When communication researchers make claims about the relationship of media to individuals or society, they use the term media to mean a variety of things. For example, a researcher might try to prove the claim "television causes X," where X might be anything from aggressive behavior to bad vision or a crisis in a sense of self. In this case, is the researcher talking about the television signals in the air, people who work at television stations, people who produce or act in television programs, the television programs themselves, television receiving sets, or some or all of these things?
Of course, the research context, setting, and constraints usually define what is meant by the term television as a medium, but how can researchers talk about new media that involve communication on computer networks? The Internet, a cooperatively run, globally distributed collection of computer networks, provides a communication forum in which an estimated 20-40 million people in 90 countries (Society, 1995) participate. The Internet provides an array of tools for people to use for information retrieval and communication in individual, group, and mass contexts, but can current notions of media be used to define communication on the Internet?
Researchers in past decades have taken many approaches to analyzing human communication on computer and networked communication systems. Using a variety of frameworks for defining units of analysis, these researchers have examined an array of communication settings. For example, some research has explored the relationships between the characteristics of media systems and the characteristics of individuals using them (Hiltz & Turoff, 1978; Johansen, Valle, & Spangler, 1979). Other researchers have examined the human component of computer-mediated communication processes in detail, examining social-psychological factors (Kiesler, Siegel, & McGuire, 1984; Kling & Gerson, 1977; Lea & Spears, 1991a, 1991b; McGuire, 1983; Spears & Lea, 1992; Spears, Lea, & Lee, 1990), as well as social context factors (Feenberg, 1989, 1992; Fulk, Schmitz, & Steinfield, 1990; Fulk, Steinfield, Schmitz, & Power, 1987; Georgoudi & Rosnow, 1985; Lea, 1992; Martin, O'Shea, Fung, & Spears, 1992; Schmitz & Fulk, 1991), and social cues (DeSanctis & Gallupe, 1987; Kiesler, 1986; McGuire, Kiesler, & Siegel, 1987; Rutter, 1987; Siegel, Dubrovsky, Kiesler, & McGuire, 1986; Sproull & Kiesler, 1986). This body of work presents a mixture of results that are very dependent on the context of the research setting. Integration of results, particularly at the theoretical level, is difficult.
Research focusing on media has likewise lead to insights, but little theoretical integration or comparison of results from study to study. Researchers have examined the diffusion and adoption of interactive media and found factors contributing to media technology adoption as well as patterns of how technology use develops in a community (Markus, 1987, 1990; Miles, 1992; Rogers, 1983, 1986; Valente, 1991). Other researchers and writers have looked at media evolution to identify societal and individual changes as a result of the use of communication technology (Harnad, 1991; Havelock, 1986; Innis, 1972; Levinson, 1990; McLuhan, 1964, 1965; McLuhan & Fiore, 1967; McLuhan & Powers, 1989; Vallee, 1982). More recently, media researchers have examined the notion of "media richness" to examine media selection in individual and organizational communication (Daft & Lengel, 1984, 1986; Daft, Lengel, & Trevino, 1987; Lengel & Daft, 1988; Rice & Shook, 1990; Trevino, Daft, & Lengel, 1990; Trevino, Lengel, Bodensteiner, Gerloff, & Muir, 1990; Trevino, Lengel, & Daft, 1987). This focus on media uses a variety of frameworks for defining units of analysis, or fails to define any units of analysis. As a result, it is difficult to piece together an integrative model to explain and predict media use, adoption, and evolution patterns, or even classify study results or theoretical statements.
Another approach to researching on-line communication is a focus on language and rhetoric. Researchers in these areas have likewise discovered many insights into the structure and content of computer-mediated communication and how literacy and orality are affected by communication technology (Baron, 1984; Black, Levin, Mehan & Quinn, 1983; Ferrara, Brunner & Whittemore, 1991; Finnegan, 1988; Gurak, 1994; Lakoff, 1982; Murray, 1991; Ochs, 1989; Ong, 1977, 1982; Shank, 1993; Spitzer, 1986). These studies have examined a variety of on-line content and used many schemes for defining or discussing units of analysis.
Murray (1991), for example, looked at electronic documents and mail used for interpersonal and small group communication on a proprietary computer network, and identified cognitive and contextual strategies for writing documents on personal computers and using electronic mail in a study of an IBM project manager and his colleagues. Shank (1993) examined electronic mailing lists involving a large number of people, many of whom are unknown to each other, on the Internet or other networks. Shank argued that communication in these on-line discussion lists is neither oral nor written, but semiotic. Can the Murray and Shank results be compared? Each researcher looked at different uses of computer-mediated communication, and thus used different units of analysis for the communication. Without careful attention to the definition of the units of analysis researchers used, it is difficult to integrate the results of these studies.
Over the decades, research in computer-mediated communication has also explored myriad on-line experiences, focusing on human and social characteristics, media, and language and rhetorical content. However, the research has not led to much successful theoretical integration or cross-study comparisons. Indeed, some research has directly contradicted previous work (Lea, 1992). Researchers have articulated the units of analysis in studies on a case-by case basis. This diversity in terminology has been necessary because of the variety of network settings and communication networks under study. These settings have included stand-alone computer-to-computer communication, electronic mail discussion lists, commercial and proprietary on-line services, commercial communication and group-ware packages, and many other communication systems (Rapaport, 1991). Similarly, the many theoretical approaches have employed diverse terminology and definitions for units of analysis.
Added to this lack of theoretical integration have been the changes and advances in Internet communication technology over the period 1969-1995. While electronic mail and Usenet news-group discussions were the early forms of communication used on the Internet, today the Internet offers a far wider range of tools for information retrieval, communication, and interaction than just text-based discussion and information dissemination (December, 1995a). The use of the Internet also has rapidly increased, with some Internet applications, such as the World Wide Web (the system for linking documents associatively using hypertext), experiencing very rapid increases in use and range of expression (December & Randall, 1995).
Those who study Internet communication may utilize this research site to identify commonalties in units of analysis. These common units can assist in research result comparisons and theoretical convergence. Without a common framework for units of analysis, definitions of what is being studied on the Internet can be clouded by a poor definition of the research setting, making cross-study (or even intrastudy) comparisons difficult.
This article proposes
an approach to defining units of analysis for Internet communication
research. To define these units of analysis, I first define the
term Internet computer-mediated communication. This definition
identifies the characteristics of Internet communication. I then
use this definition as the basis for developing a set of definitions:
media space, media class, media object, and media
instance as units of analysis for Internet communication studies.
Finally, I illustrate these units of analysis with some examples.
