Content-Type: text/html
Co-use and co-processing of news media in the family:
An explication and empirical validation
Submitted to the "Media in the Family" Competition
Mass Communication and Society Division
The Association for Education in Journalism and Mass Communication
1 April 2000
Christian Sandvig
Melissa Nichols Saphir
Steven Chaffee
Department of Communication
Stanford University
Correspondence to: Christian Sandvig
Department of Communication
Stanford University
Stanford, CA 94305-2050
E-mail: [log in to unmask]
Phone: (650) 941-0787
FAX: (650) 725-2472
Co-use and co-processing of news media in the family:
An explication and empirical validation
Abstract
This study considers the sharing of media in the family by developing two concepts similar to co-viewing and mediation but that apply to communication media other than television. Termed co-use and co-processing, this paper first explicates these concepts, then presents preliminary empirical evidence that these concepts exist from a survey of parent-adolescent pairs. We find that families widely co-use media other than television, mutually co-process content from these media, and that adolescents often initiate co-processing.
Co-use and co-processing of news media in the family
Co-use and co-processing of news media in the family:
An explication and empirical validation
Apart from parental co-viewing and mediation to help a child learn to cope with television, very little research has examined the commonplace activity of people sharing the consumption of mass media. In this paper, we develop two concepts that describe behaviors similar to co-viewing and mediation in relation to communication media other than television. The first concept is co-use, meaning the physical sharing of a given media channel. The second concept is co-processing, a term we have coined to refer to interpersonal discussion of content that comes via a channel. Like much of the literature on co-viewing and mediation, we focus in this paper on co-use of media within families and on co-processing of media content between adolescents and their parents.
Our approach differs from related research literatures in several important respects. An obvious difference is that the research on family co-viewing and mediation has been peculiar to television. Indeed, the very term "co-viewing" all but excludes media that do not have screens to be viewed. In contrast, we expect co-use and co-processing to be behaviors that cut across audio, print, and computer media, in addition to television. A less obvious difference has to do with the fact that the mediation literature has been decidedly one-sided in its primary emphasis on parents' role as mediators of television effects on children. We also consider whether the sharing of mass media offers opportunities for children and adolescents to engage in "mediation" for their parents.
After explicating co-use and co-processing, this paper presents preliminary empirical evidence that these concepts exist, from a secondary analysis based on a survey of parent-adolescent pairs. We use these data to test empirically whether parents and children co-use media other than television and whether they mutually co-process content from these media.
The legacy of co-viewing and mediation
From the co-viewing literature it is known that parents and children co-use television in the sense of being physically proximate when viewing (Borzekowski & Robinson, 1999; Dorr, Kovaric & Doubleday, 1989). Scholars differentiate between co-viewing and mediation, defined in these studies as parent-child discussion of television content, in part because these two behaviors differ in their effects on children's interpretation and understanding of television content. For example, mediation by adults increases children's comprehension (Collins, Sobol & Westby, 1983; Desmond et al., 1985) and recall (Watkins et al., 1980) of program details and the extent to which children benefit from pro-social programs (Corder-Bolz, 1980). Mediation also decreases children's perception that television programs are realistic (Austin, Roberts & Nass, 1990). In the absence of mediation, co-viewing alone has little effect on children's comprehension or learning from educational programming (Dorr, 1989). When children and parents co-view violent programming and parents do not mediate the violent content through discussion, the children are more likely to engage in aggressive behavior later (Nathanson, 1999). The presence of co-viewers also affects the choice of program viewed (Webster & Wakshlag, 1982). For example, children are more likely to see violent programming when co-viewing with their parents than when viewing alone (Chaffee & Tims, 1976).
Studies of co-viewing and mediation have contributed to our understanding of parents' role in mitigating negative effects and amplifying positive effects of television on children. As concepts for future communication research, however, co-viewing and mediation are extremely limiting. In particular, these terms unnecessarily restrict scholars' focus to the medium of television and to dependent variables within children.
Moving beyond television. Communication research might profitably theorize about the variables that are truly at work in mediated interaction, rather than treating the medium at hand as a "black box" (Nass & Mason, 1990). To the extent that co-viewing and mediation apply only to television, they have been "black box" topics of research, implying either that co-viewing is a behavior that accompanies a collection of electrical components that is known as a television set or some particular collection of content at a particular time. If the characteristics of television sets change, if other media become common in the home, and if content takes new forms it becomes difficult to generalize from co-viewing. One reason, then, to develop concepts of co-use and co-processing is to theorize about uses and effects of new media, including technologies that do not yet exist and media that may one day replace television.
