactive communication while serving on the full-time faculty at a prominent southern university. Bob directs a technology task force for one of the leading communication programs in the US. He concentrates on the integration of new technology, such as interactive multimedia, into the universitys programs.
176

Downes & McMillan: Dening interactivity

Jake has served as an instructor for the past 15 years at a prominent research and technology university. He has worked extensively in the ongoing development of his institutions computer courses and played a signicant role in shaping the related curriculum. Kent manages content development for a large student-oriented web site. He also manages the university students who are responsible for creation of the content. Will has post-graduate training in computer science with an emphasis in articial intelligence. He currently teaches university students the technical aspects of building sophisticated web sites. Rita works professionally as a multimedia-media designer. She has experience as a journalist, lm maker, and artist. She teaches university students the basics of using computers as a tool for communication. Amy works for a company that specializes in facilitating collaborative workplace strategies. She has post-graduate education in philosophy and is also active in creating several virtual communities oriented to support of professional women. Jeff is a recently retired journalist whose computer experience began with mainframes in the 1960s, progressed to minicomputers in the 1970s, and evolved to microcomputers in the 1980s. He currently uses email extensively in his teaching of university writing courses.
References
Ang, S. and L.L. Cummings (1994) Panel Analysis of Feedback-Seeking Patterns in Face-to-Face, Computer Mediated, and Computer-Generated Communication Environments, Perceptual and Motor Skills 79: 6773. Blattberg, R.C. and J. Deighton (1991) Interactive Marketing: Exploiting the Age of Addressability, Sloan Management Review (Fall): 514. Braman, S. (1989) Dening Information: An Approach for Policymakers, Telecommunications Policy 13: 23342. Chen, M. (1984) Computers in the Lives of Our Children: Looking Back on a Generation of Television Research, in R.E. Rice (ed.) The New Media: Communication, Research, and Technology, pp. 26986. Beverly Hills, CA: Sage. DeFleur, M.L., P. Kearney, and T.G. Plax (1997) Fundamentals of Human Communication (2nd edn). Mountain View, CA: Mayeld Publishing Co. Denzin, N.K. and Y.S. Lincoln (1994) Handbook of Qualitative Research. Thousand Oaks, CA: Sage Publications. Dreze, X. and F. Zufryden (1997) Testing Web Site Design and Promotion Content, Journal of Advertising Research (March/April): 7791. Finn, T.A. (1998) A Conceptual Framework for Organizing Communication and Information Systems, paper presented at the International Communication Association Conference, Jerusalem, 204 July.
177

New Media & Society 2(2)

Firestone, W.A. (1987) Meaning in Method: The Rhetoric of Quantitative and Qualitative Research, Educational Researcher (October): 1621. Grunig, J.E. and L.A. Grunig (1989) Toward a Theory of Public Relations Behavior of Organizations: Review of a Program of Research, in J.E. Grunig and L.A. Grunig (eds) Public Relations Research Annual, Volume 1, pp. 2763. Hillsdale, NJ: Lawrence Erlbaum Associates. Ha, L. (1998) Interactivity in Business Web Sites: A Content Analysis, paper presented at the American Academy of Advertising Conference, Lexington, KY, 2730 March. Heeter, C. (1989) Implications of New Interactive Technologies for Conceptualizing Communication, in J.L. Salvaggio and J. Bryant (eds) Media Use in the Information Age: Emerging Patterns of Adoption and Computer Use, pp. 2215. Hillsdale, NJ: Lawrence Erlbaum Associates. Jensen, J.F. (1998) Interactivity: Tracing a New Concept in Media and Communication Studies, Nordicom Review 19(1): 185204. Kayany, J.M., C.E. Worting, and E.J. Forrest (1996) Relational Control and Interactive Media Choice in Technology-Mediated Communication Situations, Human Communication Research 22: 399421. Lee, T.W. (1999) Using Qualitative Methods in Organizational Research. Thousand Oaks, CA: Sage Publications. Lynne, M.L. (1994) Finding a Happy Medium: Explaining the Negative Effects of Electronic Communication on Social Life at Work, ACM Transactions on Information Systems 14: 11949. Markus, M.L. (1994) Finding a Happy Medium: Explaining the Negative Effects of Electronic Communication on Social Life at Work, ACM Transactions on Information Systems 14: 11949. Marshall, C. and G.B. Rossman (1989) Designing Qualitative Research. Newbury Park, CA: Sage Publications. Mason, J. (1996) Qualitative Researching. London: Sage Publications. McMillan, S.J. (1998) Who Pays for Content? Funding in Interactive Media, Journal of Computer-Mediated Communication 4(1), URL: (consulted Feb. 1999) http:/ /www.ascusc.org/jcmc/vol4/issue1/mcmillan.html McMillan, S.J. (1999) Advertising Age and Interactivity: Tracing Media Evolution Through the Advertising Trade Press, paper presented at the American Academy of Advertising Conference, Albuquerque, NM, 269 March. Morris, M. and C. Ogan (1997) The Internet as Mass Medium, Journal of ComputerMediated Communication 1(4), URL: (consulted Feb. 1999) http://www.cwis.usc.edu/ dept/annenberg/vol1/issue4/vol1no4.html OKeefe, S. (1995) Publicity on the Internet: Creating Successful Publicity Campaigns. New York: Alfred A. Knopf. Patton, M.Q. (1987) How to Use Qualitative Methods in Evaluation. Newbury Park, CA: Sage Publications. Pavlik, J.V. (1996) New Media Technology: Cultural and Commercial Perspectives. Boston: Allyn and Bacon. Rafaeli, S. (1988) Interactivity: From New Media to Communication, in R.P. Hawkins, J.M. Wiemann and S. Pingree (eds) Advancing Communication Science: Merging Mass and Interpersonal Process, pp. 11034. Newbury Park, CA: Sage. Rafaeli, S. (1990) Interacting with Media: Para-Social Interaction and Real Interaction, in B.D. Ruben and L.A. Liverouw (eds) Mediation, Information and Communication: Information and Behavior, Volume 3, pp. 12581. New Brunswick, NJ: Transaction Publishers.
178