Whereas electronic mail has been a frequent subject area for previous communication research, the global Internet today offers a far more diverse set of tools and contexts for communication than it has in the past. Communication on the Internet also exists within developed social and cultural traditions. Therefore, what we mean when we talk about specific forms of Internet communication must be carefully defined, so that consistency in research approaches can be made and areas of inquiry for possible research can be identified.
Approaches
to Defining Communication
I define Internet communication
here by identifying characteristics of technology to define units
of study. I want to note that these units of study are just one
part of a total research project. The other concerns for researchers
include the characteristics of the people involved in the communication,
the social context of the communication, and a host of other possible
factors in communication research. Therefore, my definition here
captures the qualities of the technology so that researchers then
can use this framework as a basis for further identifying the
other units and factors important to a particular research study.
This approach makes careful distinctions among terms and units of analysis. This careful approach may seem like overkill, because such an exhaustive definition of a medium has not seemed necessary for studying other forms of media. Television and radio are familiar enough so that a researcher does not usually need to carefully define the object of study. A researcher looking at a particular genre of radio communication, such as AM talk radio, rarely would have to explain further what is meant by this term. The diversity of Internet communication systems and applications, however, motivates a more careful definition of terms and units of analysis. As I will describe below, the Internet cannot be considered to encompass a single medium, but consists of a range of media. Without careful definition of analysis units, researchers might easily obscure whatever part of the Internet research setting they are examining.
To generate a definition
of Internet communication, I first analyze the constituent parts
of the term Internet-based, computer-mediated communication. I
examine each term in successive subsections of this article, each
with a heading corresponding to the term. Following this definition,
I illustrate by describing some categories of activity the term
encompasses.
Internet-based
To say that communication
is Internet-based means that, at the data level, it conforms to
a particular set of data communications protocols. A protocol
is a set of rules for exchanging information. Computer networks
use protocols to enable computers connected to a network to send
and receive messages. The set of protocols called the TCP/IP protocol
suite defines the rules for data exchange on the Internet. This
set of protocols, originally developed for a United States Department
of Defense research project, integrates a set of services (including
electronic mail, file transfer, and remote log-in) that can occur
among many computers on local or wide-area networks.
The resulting networks connected with the TCP/IP protocol suite are highly robust. If one section of the network (or a computer host in the network) becomes inoperable, data can be rerouted around the damage in the network. Figure 1 illustrates how these protocols send data over the networks. First, the TCP (Transmission Control Protocol) breaks the data into packets of information. Next, these packets are sent over the network, possibly over different routes, according to IP (Internet Protocol). Finally, these packets are reassembled (or re-sent, in the case of data corruption or loss) in their proper order upon arrival at the destination.
What this protocol scheme does is create a system for reducing all communication to data exchange, and this commonality of protocol is the essence of Internet communication at the data level. Through cooperation and connections, TCP/IP networks can be connected in larger and larger communication systems. Individual organizations can run their own TCP/IP network (an internet) and connect it with other local, regional, national, and global internet networks. The resulting patchwork of networks, called the Internet (with a capital I), shares the TCP/IP protocol suite. However, the Internet is not a single network, but a cooperatively organized, globally distributed system for exchanging information. The data that traverses this network of networks is Internet communication.
However, the Internet is not the only global network. Other global networks employ different protocols, but can exchange data with the Internet through exchange points called gateways. Non-Internet communication flowing into a gateway point is translated to Internet communications protocols and sent on its way, indistinguishable from the packets that TCP creates when sending a message directly on the Internet. Likewise, communication can flow off the Internet at the gateway points in the same manner: The Internet packets are translated to the non-Internet protocols necessary for communication on another network.
Electronic mail is a popular form of communication exchanged across gateways. Through electronic mail gateways, users on the Internet can exchange electronic mail with other users on (non-Internet) networks, such as BITNET (Because Its Time Network), UUCP (Unix-Unix Copy Protocol), and FidoNet (a network based on personal computer communication over telephone lines). Users of the Internet can also exchange electronic mail with many commercial services such as Prodigy, Delphi, CompuServe, America on-line, and many others (Chew & Yanoff, 1995). The result is that electronic mail disseminates freely throughout the Internet, as well as many other networks. The resulting collection of worldwide networks that exchange electronic mail is called the Matrix (Quarterman, 1990).
Although the free flow of electronic mail blurs the distinction between Internet communication and non-Internet communication in the Matrix, the distinction between the Internet and the Matrix for many other communication applications is crucial. For example, communication using the Internet Gopher protocol cannot easily be shared outside the Internet. Similarly, Telnet, FTP (File Transfer Protocol), and World Wide Web communication are restricted, in most cases to users of the Internet. Commercial on-line services, recognizing the value of access to the Internet for their customers, have been creating more kinds of gateways to the Internet. Commercial on-line services now have many gateways allowing their users to access Telnet, FTP, and World Wide Web applications on the Internet.
The resulting mix of global networks makes the Internet protocol suite the lingua franca of cyberspace, creating a common ground to which many other on-line networks connect through gateways. By having many gateways to the Internet, commercial services can offer users access to the content of the Internet while offering them proprietary, value-added services and content. A researcher studying communication on an on-line service (for example, Prodigy) must be careful to realize that they are not studying Internet communication. While the users of Prodigy may enjoy the free flow of electronic mail from the Internet to Prodigy and back, there are many proprietary services that Prodigy offers that are not available on the Internet.
The emphasis in this discussion on the Internet's boundaries and gateways underscores the first work of a researcher studying Internet communication: to identify what is and what is not Internet communication, and to investigate possible relationships between Internet communication (through gateways) and non-Internet communication. Based on this discussion, I can summarize a definition for Internet-based communication: Internet-based communication takes place on the global collection of networks that use the TCP/IP protocol suite for data exchange.
There are TCP/IP networks
(internets, with a small i) that are not connected to the
larger global Internet. These internet networks, therefore, although
technically using the TCP/IP protocol suite, are not part of the
global Internet based on this definition. People using these local
internets cannot access the myriad resources or take part in the
cultural and social traditions and artifacts that exist on the
global Internet.
Computer
The term computer
in the context of defining Internet computer-mediated communication
means much more than just a device for calculation. In fact, the
salient function of a computer as used for communication is not
to provide computational capability, but to provide a platform
for the operating system and software applications to support
network data transmission and user applications.
On the Internet, the relationships among computers commonly follow the client-server model. Like the TCP/IP protocols, the client-server model is a unifying characteristic of Internet communication. A server is a computer and its associated hardware and software applications that act as a repository for information files or software programs. The server sends this information by request across the network to users of client software.