Until recently, television has certainly been a defensible object of study. In the 1970s, when co-viewing began to be examined widely (e.g., Bower, 1973; Brown & Linn‚, 1976; Chaffee & Tims, 1976; LoSciuto, 1971), television was arguably the central communication technology in the lives of families. But even at that time, some scholars recognized that media other than television are shared. For example, Johnstone (1974) documented that adolescents go to movies, listen to the radio, and collect records with their friends. In a study of adults, Katz, Gurevitch, and Haas found that "the content of conversation is contributed primarily by newspapers and books" (1973, p. 172). In other words, adolescents and adults co-use movies and music media, and adults co-process information from print media.
Newer interactive media are also shared. In video game parlors, for example, pairs and groups of teenagers engage in competitive and cooperative play and converse with each other while playing (Michaels, 1993). Sandvig's (2000) observations of a public computer lab for children revealed children share a single computer with other children about half the time, while one in ten visitors brings an older sibling or adult, and these pairs use the computers collaboratively. There is no reason to believe that such co-use and co-processing is limited to public settings. It is easy to envisage a home in which a parent encourages an adolescent to read a newspaper article at breakfast; the pair listens to the radio together in the car; and the adolescent tells the parent in the evening about a new World Wide Web site. In the most comprehensive recent assessment of media in the American home, Roberts, Foehr, Rideout, & Brodie concluded from a large national sample of 7th through 12th graders that a wide variety of media beyond television (videos, movies, video games, computer games, chat rooms, and Web sites) are co-used to at least some degree[1]-for instance 39% of children co-used World Wide Web sites (1999, p. 64).[2]
As these examples suggest, another reason to develop concepts broader than co-viewing and mediation is that an atheoretical focus on television overemphasizes viewing relative to other common media-related behaviors. Like the restricted focus on television, the emphasis on shared viewing -- to the exclusion of shared reading, listening together, and other forms of co-use -- is misplaced. Particularly in the family setting, it is the extent of shared use that has been of key theoretical interest in mediation, not any particular distinction of viewing behavior vs. other kinds of media use.
Moving beyond child dependent variables. Effects on children of adult-oriented television violence and profanity is a longstanding concern in media research. It is therefore not surprising that the purpose of early studies of co-viewing and mediation was to determine whether and how parents can mitigate harmful effects of television on their children (McLeod, Fitzpatrick, Glynn & Fallis, 1982). It follows that the dependent variables of main interest have been children's knowledge, attitudes, feelings, and behavior; parent variables have been of interest only to the extent that they "mediate" effects of television on these child behaviors. Because the dominant model of socialization has assumed top-down influence from socialization agents such as television and adults on children, the possibility that children and adolescents influence parents through discussion of media content went unexamined.
We offer the term "co-processing" as a substitute for "mediation" because the latter term implies a top-down, parent-to-child socialization model and obscures the potential role of children and adolescents as agents in their parents' socialization. We now know, for example, that adolescents influence their parents' socialization to politics (McDevitt & Chaffee, 199x), in part through adolescent-initiated discussion of news content (Saphir & Chaffee, 2000). As a result of parallel findings across many other domains, child development scholars now embrace bi-directional models of socialization (Maccoby, 1992). In these models, parent-child interactions lead to learning and development in both directions, i.e., in parents as well as in children. It therefore seems appropriate to conceptualize family discussion of media content as co-processing, a non-directional term. This conceptualization allows us to treat the direction of influence in a particular instance of co-processing as an empirical question, rather than an unquestioned assumption embedded in the name of the concept.
Summary. Related research to date has somewhat arbitrarily focused on shared use of television and interpersonal communication activity that surrounds it. These investigations of co-viewing and mediation have privileged one-way parent-to-child mediation of television content, mostly because the research has been geared toward parents as advice on child-rearing. The proliferation of media in the home and evidence for mutual socialization call into question, respectively, the primacy of television and the child-as-target assumptions in the co-viewing and mediation literature. We understand the co-viewing/mediation literature as a line of research concerned fundamentally with socialization in the family, where the effects of mediated messages are modified by, first, shared use of the medium, and second, interpersonal communication about the messages. Given this perspective, there is no a priori reason to assume that interpersonal interaction would be restricted to television, nor why its effects would be restricted to children alone.
It is not lightly that we argue for the addition of two new concepts to the communication lexicon, but expanding co-viewing and mediation to breach the traditional boundaries of television, viewing, and children-while remaining within the conceptualizations of the literature to date-is a near-impossible task. This leads us inevitably to a new theoretical view of these phenomena.