Downes & McMillan: Dening interactivity

Rafaeli, S. and F. Sudweeks (1997) Networked Interactivity, Journal of ComputerMediated Communication 2(4), URL: (consulted Feb. 1999) http://www.usc.edu/dept/ annenberg/vol2/issue4/rafaeli.sudweeks.html Rheingold, H. (1993) The Virtual Community: Homesteading on the Electronic Frontier. New York: Addison-Wesley. Rice, R.E. (1984) New Media Technology: Growth and Integration, in R.E. Rice (ed.) The New Media: Communication, Research, and Technology, pp. 3354. Beverly Hills, CA: Sage. Rice, R.E. and F. Williams (1984) Theories Old and New: The Study of New Media, in R.E. Rice (ed.) The New Media: Communication, Research, and Technology, pp. 5580. Beverly Hills, CA: Sage. Rogers, E.M. (1995) Diffusion of Innovations (4th edn). New York: Free Press. Rosoneau, P.M. (1992) Post Modernism in the Social Sciences. Princeton, NJ: Princeton University Press. Schaffer, L.C. and M.J. Hannan (1986) The Effects of Progressive Interactivity on Learning from Interactive Video, Educational Communication and Technology 34: 8996. Singleton, R.A. (1993) Approaches to Social Research. New York: Oxford University Press. Snyder, I. (1996) Hypertext: The Electronic Labyrinth. Melbourne: Melbourne University Press. Sproull, L. and S. Kiesler (1991) Computers, Networks, and Work, Scientic American (September): 11623. Steur, J. (1992) Dening Virtual Reality: Dimensions Determining Telepresence, Journal of Communication 42: 7393. Stolz, C. (1995) Silicon Snake Oil: Second Thoughts on the Information Highway. New York: Doubleday. Strauss, A.L. (1996) Qualitative Analysis for Social Scientists. New York: Cambridge University Press. Thomas, P.J. (1995) Introduction: The Social and Interactional Dimensions of HumanComputer Interfaces, in P.J. Thomas (ed.) The Social and Interactional Dimensions of Human-Computer Interfaces, pp. 110. Cambridge: Cambridge University Press. Walther, J.B. (1994) Anticipated Ongoing Interaction Versus Channel Effects on Relational Communication in Computer-Mediated Interaction, Human Communication Research 20: 473501.
EDWARD J. DOWNES is an assistant professor at Boston University. His expertise is in communication management, public relations, public affairs, program administration, and qualitative research. His most recent work was published in the Journal of Public Relations Research. Address: Boston University, College of Communication, 640 Commonwealth Ave, Boston, MA 02215. [email: edownes@bu.edu] SALLY J. MCMILLAN is an assistant professor at the University of Tennessee. Her research focuses on the individual, organizational and social impacts of interactive communication technologies and has been published in journals such as the Journal of Computer-Mediated Communication and Health Communication as well as proceedings of the American Academy of Advertising conference and the Advertising and Consumer Psychology conference. Address: The University of Tennessee, College of Communication, 476 Communication Building, Knoxville, Tennessee, TN 37996. [email: sjmcmill@utk.edu]

179