Figure 2 summarizes the client-server relationships. First, a request for information flows from a client to a server. Based on this request, the server sends information back to the client.
Client-server communication also follows a set of protocols. These protocols define the particular application the client and server are using. For example, the Internet Gopher protocol defines an application for structuring information into a system of menus, submenus, and entries. A user of a Gopher client makes a request from a Gopher server for a list of menu items. The Gopher server sends back this list, and the Gopher client displays the list for the user.
The distributed form of the client-server request and serve scheme allows for many efficiencies. Since the client software interacts with the server according to a standard data exchange protocol, the client software can be customized for the user's particular computer host. This means that the server doesn't have to "worry" about the hardware or software particularities of the computer on which client software resides. The client software doesn't have to "worry" about which server of a particular type it requests information from, as all servers of a particular protocol behave the same way.
For example, a Web client that can access any Web server can be developed for Mcintosh computers. This same Web server might be accessed by a Web client written for a Unix workstation running the X-Window System. This sets up a clear demarcation of duties between the client and the server. This makes it easier to develop information because separate versions of information distributed from a server need not be developed for any particular hardware platform. All the customizations necessary for the user's computer are written into client software for that platform.
An analogy to the client-server model is the television broadcast system: A customer can buy any kind of television set (client) to view broadcasts from any over-the-air broadcast tower (server). Whether the user has a wristband TV or a projection screen TV, the set receives information from the broadcast station in a standard format and displays it in a manner appropriate to the user's TV set. Separate TV programming need not be created for each kind of set, regardless of such characteristics as screen size. Further, newly created television stations will be able to send signals to all television sets currently in use.
The client-server model
is the key characteristic of the Internet's scheme for communication
among applications.
Mediated
Mediation is the process
of intervening or coming between. For Internet communication,
mediation also involves literally putting a message into media,
or encoding a message into electronic, magnetic, or optical patterns
for storage and transmittal. A message on the Internet is encoded,
stored, and transmitted according to the rules of the client-server
application and the TCP/IP protocol suite.
The transmittal of this mediated message on the Internet can also have a variety of mediation characteristics, including time, distribution, and media type.
The time delay between sending and receiving messages can vary. This time delay can be nearly instantaneous in applications where users take part in (nearly) "real time" text interchange. Internet applications where users engage in this communication include mass and group communication systems, such Internet Relay Chat (IRC), and Multiple User Dialogue/Dimension/Dungeon (MUDs) or variants (known as MU* systems). In other network information retrieval systems, communication can be persistent, that is, exist on a server for on-demand retrieval, such as file of hypertext on a Web server, information on a Gopher server, or files available from an FTP server.
A message on the Internet can be distributed from a sender to receivers according to a variety of schemes. These variations include
Internet applications display information in a variety of media types, including text, sound, graphics, images, video, or binary (executable) files. The Multipurpose Internet Mail Extensions (MIME) specifications define these multimedia types commonly used in Internet communication. Text that uses a system of associative linking is called hypertext. Hypertext that employs multimedia is called hypermedia.
The range of characteristics outlined above highlight the range of possibilities for communication on the Internet involving variations in time, distribution scheme, and media type. A researcher should be able to identify each of these characteristics for a particular form of Internet communication under study.
Note that other media may exhibit these same characteristics. For example, radio broadcasts are available in real time to anyone with an appropriate client (radio receiver) within range of the broadcast station. A person calling into a radio program with their voice broadcast as part of the program is using a point to server broadcast form of distribution. The characteristics cited here for time, distribution, and media, typify the mediation process in Internet communication content.
Communication
Although data communication
serves as the basis for transmitting messages on computer networks,
a communication researcher, unless involved in the detailed technical
examination of data transmission, is concerned with the human
communication issues involved when people communicate on the Internet.
Indeed, Internet-based communication is essentially human communication
via the Internet computer network, so any definition of Internet
communication necessary involves defining human communication
itself. Human communication can be characterized as a process
in which people exchange symbols (Littlejohn, 1989). The process
of symbol exchange occurs in the context of Internet communication
with mediation characteristics as described above, following the
client-server model for information exchange and the TCP/IP protocol
suite for data exchange.
The content of Internet communication, however, is more complex to describe, but it can often be captured for study. Internet content can be encoded and decoded using a variety of media types (text, graphics, sound, video, executable file). Because of its mediated form, Internet communication often leaves a visible trail of artifacts that can be collected for study. For example, Usenet discussion groups produce a set of articles that can be examined and analyzed. Real-time discourse among a group of participants, such as in MU* systems, can be recorded in the form of a transcript. Other Internet communication artifacts include Web pages, files on Gopher sites, and transcripts of Listserv discussion lists.
The interpretation
of these artifacts depends on the goals of the study. Likewise,
the symbols important to the researcher will vary. For example,
Internet communication using real-time audio-video conferencing
can occur with an application called CU-See Me. This application
allows participants to transmit pictures and sound, thus displaying
many nonverbal and paralinguistic cues not present in much text-based
communication on the Internet. Thus, communication on the Internet
represents a range of possibilities for symbol creation, and some
of these symbols can be similar to those examined in unmediated
human communication.
Integration
The above discussion
outlines the boundaries of the definition of Internet-based, computer-mediated
communication. Using this definition, a researcher can identify
what is not Internet communication (e.g., electronic mail exchanged
among users of a commercial on-line service such as America Online)
as well as what is Internet communication (e.g., users communicating
via text exchange in Diversity University's MU*. Clearly, non-Internet
communication may include a range of activities, including commercial
voice mail, telephone conversations, Local Area Network (LAN)
communication, or Computer-Supported Cooperative Work on LANs,
that may exhibit many of the qualities of Internet communication.
However, users of the Internet experience a much different communication
context than non-Internet users. Internet users can access multiple
applications for communication (for example, using a Windows environment
on a PC or a workstation to use electronic mail, a Web client,
and a MU* client all at once). Users of the Internet have access
to a large set of possible communication partners and information
sources. Internet users also communicate within a particular cultural
context on the Internet, with its own shared cultural traditions
and symbols, such as typographic symbols representing emotions
called smilies.
Based on the above
discussion, we arrive at the following definition: Internet-based,
computer-mediated communication involves information exchange
that takes place on the global, cooperative collection of networks
using the TCP/IP protocol suite and the client-server model for
data communication. Messages may undergo a range of time and distribution
manipulations and encode a variety of media types. The resulting
information content exchanged can involve a wide range of symbols
people use for communication.
Examples of Internet
communication
People use Internet
communication for many purposes. I identify some of these major
purposes with the broad categories of communication, interaction,
and information. These categories are not mutually exclusive:
Someone can participate in Internet communication for a combination
of communication, information, and interaction at the same time.