Conceptualizing Co-Use and Co-Processing
Defining co-use. Put simply, co-use describes the degree to which two or more individuals use a medium together. In addition to co-viewing of television, co-use refers to the sharing of any medium, including print, audio, video, or computer-based technologies. This shared use might be two family members viewing a television program at the same time, it might be sitting at the computer together while browsing the Web, but it might also be asynchronous sharing of a newspaper article, passed around at breakfast.
Defining co-processing. While the term "co-processing" implies cognition as well as behavior, in this study we define it operationally as discussion of mediated messages. Co-processing is interpersonal discussion of mediated messages. Whereas "mediation" implies a top-down model of parental influence on children, the term co-processing is intentionally ambiguous with regard to direction. This ambiguity reflects our openness to the possibility that adolescents influence their parents' use and processing of media. Thus, co-processing includes parent mediation of television content for children, and also includes child-initiated discussion of television with parents, as well as discussions initiated by either party with regard to other media. While co-processing may take the form of discussion of current news, co-processing is more than just discussion about current events in that co-processing is always centered around a message and the media that send and deliver it. Co-processing includes conversations about what to pay attention to and what to ignore in the media, about how to understand it, and cues as to whether one should laugh or cry at the result. In general, these activities involve mutual mental activation of the two individuals, which is the essence of co-processing conceptually.
Bryce and Leichter (1983) criticized research on mediation for assuming that mediation takes place only during co-viewing. They argued, based on their ethnographic observations of families, that important conversations about television occur when family members are not in the presence of the TV set. In our preliminary explication of co-use and co-processing, we accordingly propose that these behaviors are conceptually independent and behaviorally exclusive: family members may co-use, co-process, or both for any medium. Even if they both co-use and co-process a particular medium, they may not necessarily engage in both of these two activities at the same time. For example, several family members might go to the cinema together (co-use) but not discuss the movie until afterward (co-processing). Separating these behaviors conceptually, instead of assuming that one is a necessary or sufficient condition for the other, offers an opportunity to theorize separately about media differences, individual differences, and situational factors that could affect the degree of correlation between co-use and co-processing.
Summary. To validate our explication of the distinction between co-use and co-processing, we need to demonstrate that shared media use is not limited to TV and that it involves inter-generational relations that can be mutual, not simply unilateral.
Most of the empirical analyses presented in this paper address the validity of these assumptions. A prior issue, though, is that of the sheer existence of the activities these concepts describe. In our first analyses, we are therefore concerned with the question of whether co-use and co-processing in our broad, across-media and mutual sense can be said to exist in the real lives of American families. To propose a new concept the theoretical rationale and empirical evidence need to be clear and compelling.
To further focus our analyses, this study deals mostly with the sharing of news media, including co-use of particular news channels and co-processing of news content. Our data come from a field study of political socialization, a developmental process (in adults as well as in children) in which family discussion of news plays an important role (McDevitt & Chaffee, 1998). News is one kind of content that can be found across most media, a feature that enables us to examine co-use and co-processing as a concept across media. News is also relevant to all citizens in some measure, not exclusively a parental domain; as adolescents approach the age of voting news becomes relevant to them, as well. On some news topics, as on some media, adolescents may even be more expert than their parents.
Method
To validate our concepts of co-use and co-processing, we conducted secondary analysis of data from an evaluation of a voting-centered civics curriculum. The curriculum was used in some middle school and high school classes in California, during the fall 1998 state election campaign. The evaluation project provided a rich dataset for a preliminary examination of co-use and co-processing of news, in part because news content was widely available to the study participants, across many media. In particular, newspaper and television coverage of the election campaigns for governor and U.S. senator was extensive throughout the fall, and the major candidates maintained elaborate Web sites (Seipel, 1998). We will describe here only those aspects of the evaluation project that enter into our analysis.
Participants
The sampling frame consisted of public middle and high schools on the East Side of San Jose. Telephone interviews were conducted at three time points: In the summer of 1998 (wave 1), parents of students in these schools were interviewed and asked for permission to interview their adolescent children later in the year. Parents who agreed were re-interviewed in November, and the students were also interviewed during that period (wave 2). Finally, in the spring of 1999, parents and students were re-interviewed (wave 3). Unless noted otherwise, only the complete parent-student pairs that participated in all three waves are included in this analysis (N = 311 parent-student pairs).
The purposive sample of the original study was heterogeneous in terms of age, ethnicity, and education. The parent sample (which included some non-parent guardians) ranged in age from 31 to 78 years old; the median age was 44 years (SD = 7 years). Just over 75 percent of the adolescents were evenly divided among tenth, eleventh, and twelfth grades; the remaining 24 percent were evenly divided across the sixth through ninth grades. Nearly half of the parents (48%) were college graduates, and almost all (96%) had graduated from high school. As slightly more than half (56%) described themselves as white, there were substantial numbers of minority respondents -- notably those describing themselves as Hispanic (22%), Asian-American (9%), and African-American (6%).