Communication: People use the Internet for communication in a one-to-one, one-to-many, or many-to-many setting. This communication can be used for scholarly activity and research or for personal and group communication and discussion. Examples: Usenet, electronic mail, and Listserv.
Interaction: People can use the Internet for the purpose of play or learning, not just for information transfer or discussion. Interaction spaces are often used for social activity and for group interaction and education. Examples: MU*s and IRC.
Information: People use the Internet for dissemination and retrieval of information. This information deals with subject matter covering a wide range of human activities and knowledge. Examples are the World Wide Web, Gopher, and FTP.
Defining what Internet
communication is and showing how this definition encompasses a
broad range of human communication, interaction, and information
purposes are the first steps in characterizing the units of analysis
for Internet communication research. In the next section, I use
and expand on the terms from this definition of Internet-based,
computer-mediated communication to describe more specific units
of analysis.
The discussion to this point should help an Internet researcher identify a research study's parameters for data exchange, client-server communication, message mediation characteristics, and communication symbols. The next step is to more precisely define what specific area of communication on the Internet is being examined.
If one researcher states that he or she will study MU* interaction, this could involve observations on a variety of MU* systems, including, potentially, discourse in systems for real-time text interchange, including MUDs (Multiple User Dialogues), MOOs (an object-oriented MUD), or MUSEs (Multiple User Simulation Environment). Even further, the particular MU* studied is extremely important, as MU*s vary in their layout, inhabitants, and activities. Jay's House MOO is very different from Diversity University MOO, for example. Yet, activities among MU*s do share some commonalities that may allow for some cross-comparison of research results. The purpose of this section is to define the units of analysis that help researchers identify what types of communication they are studying on the Internet.
I would like to emphasize again that defining the units of analysis with this framework should not imply that only these units can be profitably used to define a research setting in isolation of other factors. Indeed, as I noted above, a communication research study will involve a wide range of other considerations, such as social setting and context, user characteristics, or the purpose of the communication. But by defining the units of analysis described here, a researcher can help identify what kind of communication system and setting is under study.
The Server-Client-Content Triad
I establish this framework
for units of analysis by developing definitions for a set of terms.
I use the terms server, client, and content
as an organizing triad for these definitions. Defined briefly
above, I expand on the definitions of these terms:
Server-A computer and associated software that provide access to information through the Internet in response to requests from client software based on a particular protocol for data exchange. An example is a World Wide Web Server using the NCSA (National Center for Supercomputing Applications) software.
Client-Software that operates on a user's computer for accessing information distributed from servers according to one (or more) protocol(s) for data exchange. An example is a Netscape Communications Corporation World Wide Web client. The Netscape client can access Web servers, and also FTP, Gopher, Telnet, and other protocol servers.
Content-Information that is exchanged, distributed, or available for retrieval or transmittal on networks. Examples are the content of the Usenet newsgroup alt.hypertext, or the text exchanged among users in the Communications Center (a particular room) in the Diversity University MOO.
Internet communication is not a single medium sharing common time, distribution, and sensory characteristics, but a collection of media that differ in these variables. I define a unit of analysis called a media space, which uses the client-server-content triad as the basis for its definition. This concept of media space is one way to describe how the Internet consists of a range of media.
Media space-A media space consists of the set of all servers of a particular type that may provide information in one or more protocols, the corresponding clients that are capable of accessing these servers, and the associated content available for access on these servers.
Note that this definition allows for a space to include servers that may provide information in a variety of protocols, but with the unifying idea that these servers all share the ability to provide information to clients with at least one common protocol. This common protocol usually identifies the type of the servers. This allows for the possibility that a space may include servers with the ability to deliver information in multiple protocols to clients capable of interpreting all of these protocols.
For example, we can consider one Internet media space to be Gopher space: the set of all information (content) provided by Gopher servers, accessible by people using Gopher clients. Note that Web clients can also access Gopher servers, so that Web clients are components of Gopher space.
Another example of an Internet media space is defined by Internet Relay Chat (IRC): IRC space consists of IRC clients accessing the text exchanged among participants from any one of many IRC servers. A user wishing to enter IRC space would need a client, such as Telnet, or a specialized IRC client, to access information on a server. The user potentially has many servers to access worldwide. The discussion on servers, generated and observed by users employing clients, constitutes IRC space. Note that IRC space is disjointed from Gopher space: A Gopher client cannot be used to observe IRC content, nor can IRC clients be used to observe IRC content.
Web space, or the set of all Hypertext Transfer Protocol (HTTP) servers, Web clients, and content on HTTP servers, is itself composed of several spaces. Using the Java programming language, developers can create specialized protocols that can be used to deliver information from Web servers to Java-enabled clients (December, 1995b). This delivery mechanism for these specialized Java-defined protocols is still the Web server. However, only Java-enabled Web clients can be used to observe Java-defined content. Therefore, the space defined by "Web hypertext, Web servers, and Web clients" is not the same as "Web hypertext plus Java-defined content, Web servers, and Java-enabled Web clients." These two spaces share common components (the non-Java content of the Web) and could then be considered to overlap.
This definition of a space corresponds to informal descriptions of other media. Colloquially, we might say that television is a medium, and mean that television is not just a collection of TV sets (clients), nor all of the broadcast TV stations (the servers). Instead, our concept of television as a medium encompasses all TV sets, all broadcast (and cable and satellite) systems, and all programming and production in combination with viewers' observation of this content through television sets (clients). Similarly, on the Internet, a space is not just one client type (e.g., Mosaic as an example of a Web client), nor the collection of all servers of one type, but the entire set of clients and servers along with the content on those servers potentially observable by those clients.
This definition of an Internet media space is useful to capture the idea that there are many different (sometimes overlapping, sometimes disjointed) spheres of activity on the Internet. A space, defined this way, is a seamless forum, in which users can observe any of the content from the servers in that space using their clients.
However, each Internet media space encompasses a vast amount of activity, probably too vast for a single research study. So just as communication researchers may want to focus on a subset of a television for study, such as television news programs or late-night talk shows, so, too, might the researcher in Internet communication need to focus on a subset of an Internet media space. The definitions below subdivide two analysis units to describe more precise units: I use the term media class to define a particular set of content, servers, and clients; the term media object defines a specific unit in a media class with which the user can observe and interact.
Media Class-A media class consists of content, servers, and clients that share a defined set of characteristics.
According to this definition, a media class can be equivalent to a media space. For example, all the Gopher content, servers, and clients comprise a media class, which is also the definition for Gopher space. A researcher, however, can define a subclass of a single space or several spaces in terms of a media class to focus on a particular set of Internet communication for study.
Here are some examples of media classes defined by shared characteristics of servers, clients, or content: (a) The hypertext (content) available from the Web server www.w3.org, observable through any Web client. This media class is a subset of Web space. (b) All the available information on the Internet accessible using the NCSA Mosaic Web client. This media class consists of several Internet media spaces, including Gopher space, Web (HTTP) space, FTP space, and others. (c) The set of all hypertext listed in the Clearinghouse for Subject-Oriented Internet Guides (http://www.lib.umich.edu/chhome.html). This media class is a subset of Web space. (d) All the available information on the Internet accessible using the HotJava Web client. This media class consists of several Internet spaces, including Gopher space and Web space, plus the specialized protocols and content created with the Java programming language (Java space).
Note how media class differs from media space. An Internet space includes all servers, corresponding clients, and content available on these servers that can be delivered to users. In contrast, a media class could include subsets of one or more media spaces, or even several media spaces. For example, as examples 2 and 4 above, a media class defined by all content accessible through a Web client includes not only Web space, but also the spaces defined by FTP, Gopher, Telnet, and Usenet.
This definition of a media class gives the researcher a way to define areas of study that may include subsets of several media spaces. However, the definitions of media space and media class are still at too high a level of abstraction to actually define something a user can interact with or observe. People cannot watch abstractions such as television (a medium) or even a media class defined by a genre of television (late night talk shows). Instead, television users observe specific programs, such as Late Night with Conan O'Brien, or The Late Show with David Letterman, or The Tonight Show starring Jay Leno, which are broadcast from a specific station and viewed with a particular television set. Similarly, analyzing Internet communication can benefit from having a term for a unit of analysis that is specific enough so that it defines something someone can observe. I use the term media object for this concept.
Media Object-A media object is a member of a media class for which the server, client, and content are completely and unambiguously specified. A media object is concrete, whereas a media class is simply a template for defining media objects. For example, a media object can be defined in this way: The World Wide Web (WWW) Frequently Asked Questions (FAQ) List on the SunSITE Web server sunsite.unc.edu accessed through the Netscape Navigator client for X, version 1.1.
Note that a particular media object might be a member of several different media classes. For example, let the media object, W, be defined by the above example ("the WWW FAQ as viewed through Netscape Navigator for X version 1.1 Web client"). Media object W is a member of the media class that includes all the information at sunsite.unc.edu. W is also part of the media class that includes all content accessible by the Netscape Navigator for X version 1.1 Web client. Still further, W, is a member of the media class whose content is defined by all the FAQ lists available on the Internet.
While the definition of a media object is specific enough to define something a user can observe, a medium may change over time. Therefore, I use the term media instance to capture this idea:
Media instance-A media instance is a media object at a particular time.
Under this definitional scheme, people perceive media instances, not media objects, classes, or spaces. However, colloquially, people often talk in terms of media objects and classes: "I watch TV" (space), "I watch TV news" (class), or "I watch the CBS news on my TV" (object). We, however, can actually experience only media instances, for example, "the CBS news on my TV last night for the first twenty minutes" (instance).
To specify a media instance, a researcher needs to specify a particular point in time for observing a media object. For example, the media object W, defined above, can be used to define the instance: W as observed on August 21, 1995, at 10:25 p.m.
However, a media instance is not yet tied to a particular user's experience and may be still too abstract for some studies. I define media experience to link a media instance to a particular user's perception:
Media experience-A media experience is a particular user's perception of a set of media instances.
Note that a user's media experience might include observation of many media objects, and these media objects might be from different media classes,which may consist of several media spaces. Here are some examples of media experience:
· Susan watched the CBS news broadcast on Wednesday, August 30, 1995, for the first 10 minutes. (Implied in this description is the specification of the server as Susan's local television broadcast station or cable service and her client as her particular model of TV set).
· Tom watched the CBS television program, Late Show with David Letterman, on August 30, 1995, while using his Netscape Web client for Windows '95 connected to the Web support site for that program, http://www.cbs.com/lateshow/.
· Chris participated
in the Internet Relay Chat channel #current-events using a Telnet
client from 10 p.m., August 30, 1995, to 2 a.m., August 31, 1995,
and used his FTP client to retrieve the file ftp://ftp.rpi.edu/pub/communications/internet-cmc.txt
at 2 a.m.
Contextualizing
Media Units of Analysis
The importance of the
above careful delineation of media space, class, object, instance,
and experience becomes more apparent when placed in the context
of a particular research study. As an example of how this definition
system can be applied, I will describe how these terms can help
in studying Web-based mass communication using the uses-and-gratifications
approach from mass communication theory.
The uses-and-gratifications approach considers consumers of media to be purposive in their choice of media and to actively seek media to fulfill their needs for a variety of uses (Infante, Rancer, & Womack, 1993, p. 408). Typologies of uses for mass media consumption, as well as other research and studies, have identified a broad range of gratifications people have in broadcast (one-way, one-to-many messaging) in several mediated communication contexts (McGuire, 1974). Other typologies of uses (Blumler, 1979; Palmgreen, Wenner, & Rayburn, 1980; Wenner, 1985, 1986) organize human needs met by media in many categories, including, for example, Wenner (1986):
· Surveillance (e.g., to obtain information about daily life)
· Entertainment/Diversion (e.g., to get away from usual cares and problems)
· Interpersonal utility (e.g., to get interesting things to talk about)
· Parasocial interaction (e.g., to encounter human qualities)
Researchers can use this typology as a basis to test attitudes people have toward media consumption with regard to variables. Two such variables are as gratification obtained (GO), defined as "the perceived outcome of engaging in a particular behavior," and gratification sought (GS), "the seeking of a value outcome mediated by the expectancy of obtaining that outcome" (Palmgreen, Wenner, & Rayburn, 1981, p. 473). Using these variables, models of user attitudes, such as attitudes of satisfaction, can be tested (Palmgreen & Rayburn, 1985).
In applying uses-and-gratifications to communication, an important consideration is to make sure that the variables such as GS and GO are measured at the level of abstraction for use in a model. In uses-and-gratifications studies, "measurement of GS and GO at the same level of abstraction is crucial to direct comparison of communication outcomes with what is sought" (Palmgreen, 1984, p. 34). A GO measured from a subject's "favorite TV news program" (media object) can substantially exceed GS as measured at the level of program type (TV news, which is a media class; Palmgreen, Wenner, & Rayburn, 1981). The result is that a model of attitudes that uses these measures of GS and GO at different levels, particularly a discrepancy model that uses the difference between GS and GO in its equation, is faulty in its construction (Palmgreen, 1984, p. 34).