Measures
Questionnaires, administered via computer-assisted telephone interviewing (CATI), were as similar as possible for both parents and adolescents. While the data are from a three-wave panel study, co-processing items were asked in identical form only in waves 2 and 3. Co-use items appeared in all three waves but were asked about different media at different time points, as noted in the tables that follow. In this analysis the data are treated cross-sectionally, regardless of which wave items came from.
The questions asked about co-use and co-processing of three media: television news, newspapers, and candidate information on the World Wide Web. The time frame specified for all of the questions that follow was "In the last month or so."
Co-use. We operationalized co-use with a set of two to three (generally yes/no) branching questions per medium. The first question determined the respondent's use, followed by a parallel question about the other interviewed family member. In other words, parents who said they had used the medium in the last month or so were then asked whether they had used the medium with the interviewed adolescent; in parallel fashion, adolescents who said they had used a medium in the last month or so were asked whether they had used that medium with the interviewed parent. Question wordings necessarily varied by medium, as the form of "use" changes across media. For television, "Have you watched any news on TV?" was followed by, "Has [child/parent] watched the news with you?" For newspapers, "How many days a week have you been reading a daily newspaper?" was followed by, "Does [child/parent] read a daily newspaper?" and then, "Do you and [child/parent] ever share the newspaper?" For the World Wide Web, "Have you personally used the World Wide Web?" was followed by, "Have you used the Web with [child/parent]?" The questions about television co-use were asked of parents only, so parent-adolescent agreement could not be calculated for television. Agreement on co-use was 62 percent for newspapers, 58 percent for the World Wide Web.
Co-processing. In contrast to traditional operationalizations of mediation that focus on what parents say to adolescents about television, we operationalized co-processing with a set of reciprocal yes/no questions for each medium. Each question in a set measured one direction of our bilateral model of influence, that is, whether the parent or the adolescent initiates co-processing of content from that medium. We asked about parents and adolescents encouraging each other to pay attention to news media, recognizing that this particular kind of conversation is only one of many possible indicators of co-processing. As with co-use, we necessarily operationalized co-processing slightly differently for different media because the language people use to encourage others to use a medium varies across media. Unlike the co-use items, however, the co-processing questions about television and newspapers were asked of all respondents, regardless of whether they reported having used these media in the last month or so. This follows Bryce and Leichter's (1983) observation that important contextual communication about television occurs even when family members are not in the presence of the television set.
For television, "Have you encouraged [child/parent] to pay attention to news that you were watching?" was followed by, "And has [child/parent] encouraged you to pay attention to news that [he/she] was watching?" For newspapers, "Have you suggested that [child/parent] read a news article?" was followed by, "And has [child/parent] suggested that you read a news article?" For the World Wide Web this screening question, "Have you visited a Web site with political news or voting information?" restricted the domain of co-processing. Respondents who answered yes to the screening question were asked, "Have you told [child/parent] about any Web sites with political news or voting information?" and, "And has [child/parent] told you about any Web sites with political news or voting information?" Parent-adolescent agreement on parent-initiated co-processing behavior was 57 percent for television, 70 percent for newspapers, and 64 percent for the Web. Agreement on adolescent-initiated co-processing behavior was 53 percent for television, 63 percent for newspapers, and 60 percent for the Web.
We consider the degree of parent-adolescent agreement on the co-use and co-processing questions, which ranged from 52% to 70% with a median of 61%, to be evidence of substantial validity of our measures. Not only are the two respondents in each family untrained participant observers, they were given no reason to expect any of these questions in advance nor to be cued to pay attention to incidents of co-use or co-processing. The fact that they agree to a significant degree marks a first step toward validation of the two concepts, even though this level of agreement might be considered weak if interpreted as coefficients of measurement reliability. It is notable that less agreement was obtained on reports of co-processing initiated by the student, who is the family member less widely expected to act in this role. Given that some of these interviews were conducted several weeks apart, and these questions asked for mutual reports about several behaviors of the two persons, to find greater than chance inter-observer concurrence encourages us that we are exploring a kind of activity that does indeed occur.
Analyses and Results
We conducted three analyses to validate our explication of co-use and co-processing. First, to test our assumption that co-use and co-processing are cross-media phenomena, we examine whether parents and adolescents engage in these behaviors with media other than television. Second, to test our assumption that co-processing is a mutual -- as opposed to parent-dominated -- interaction, we examine whether adolescents as well as parents initiate co-processing. To explore co-use and co-processing across media, our third analysis focuses on intercorrelations between the items.