In order to illustrate how the definitions of media class, object, and instance can assist a researcher using uses-and-gratifications measurements in a communication context, I define the following media class, object, and instance:
· Media class WebMag-the set of all Web-based magazines delivered through Web servers and observed using a Web client.
· Media object-the International Teletimes, Web-based magazine at http://www.wimsey.com/teletimes.root/teletimes_home_page.h tml as observed through the Netscape 1.1 for X client.
· Media instance t-media object T during its June 1995 circulation period.
We can then measure a subject's attitude in terms of the variables GS and GO based on their experience of media instance t, and also their attitudes about media object T as well as media class WebMag. Typically, these measurements are made on a five-point Likert scale asking users to indicate how much they agree with a given statement on a scale ranging from strongly agree to strongly disagree. A model that questions users at the media class level would ask the user questions such as:
· GS question-I read Web-based magazines to relax.
· GO question-Web-based magazines help me relax.
At the media object level, the questions would be specific to particular content:
· GS question-I read International Teletimes to relax.
· GO question-International Teletimes helps me relax.
Note that at this object level, the client is implied to be the user's own Web client.
At the instance level, the particular media object occurring at a fixed time would be the focus:
· GS question-I want to read the June 1995 International Teletimes in order to relax.
· GO question-The June 1995 International Teletimes helped me relax.
The flexibility of these units can permit many media classes, objects, instances, and experiences to be defined for analysis. For example:
· Media class UsenetAlt-the set of all Usenet alt.* newsgroups.
· Media object H-alt.hypertext viewed through Netscape 1.1 for X as served on the usenet.foo.edu server.
· Media instance h-alt.hypertext viewed through Netscape from the usenet.foo.edu server, articles 2498 to 2523 (article numbering may vary across Usenet servers, but, of course, viewers accessing articles on the same server will see the same articles with the same numbers).
The overall purpose of a study may be to make cross-comparisons at the media class level:
· Media class DuComm-the content (discourse plus static MOO structures) of the Diversity University MOO's Communications center.
· Media class LmEntry-the content (discourse plus static MOO structures) of the Lambda MOO's new user entry room.
The purpose of a study might be to compare the discourse in these two different contexts of MOO use (e.g., comparing discourse in these two "rooms" on different MOOs). In order to collect more specific data on user experiences in these classes, the researcher must define objects and instances. The server and content were specified in the class definition. The object definition requires the researcher to specify the client and then instances based on time. For example:
· Media object D-a member of DuComm as viewed through the XYZ MOO client
· Media instance d-D from 10:00 a.m. EST to 11:00 a.m. EST on June 24. 1995.
· Media object L-elements of LmEntry as viewed through the XZY MOO client
· Media instance l-D from 10:00 a.m. EST to 11:00 a.m. EST on June 24, 1995.
The researcher could either analyze the transcript of the discourse (a language approach) or question users' experiences of d and l, or attitudes with regard to D, L, DuComm, or LmEntry. The key is that by carefully defining the levels at which these measurements take place, the research can have a consistent framework for comparing different media experiences.
Media classes can also be described according to their mediated characteristics, as listed above, for time delay, distribution, and media type. For example, we can define media class JavaSun as "all the content available on the server java.sun.com that is observable through a Java-enabled Web client."
The characteristics of individual objects of type JavaSun share these characteristics: time (persistent), distribution (server broadcast), and media type (hypertext with applet, which is an executable content written using the Java language).
All objects in media class J then have these class characteristics. For example, we can define an object as follows: Object S is the spreadsheet at http://java.sun.com/applets/applets/spreadsheet/index.html viewed through HotJava alpha release 3 for Windows '95.
This object has its content and server specified by a URL and its client specified as a particular version of a Java-enabled browser. Object S shares the time, distribution, and media type characteristics as all other objects of type JavaSun. A subclass of JavaSun might be defined in terms of a specific browser: Media class JavaSunW95 includes all the content available on the server java.sun.com that is observable through the HotJava browser for Windows '95.
Clearly, S is an object of type JavaSunW95, and all objects of type JavaSunW95 share the same time, distribution, and media type characteristics as objects of type JavaSun. Using class hierarchies consisting of classes and subclasses, related classes can inherit characteristics.
Media classes and objects can be represented graphically. Figure 3 shows symbols that can be used for content, server, and a client.
Figure 4 shows these symbols used to illustrate several media classes. The left figure shows a media class called Web, which is defined by any Web content on any Web server through any Web browser. The dashed lines indicate a variety of possible values for the content, server, or client in the class.
The diagram in the middle is for the class JavaSun, a subclass of Web that consists of all the applets on the server java.sun.com that are observable through a Java-enabled client.
Finally, the right diagram in Figure 4 shows the class JavaSunW95, which is a subclass of JavaSun. This class consists of all applets on the java.sun.com server as viewed through the alpha 3 release of the HotJava browser for Windows '95.
Media objects can be shown using the same notation as for classes, but with solid lines for the symbols, as a media object has all of its content, server, and client attributes specified. Figure 5 shows three media objects of type JavaSun. The content of each object is the spreadsheet http://java.sun.com/applets/applets/spreadsheet/index.html. The server is java.sun.com, and the three Java-enabled Web clients shown in the diagram complete the definition for three media different objects.
Articulating the vast possibilities for communication on the Internet can be approached using a system of definitions. A definition for Internet communication relies on the precise meaning and technical implication of the words, Internet, computer, mediated, and communication. In approaching Internet communication as a range of media, we can define units of analysis: media space, media class, media object, media instance, and media experience, based on an articulation of levels of abstraction. Using the multiple levels of abstraction inherent in these definitions, communication researchers can define a variety of research settings with a consistent treatment of units of analysis. The overall benefit of this definition of Internet communication is that the diverse landscape of Internet communication can be defined, opening up possibilities for cross-study comparisons as well as theoretical integration.
In addition to using the above definition of Internet communication and the definition framework for media classes, objects, instances, and experiences, communication researchers can define their research setting using consistent terminology and units of analysis. Specifically, the benefits of this use are (a) to provide ways to define units of analysis for measuring variables in many communication contexts, (b) to allow consistent articulation of units and of analysis for study and for cross-study comparisons, and (c) to allow media class and object definitions to articulate the differences in levels of abstraction for measuring experiences based on different media objects.