As a pre-requisite for examining co-use and co-processing of news, it is important to note that even in the predominantly working class area from which study participants were drawn, contact with news media and the Internet was fairly widespread. As indicated in Table 1, 95 percent of parents and 90 percent of adolescents had watched television news in the past month; 83 percent of parents and 78 percent of adolescents had read a newspaper; and 70 percent of parents and 91 percent of adolescents had used the World Wide Web. Thus we can assume our study was relevant to most if not all our respondents.
Co-use across media. Table 1 also shows that parent-adolescent co-use of these three news media. As the literature would imply, television was a fairly common activity. A majority of parents reported that they had watched TV news together with the interviewed adolescent in the last month. (Adolescents were not asked about co-use of television news.) Smaller but still substantial numbers of parents and adolescents reported that they had co-used other news media. Fully a third of the parents reported sharing a newspaper or using the Web together with the interviewed adolescent. More than a third of the adolescents reported sharing a newspaper, while 19 percent reported co-using the Web with the interviewed parent.
Table 2 presents frequencies for parent-adolescent co-processing of news media. As expected based on the results for co-use (Table 1), more respondents reported co-processing news from television (63 percent by parent report and 70 percent by adolescent report) than from other media. Nevertheless, more than half of parents and of adolescents reported co-processing newspapers, and 10 percent of parents and 11 percent of adolescents reported co-processing the World Wide Web-surprisingly high numbers given the restrictive nature of the co-processing question for the Web (only telling about political or voting Web sites). In summary, the results Tables 1 and 2 are consistent with our assumption that co-use and co-processing are general behaviors, not limited to television.
Mutuality of co-processing. Comparison of the "parent initiated" row with the "child initiated" and "both initiated" rows in TableE2 (in boldface) makes clear that in the aggregate, adolescents are nearly as likely as parents to initiate co-processing of media. In some families, the student is more likely to do so. For television news, over a third of respondents reported either that both family members initiated co-processing or that only adolescents initiated co-processing of television news in the last month or so; nearly identical percentages reported that only parents initiated co-processing. More than 20 percent of respondents reported that co-processing of newspapers was initiated by both family members or by adolescents only. Co-processing of political Web sites was much less common than co-processing of the other two media. Nevertheless, mutuality prevails. Nearly equal percentages of respondents reported that both family members or adolescents only initiated co-processing, as compared to 6 percent reporting that only parents initiated co-processing of political Web sites. These results are consistent with our proposition that co-processing is a bilateral, mutual interaction between parents and adolescents.
Co-use and co-processing across media. To explore the relationship between co-use and co-processing for each medium of communication, we turn to an examination of bivariate Spearman correlations among the items in Table 3. Table 3 only correlates parent reports with each other, and adolescent reports with each other. The correlation matrix for parent reports is below the diagonal, for adolescents it is above.
While co-use and co-processing could be completely independent in some circumstances, we would generally expect that people who use a medium together would also be more likely to talk about content from that medium. To explore the relationship between co-use and co-processing, we examine the lower-left quadrant below the diagonal and find that parents are more likely to initiate co-processing about the same medium for which they report co-use. That is, the correlations in bold are somewhat larger than the surrounding correlations. Examining the upper-right quadrant above the diagonal, we find that children are not noticeably more likely to initiate co-processing about the same medium for which they report co-use. That is, it is difficult to distinguish the bold correlations from the surrounding correlations. Our expectation is supported for parents, but children do not seem to discriminate between media as much as parents.
Further, we can examine differences between media with respect to co-use or co-processing. Some media may be harder to co-process while not co-using. For example, it might be easier to tell you about something that was said on a TV show the next day that it is to explain a complicated Web page (especially one that is interactive). The more decisions that have to be made to operate the medium, the more co-processing we would expect while co-using. For television, this involves changing channels periodically, but many Web sites require frequent clicking and decision-making. Of the correlations in bold, we find that the Web is always the highest, supporting this inference.
Next, we consider co-use as a cross-media or media specific concept. Examining the upper-left quadrant below the diagonal, we see that for parents, reported co-use of a particular medium is not related to co-use of another medium. All three correlations are low, and only the correlation between co-use of newspapers and the Web reaches significance. For adolescents (upper-left quadrant above diagonal), reported co-use of a particular medium is not related to co-use of another medium to the extent that we can tell with the data available (only the World Wide Web and newspapers can be compared). We have shown in table 1 that families co-use at least three media, but the results in table 3 demonstrate that the pattern of media use in the home is one where when one medium is co-used there is no significant relationship with the co-use of other media. Families may have a habit of watching TV together, reading the newspaper together, or using the Web together, but we do not find that they tend to have all three of these habits.
Finally, we explore co-processing as cross-media or media specific concept. Because we have operationalized co-processing in this study using news content that may not be specific to a particular medium, we expect co-processing to be correlated across media. Turning to the lower-right quadrant below the diagonal, we find that parents that initiate co-processing for one medium may do so for any other. The lower-right quadrant below the diagonal shows the same pattern for adolescents. Note that whether this result should be interpreted as between-media consistency within families or within-topic (i.e., news and political information) consistency can not be determined with this data.
Discussion
Having established a case for the existence, and valid measurement, of co-use and co-processing of media, we should consider the implications of these concepts for other research concepts in mediated communication. We first address the implications for the family as a site of socialization, then consider the implications of applying these concepts across media, and finally conclude.
Socialization and the Family. The family as a unit of analysis is often neglected in communication research, yet it is a vital group for understanding socialization processes. Co-use and co-processing can be conceptualized as applying to the family unit as a whole (indeed, we concede that any group might display this behavior). Effects of co-processing might be conceptualized as applying either to the family or to an individual within it. Newcomb's A-B-X model describes the orientation of individuals A and B to a source, X, and to each other (Newcomb, 1953). Put in Newcomb's terms, co-processing is the variable describing the degree to which A and B select X, and having shared X, affect each other's understanding of X. Co-processing could then be considered a prerequisite for co-orientation. Newcomb's conception of the relationship of two individuals to the media where both individuals are on the same level, not hierarchically one above the other, has been largely disregarded in the literature that applies to socialization in the family. Rather than viewing the family as containing multiple actors with agency, research overwhelmingly addresses the influence of parents on children or media on children. As stated earlier, we prefer a bi-directional model where co-processing can involve any member of the family and be initiated by any member of the family. In this study we have found empirical support for such a model.
If co-processing can involve and be initiated by any family member, this implies it can affect any family member. We know from this study and others that parents and adolescents share many media and discuss media content, while previous television research has shown that mediation affects comprehension and recall. If empirical research allows us to expect change in the child as a result of this, it is unclear what would restrict the possibility of effects to only one half of this dyad. To understand the child as continually the dependent variable in an interaction is to neglect other communicative interactions within the family.
Co-viewing scenarios often involve the application of parental expertise to the child's inexperience. Upon consideration of our findings, it is certainly realistic to consider a scenario where these roles are reversed, particularly with new media, which children tend to adopt more readily than their parents. Adolescents can also be experts at some topics, and mediate the effects of a parent's use of media through discussion. Even when expertise is not at issue, it is possible to theorize effects on parents-in merely participating in the physical act of sharing a medium with a child and observing a child's reaction to content, a parent may come to view the child, that content, or the world it describes in a new way. Future research might then usefully investigate both the cross media and bi-directional (or "trickle-up") settings and consequences of co-processing and co-use.
Further, the possible independence of co-use and co-processing activities suggest that family members who co-process media content may not need to co-use a medium to be affected by it. The source of this content may potentially be any medium, but of critical interest are the technological media (print, television, radio, the Internet, and perhaps others) that are agents of socialization duplicated in every modern home.
Co-use and co-processing across many media. Questioning the limitation of previous research to "viewing" leads to a more subtle understanding of the importance of the underlying variable interactivity (Rafaeli, 1988). Television, during the time that the bulk of co-viewing research was performed, was a unidirectional source of content whose interactivity was limited to the choice of channel. Viewing television may not be entirely passive in a cognitive sense, but when compared to some other media (e.g., computers) it can seem a relatively passive act. No conceptual reason restricts the effects of co-processing to media that limit interactivity, however. Indeed, our findings allow us to preliminarily suggest that media that allow a greater degree of interactivity may increase the likelihood of co-processing, as these media offer more opportunity for decision-making and hence more potential triggers for discussion when they are co-used. Increased interactivity might, however, negatively affect co-use. Roberts et al found that "Interactive screen media_tend to support activities that a substantial proportion of children engage in alone." (1999: 65). Whether this is a function of the content, the degree (or type) of interactivity, or some other factor is not known.