More work remains in
exploring how this system of definitions can be applied to more
general on-line communication contexts. The client-server component
of this definition is based on a communications system employing
client-server communication on the Internet. The growing interconnections
among Internet communication and non-Internet services will open
up more possibilities for communication and an even more diverse
on-line communication environment.
Baron, Naomi, S. (1984).
Computer-mediated communication as a force in language change.
Visible Language, 18(2), 118-141.
Black, S., Levin, J.,
Mehan, H., & Quinn, C. N. (1983). Real and non-real time interaction:
Unraveling multiple threads of discourse. Discourse Processes,
6(1), 59-75.
Blumler, J. G. (1979).
The role of theory in uses and gratifications studies. Communication
Research, 6(1), 9-36.
Chew, J., & Yanoff,
S. (1995). Inter-network mail guide. [On-line]
Available:
file://ftp.csd.uwm.edu/pub/internetwork-mail-guide.
Daft, R. L., &
Lengel, R. H. (1984). Information richness: A new approach to
managerial behavior and organizational design. Research in
Organizational Behavior, 6, 191-233.
Daft, R. L., &
Lengel, R. H. (1986). Organizational information requirements,
media richness and structural design. Management Science, 32,
554-571.
Daft, R. L., Lengel,
R. H., & Trevino, L. K. (1987). Message equivocality, media
selection, and manager performance: Implications for manager performance.
MIS Quarterly, 11, 355-66.
December, J. (1995a).
Internet tools summary. [On-line] Available: http://www.rpi.edu/Internet/Guides/decemj/internet-tool
s.html.
December, J. (1995b).
Presenting Java. Indianapolis, IN: Sams.Net.
December, J., &
Randall, N. (1995). The World-Wide Web unleashed (2nd ed.).
Indianapolis, IN: Sams.Net.
DeSanctis, G., &
Gallupe, R. B. (1987). A foundation for the study of group decision
support systems. Management Science, 33, 589-609.
Feenberg, A. (1989).
The written world: On the theory and practice of computer conferencing.
In R. Mason & A. Kaye (Eds.), Mindweave: Communication,
computers and distance education (pp. 22-39). Oxford: Pergamon
Press.
Feenberg, A. (1992).
From information to communication: The French experience with
videotex. In M. Lea (Ed.), Contexts of computer-mediated communication
(pp. 168-187). New York: Harvester Wheatsheaf.
Ferrara, K., Brunner,
H., & Whittemore, G. (1991). Interactive written discourse
as an emergent register. Written Communication, 8(1), 8-34.
Finnegan, R. H. (1988).
Literacy and orality: Studies in the technology of communication.
Oxford, England: Blackwell.
Fulk, J., Schmitz,
J., & Steinfield, C. (1990). A social influence model of technology
use. In J. Fulk & C. Steinfield (Eds.), Organizations and
communication technology (pp. 117-140). Newbury Park, CA:
Sage.
Fulk, J., Steinfield,
C. W., Schmitz, J., & Power, J. G. (1987). A social information
processing model of media use in organizations. Communication
Research, 14(5), 529-552.
Georgoudi, M., &
Rosnow, R. L. (1985). The emergence of contextualism. Journal
of Communication, 35(1), 76-88.
Gurak, L. (1994). The
rhetorical dynamics of a community protest in cyberspace: The
case of Lotus marketplace. Unpublished doctoral thesis, Rensselaer
Polytechnic Institute, Troy, NY.
Harnad, S. (1991).
Post-Gutenberg galaxy: The fourth revolution in the means of production
of knowledge. The Public-Access Computer Systems Review, 2(1),
39-53.
Havelock, E. A. (1986).
The muse learns to write: Reflections on orality and literacy
from antiquity to the present. New Haven, CT: Yale University
Press.
Hiltz, S. R., &
Turoff, M. (1978). The network nation: Human communication
via computer. Reading, MA: Addison-Wesley.
Infante, D. A., Rancer,
A. S., & Womack, D. F. (1993). Building communication theory
(2nd ed.). Prospect Heights, IL: Waveland Press.
Innis, H. A. (1972).
Empire and communications (Rev. ed.). Toronto: University
of Toronto Press.
Internet Society (1995).
Frequently asked questions: What is the internet? [On-line] Available:
gopher://gopher.isoc.org:70/00/isoc/faq/what-is-internet.txt.
Johansen, R., Vallee,
J., & Spangler, K. (1979). Electronic meetings: Technical
alternatives and social choices. Reading, MA: Addison-Wesley.
Kiesler, S. (1986).
The hidden message in computer networks. Harvard Business Review,
64(1), 46-58.
Kiesler, S., Siegel,
J., & McGuire, T. W. (1984). Social psychological aspects
of computer-mediated communication. American Psychologist,
39, 1123-1134.
Kling, R., & Gerson,
E. M. (1977). The social dynamics of technical innovation in the
computing world. Symbolic Interaction, 1(11), 132-46.
Lakoff, R. T. (1982).
Some of my favorite writers are literate: The mingling of oral
and literate strategies in written communication. In D. Tannen
(Ed.), Spoken and written language: Exploring orality and literacy,
Advances in discourse process (pp. 239- 260). Norwood, NJ:
Ablex.
Lea, M. (1992). Introduction.
In M. Lea (Ed.), Contexts of computer-mediated communication
(pp. 1-6). New York: Harvester Wheatsheaf.
Lea, M., & Spears,
R. (1991a). Computer-mediated communication, de-individuation,
and group decisionmaking. In S. Greenberg (Ed.), Computer-supported
cooperative work and groupware, people and computers (pp.
155-173). London: Academic Press.
Lea, M., & Spears,
R. (1991b). Computer-mediated communication, de-individuation,
and group decisionmaking. International Journal of Man-machine
Studies, 34, 283-301.
Lengel, R. H., &
Daft, R. L. (1988). The selection of communication media as an
executive skill. Academy of Management Executive, 2(3),
225-232.
Levinson, P. (1990).
Computer conferencing in the context of the evolution of media.
In L. M. Harasim (Ed.), On-line education: Perspectives on
a new environment (pp. 3-14). New York: Praeger.
Littlejohn, S. W. (1989).
Theories of human communication (3rd ed.). Belmont, CA:
Wadsworth.
Markus, M. L. (1987).
Toward a "critical mass" theory of interactive media:
Universal access. Communication Research, 14(5), 491-511.
Markus, M. L. (1990).
Toward a critical mass theory of interactive media: Universal
access, interdependence and diffusion. In J. Fulk & C. Steinfield
(Eds.), Organizations and communication technology (pp.
194-218). Newbury Park, CA: Sage.
Martin, L., O'Shea,
T., Fung, P., & Spears, R. (1992). 'Flaming' in computer-mediated
communication: Observations, explanations, implications. In M.