Questioning the limitation of previous research to "television" leads us to a more flexible conceptualization of media. The time-specific conceptualization of media in previous research obscures both limitations of some findings and the potential for broader generalizations of others. Both communication technologies and the people that use them change as the technology diffuses through society (Rogers, 1995). Scholars of the social history of communication technology provide many illustrations of ways that the social context of media use changed as the medium diffused. When radios, television sets, and other media technologies were scarce and expensive they were a spectacle and generally displayed in social contexts -- families would often travel to other households to see and hear them (Sterling & Kittross, 1978; Douglas, 1987; Marvin, 1988). Use was then rare, but co-use was almost always the case. Co-use and co-processing also apply to communication with media not generally considered in media effects research. When even telephones were scarce they were often installed in local businesses where people would gather to meet and use them (Fischer, 1992) -- and this would often lead to what we would call co-processing; that is, discussion of news received by telephone.
Conclusion. After proposing co-use and co-processing as concepts, we have empirically established that a substantial number of families co-use not only television, but also newspapers and World Wide Web sites. While co-use of different media is widespread, little to no relationship exists between the co-use of one medium and the co-use of another. Families also co-process content from these media, and when this co-processing occurs, up to half of it is initiated by adolescents or by both adolescents and adults together. The co-processing of news and political content does not appear to be media dependant -- the initiation of co-processing of content from any particular medium is always related to some degree to co-processing content from any other medium. Whether it is the medium or the content that is co-processed is an open question pending further exploration, as different media may provide different content.
Finally, while the increasing spread of computer technology has fueled fears that new media will individuate the family, between one-third and one-half of the adolescents and parents we studied use the Web with each other. Our findings assert that the World Wide Web is bringing families together, as have other media that we have studied in the past. The lesson in this may be that the polemic regarding the possible negative effects of new media is more of a cultural response to new technology than a feature of behavioral science. There are surely effects, both positive and negative, that have not yet been discovered, but a medium in the home that prevents families from spending time together is not one of them.
References
Austin, R.W., Roberts, D.F., & Nass, C.I. (1990). Influences of family communication on children's television-interpretation processes. Communication Research, 17(4): 545-564.
Borzekowski, D.L.G., & Robinson, T.N. (1999). Viewing the viewers: Ten video cases of children's television viewing behaviors. Journal of Broadcasting and Electronic Media, 43(4): 506-528.
Bower, R.T. (1973). Television and the Public. New York: Holt, Rinehart, and Winston.
Brown, J.R., & Linn‚, O. (1976). The family as a mediator of television's effects. In R. Brown (ed.), Children and Television, p. 184-198. Beverly Hills, CA: Sage.
Bryce, J.W., & Leichter, H.J. (1983). The family and television: Forms of mediation. Journal of Family Issues, 4(2): 309-328.
Chaffee, S.H., Saphir, M.N., Graf, J., Sandvig, C., & Hahn, K.S. (1999). Attention to Counter-Attitudinal Messages in a 1998 State Election Campaign. Paper presented at the Association for Ecuation in Journalism and Mass Communication, New Orleans, LA.
Chaffee, S.H., & Tims, A.R. (1976). Interpersonal factors in adolescent television use. Journal of Social Issues, 32(4): 98-115.
Collins, W.A., Sobol, B.L., & Westby, S. (1981). Effects of adult commentary on children's comprehension and inferences about a televised aggressive portrayal. Child Development, 52(1): 158-163.
Corder-Bolz, C.R. (1980). Mediation: The role of significant others. Journal of Communication, 30(3): 106-118.
Douglas, S.J. (1987). Inventing American broadcasting, 1899-1922. Johns Hopkins University Press: Baltimore.
Desmond, R.J., Singer, J.L., Singer, D.G., Calam, R., & Colimore, K. (1985). Family mediation patterns and television viewing: Young children's use and grasp of the medium. Human Communication Research, 11(4): 461-480.
Dorr, A., Kovaric, P., & Doubleday, C. (1989). Parent-child coviewing of television. Journal of Broadcasting and Electronic Media, 33(1): 35-51.
Fischer, C.S. (1992). America calling: A social history of the telephone to 1940. University of California Press: Berkeley, CA.
Johnstone, John W.C. (1974). Social integration and mass media use among adolescents: A case study. In J.G. Blumler & E. Katz (eds.), The Uses of Mass Communications: Current Perspectives on Gratifications Research, p. 35-48. Beverly Hills, CA: Sage Publications.
Katz, E., Gurevitch, M., & Haas, H. (1973). On the use of the mass media for important things. American Sociological Review, 38(2): 164-181.
LoSciuto, L. A. (1971). A national inventory of television viewing behavior. In E. A. Rubinstein, G.A. Comstock, & J.P. Murray (eds.), Television and Social Behavior Reports and Papers Vol. IV: Television in Day-to-Day Life: Patterns of Use. Washington, D.C.: U.S. Government Printing Office.
Maccoby, E.E. (1992). The role of parents in the socialization of children: An historical overview. Developmental Psychology, 28(6): 1006-1017.
Marvin, C. (1988). When old technologies were new: Thinking about electric communication in the late nineteenth century. Oxford University Press: New York.
McDevitt, M., & Chaffee, S. H. (1998). Second chance political socialization: "Trickle-up" effects of children on parents. In T. J. Johnson, C. E. Hays, & S. P. Hays (Eds.), Engaging the Public: How Government and the Media Can Reinvigorate American Democracy (pp. p. 57-66). Lanham, MD: Rowman & Littlefield.
McLeod, J.M., Fitzpatrick, M.A., Glynn, C.J., & Fallis, S.F. (1982). Television and social relations: Family influences and consequences for interpersonal behavior. In D. Pearl, J. Lazar, & L. Bouthilet (Eds.), Television and Behavior: Ten Years of Scientific Progress and Implications for the 80s, Vol. 2, p. 272-286. Washington, DC: US Government Printing Office.
Michaels, J.W. (1993). Patterns of video game play in parlors as a function of endogenous and exogenous factors. Youth & Society, 25(2): 272-289.
Nathanson, A.I. (May, 1999). The relationship between parental mediation of violent TV and children's aggressive tendencies: The intervening role of children's orientations toward violent TV. Paper presented at the International Communication Association conference, San Francisco, CA.
Nass, C. & Mason, L. (1990). On the study of technology and task: A variable-based approach. In J. Fulk & C. Steinfeld (Eds.), Organizations and communication technology (pp. 46-67). Newbury Park: Sage.
Newcomb, T. (1953). An approach to the study of communicative acts. Psychological Review, 50(6), 393-404.
Rafaeli, S. (1988). Interactivity: From new media to communication. Sage Annual Review of Communication Research: Advancing Communication Science. 16: 110-134, Sage: Beverly Hills, CA.
Rogers, E. M. (1995). Diffusion of Innovations. (4th ed.) Free Press: New York.
Sandvig, C. (2000). The information apologue: Play and Internet access in the children's library. Paper presented at the International Communication Association conference, Acapulco, Mexico.
Saphir, M.N., & Chaffee, S.H. (2000). Adolescents' contributions to family communication patterns. Paper presented at the International Communication Association conference, Acapulco, Mexico.
Seipel, T. (October 8, 1998). Political sites explode on Internet. San Jose Mercury News, p. 1A.
Sterling, C. H. & Kittross, J. M. (1978). Stay tuned: A concise history of American broadcasting. Wadsworth: Belmont, CA.
Watkins, B., Calvert, S., Huston-Stein, A., & Wright, J.C. (1980). Children's recall of television material: Effects of presentation mode and adult labeling. Developmental Psychology, 16(6): 672-674.
Webster, J.G., & Wakshlag, J.J. (1982). The impact of group viewing on patterns of television program choice. Journal of Broadcasting, 26(1): 445-455.
Table 1.
Co-use frequency across media
Parent
Report
Adolescent
Report
Television a
Do not watch TV news
5%
10%
Watch news, but not together
31
*
Watch news together
60
*
Total (N = 311)
100
100
Newspaper b
Do not read newspaper
17%
22%
Read newspaper, do not share
41
35
Read newspaper and share
38
37
Total (N = 311)
100
100
World Wide Web b
Do not use Web
30%
9%
Use Web, but not together
39
60
Use Web together
32
19
Total (N = 311)
100
100
* = Not asked
a Parents asked at wave one, Summer 1998; Children asked at wave two, Fall 1998
b Asked at wave three, Spring 1999
Table 2.
Co-processing frequency and directionality across media
Parent
Report
Adolescent
Report
Television
No co-processing reported
37%
30%
Only parent initiated
31
35
Only child initiated
6
6
Both initiated co-processing
26
29
Total (N = 311)
100
100
Newspaper
No co-processing reported
47%
41%
Only parent initiated
30
36
Only child initiated
6
4
Both initiated co-processing
17
19
Total (N = 311)
100
100
World Wide Web
No co-processing reported
90%
89%
Only parent initiated
6
6
Only child initiated
2
4
Both initiated co-processing
2
2
Total (N = 311)
100
100
Table 3.
Spearman correlations between co-use and co-processing by medium (N in parentheses)
[1] The degree to which this co-use was with parents vs. with peers, siblings and others varied widely by age and medium.
[2] Specifically, this is the percentage of children that did not report that they "viewed or used" World Wide Web sites "mainly alone" in the last day.