Lea (Ed.), Contexts of computer-mediated communication (pp.
89-112). New York: Harvester Wheatsheaf.
McGuire, T. W., Kiesler,
S., & Siegel, J. (1987). Group and computer-mediated discussion
effects in risk decision making. Journal of Personality and
Social Psychology, 52(5), 917-930.
McGuire, W. J. (1974).
Psychological motives and communication gratification. In J. G.
Blumler & E. Katz (Eds.), The uses of mass communications:
Current perspectives on gratifications research (pp. 167-196).
Beverly Hills, CA: Sage.
McGuire, W. J. (1983).
A contextualist theory of knowledge: Its implications for innovation
and reform in psychological research. In L. Berkowitz (Ed.), Advances
in experimental social psychology, Vol. 16 (pp. 1-47). New
York: Academic Press.
McLuhan, M. (1964).
Understanding media: The extensions of man. New York: McGraw-Hill.
McLuhan, M. (1965).
The Gutenberg galaxy: The making of typographic man. Toronto:
University of Toronto Press.
McLuhan, M., &
Fiore, Q. (1967). The medium is the massage. New York:
Bantam Books.
McLuhan, M., &
Powers, B. R. (1989). The global village: Transformations in
world life and media in the 21st century. New York: Oxford
University Press.
Miles, I. (1992). When
mediation is the message: How suppliers envisage new markets.
In M. Lea (Ed.), Contexts of computer-mediated communication
(pp. 145-167). New York: Harvester Wheatsheaf.
Murray, D. E. (1991).
The composing process for computer conversation. Written Communication,
8(1), 35-55.
Ochs, Elinor(1989). Language, affect, and culture. Cambridge, England:
Cambridge University Press.
Ong, W. J. (1977).
Interfaces of the word: Studies in the evolution of consciousness
and culture. Ithaca, NY: Cornell University Press.
Ong, W. J. (1982).
Orality and literacy: The technologizing of the word. New
Accents Series. London: Methuen.
Palmgreen, P. (1984).
Uses and gratifications: A theoretical perspective. Communication
Yearbook, 8, 20-55.
Palmgreen, P., &
Rayburn, J. D. (1985). A comparison of gratification models of
media satisfaction. Communication Monographs, 52, 334-346.
Palmgreen, P., Wenner,
L. A., & Rayburn, J. D. (1980). Relations between gratifications
sought and obtained: A study of television news. Communication
Research, 7, 161-192.
Palmgreen, P., Wenner,
L. A., & Rayburn, J. D. (1981). Gratification discrepancies
and news program choice. Communication Research, 8, 451-478.
Quarterman, J. S. (1990).
The matrix: Computer networks and conferencing systems worldwide.
Bedford, MA: Digital Press.
Rapaport, M. (1991).
Computer mediated communications: Bulletin boards, computer
conferencing, electronic mail, and information retrieval.
New York: Wiley.
Rice, R. E., &
Shook, D. E. (1990). Voice messaging, coordination, and communication.
In J. Galegher, R. E. Kraut, & C. Egido (Eds.), Intellectual
teamwork: Social and technological foundations of cooperative
work (pp. 327-350). Hillsdale, NJ: L. Erlbaum Associates.
Rogers, E. M. (1983).
Diffusion of innovations (3rd ed.). New York: Free Press.
Rogers, E. M. (1986).
Communication technology: The new media in society. New York:
Free Press.
Rutter, D. R. (1987).
Communicating by telephone. Oxford, England: Pergamon Press.
Schmitz, J., &
Fulk, J. (1991). Organizational colleagues, media richness, and
electronic mail: A test of the social influence model of technology
use. Communication Research, 18(4), 487-523.
Shank, G. (1993). Abductive
multiloguing the semiotic dynamics of navigating the net. The
Arachnet Electronic Journal on Virtual Culture, 1(1). [Online].
Available:
ftp://ftp.lib.ncsu.edu/pub/stacks/aejvc/aejvc-v1n01-shank-abductive.
Siegel, J., Dubrovsky,
V., Kiesler, S., & McGuire, T. (1986). Group processes in
computer- mediated communication. Organizational Behavior and
Human Decision Processes, 37, 157-187.
Spears, R., & Lea,
M. (1992). Social influence and the influence of the 'social'
in computer- mediated communication. In M. Lea (Ed.), Contexts
of computer-mediated communication (pp. 30-65). New York:
Harvester Wheatsheaf.
Spears, R., Lea, M.,
& Lee, S. (1990). De-individuation and group polarization
in computer- mediated communication. British Journal of Social
Psychology, 29, 121-134.
Spitzer, M. (1986).
Writing style in computer conferences. IEEE Transactions
on Professional Communications, 29(1), 19-22.
Sproull, L., &
Kiesler, S. (1986). Reducing social context cues: Electronic mail
in organizational communication. Management Science, 32(11),
1492-1513.
Trevino, L., Lengel,
R. H., Bodensteiner, W., Gerloff, E., & Muir, N. K. (1990).
The richness imperative and cognitive style: The role of individual
differences in media choice behavior. Management Communication
Quarterly, 4, 176-197.
Trevino, L. K., Daft,
R. L., & Lengel, R. H. (1990). Understanding managers' media
choices: A symbolic interactionist perspective. In J. Fulk &
C. Steinfield (Eds.), Organizations and communication technology
(pp. 71-94). Newbury Park, CA: Sage.
Trevino, L. K., Lengel,
R. H., & Daft, R. L. (1987). Media symbolism, media richness,
and media choice in organizations: A symbolic interactionist perspective.
Communication Research, 14(5), 553-74.
Valente, T. (1991).
Thresholds and critical mass: Mathematical models of diffusion
of innovation. Unpublished doctoral dissertation, University
of Southern California, Annenberg School for Communication.
Vallee, J. (1982).
The network revolution: Confessions of a computer scientist.
Berkeley, CA: And/Or Press.
Wenner, L. A. (1985).
The nature of news gratifications. In K. E. Rosengren, L. A. Wenner,
& P. Palmgreen (Eds.), Media gratifications research: Current
perspectives (pp. 171- 193). Beverly Hills, CA: Sage.
Wenner, L. A. (1986). Model specification and theoretical development in gratifications sought and obtained research: A comparison of discrepancy and transactional approaches. Communication Monographs, 53, 160-179.
Notes:
John December is a PhD candidate in Communication and Rhetoric at Rensselaer Polytechnic Institute in Troy, New York.
Copyright 1996 Journal of Communication 46(1) Winter. 0021-9916/96