11.5 Family-viewing context

By the late 1970s, two reviews of research on child development had concluded that television was more than a communicator of content because it organized and modified the home environment (Atman & Wohlwill, 1978; Majoriebanks, 1979). Conversely, it was known that the home environment organized and modified television viewing. For example, Frazer (1976) found that the family routine established the viewing habits* of preschoolers, not vice versa. Today we know that demographic differences, such as edinicity (Tangney & Feshbach, 1988) and individual differences, such as genetics (Plomin, Corley, DeFxies & Fulker, 1990), also influence the family-viewing context. This section deals with variables that mediate the effects of television in the home setting, including the home environment, coviewing, and viewing habits. For "television viewing occurs in an environmental context that influences what and when viewing occurs, as well as the ways in which viewers interpret what they see" (Huston et aL, 1992, p. 98).

11.5.1 Variables That Mediate

The variables in the farnily context for television viewing can be grouped into three categories: (a) the environment, which encompasses the number and placement of sets, the toys and other media available, options for other activities, rules for viewing, and parental attitudes and style; (b) coviewing, which includes the nature and frequency of interactions, the effect of attitudes, and the effect of age and roles; and (c) viewing habits, which are based on variables such as amount of viewing, viewing patterns or preferences, and audience involvement.* These variables interact to create a social environment that mediates the effects of viewing.

Mediating variables can be separated into two types of variables: dire " ct and indirect. Direct mediating variables are those that can be controlled, such as the situation or habits. Indirect mediating variables are those that are fixed, such as educational or socioeconomic level.

The research on television as a socializing agent is extensive and will be discussed later in this chapter. Although research on family context abounds, many findings are contradictory or inconclusive. Nevertheless, there is enough research to suggest some important interactions.

One approach to visualizing the relationship between program variables (e.g., formal features, content), context variables (e.g., environment, habits, coviewing), and outcome variables (e.g., attention, comprehension, attitudes) was presented by Seels in 1982 (see Fig. 11-2).

Another approach to conceptualizing visually the relationship of some of these mediating variables to exposure and outcomes was presented by Carolyn A. Stroman (1991) in Figure 11-3, which appeared in the Journal of Negro Education.

11.5.2 Theoretical Assumptions At the level of operational investigation of these variables, assumptions are made that affect the questions researched, methodologies used, and interpretation of findings. One such issue is how television viewing should be defined. As discussed in the message design and cognitive-processing section, classic studies by Allen (1965) and Bechtel, Achelpohl, and Akers (1972) found there was a great deal of inattention while the television set was turned on. If viewing is defined as a low level of involvement, i.e., nothing more then being in the room when the television set is on, the result is estimates of the big role of television in children's lives. When estimates of viewing by 5-year-olds made from parent-kept viewing diaries and time-lapse video recordings are compared, diaries yield estimates of 40 hours a week and time-lapse video recordings analyzed for attentive viewing yield 3 112 hours a week (Anderson, Field, Collins, Lorch & Nathan, 1985, cited in Comstock & Paik, 1987). Viewing is often defined as "including entering and leaving the room while intermittently monitoring what is unfolding on the screen" (Comstock & Paik, 199 1, p. 19).

On the other hand, current research on mental activities that occur during the television experience suggests that a great deal of mental activity can occur while viewing. Comstock and Paik (1991) suggest that a distinction be made between monitoring (paying attention to audio, visual, and social cues that indicate the desirability of attention to the screen) and viewing (paying attention to what is taking place on the screen).

The issue of whether the viewer is active or passive arises from differing conceptions of viewing and from the fact that research has established that the viewer can be either, depending on programming and the mediating variables. Comstock and Paik (1991) cite several classic and recent studies that established a high level of mental activity despite an often low level of involvement (Bryant, Zillmarm & Brown, 1983; Huston & Wright, 1989; Krendl & Watkins, 1983; Krull, 1983; Lorch, Anderson & Levin, 1979; Meadowcroft & Reeves, 1989; Thorson, Reeves &Schleuder, 1985). As previously noted in the section on message design and cognitive processing, the notion of hypnotic watching of television has been largely discredited (Anderson & Lorch, 1983; Bryant & Anderson, 1983).

Three studies by Argenta, Stoneman, and Brody (1986), Wolf (1987), and Palmer (1986) reinforce this conclusion. Wolf and Palmer interviewed children about their viewing to determine interest, thoughtfulness, and insight. Their study, therefore, is susceptible to the biases of self-reporting. Argenta et al. analyzed the visual attention of preschoolers to cartoons, Sesame Street, and situation comedies. They observed social interaction, viewing, and use of toys. With Sesame Street and situation comedies, attention was divided among social interaction, viewing, and toys. Only with cartoons did social interaction decrease. "The image of children mesmerized in front of the television set, forsaking social interaction and active involvement with their object environment, held true for only one type of programming, namely, cartoons" (Argenta et al., p. 370). Thus, findings will differ depending on how viewing is defined.

Another assumption is that incidental learning and intentional learning are separate during the television experience. Yet, if an adult reinforces or intervenes while coviewing a program for children, intentional learning will increase. And if a child learns indirectly through informative programming, incidental learning will increase. The nature of the television experience today, especially with cable and videocassette recorder (VCR) technology, may be that incidental and intentional learning happen concurrently and may even interact or reinforce each other. Coviewing with discussion may be a way to join incidental and intentional learning. In an article on "Family Contexts of Television," Leichter et al. (1985) point out that ways of representing and thinking about time may be learned from the television experience. Children can incidentally learn to recognize the hour or the day from the programming schedule. They can intentionally learn time concepts by watching Mister Rogers'Neighborhood and Sesame Street.

A methodological assumption underlying much research on the television viewing environment is the acceptability of self-reporting instruments and diaries. Although these techniques are valid, often they need to be compared with research results from other methodologies. This may be especially true in television research, because self-reporting techniques are used so extensively, particularly in studies on the family-viewing context.

11. 5.3 The Television Viewing Environment

The television viewing environment is part of the television viewing system, which results in a television viewing experience. This section will next address several categories and subcategories of mediating variables starting with the viewing environment.

11.53.1. Number and Placement of Sets. Leichter and her colleagues (1985) discuss the temporal and spatial organization of the television viewing environment. According to Leichter et al., there are symbolic meanings associated with the placement of television sets in the home. In their discussion of the methodological approaches to the study of family environments, they stress the need "to obtain a detailed picture of the ways in which television is interwoven with the underlying organization of the family" (p. 3 1). They decided that ethnographic or naturalistic data gathering through a variety of observation techniques was best. Therefore, they used participant observation, interviewing, recording of specific behaviors, and video and audio recording of interactions. To gather data over a sufficient time span, one observer moved in with the family. Leichter and her colleagues generated research questions through a study of three families followed by a study of ten families. They compared the data generated with a similar cross-cultural study done in Pakistan (Ahmed, 1983) and concluded that placement varies with the architecture of the home and with family perceptions. As a result, a set can be "fixed" or "static" in terms of its placement, just as individual position for viewing can be fixed or static. The area of placement can be close to traffic patterns or places of activity, or it can be set in out-of-the-way places reserved just for viewing. Where the set is placed may lead to conflict because of other activities.

Figure 11-2. Relationship among variable in the family-viewing context.(from Seels, 1982).

Figure 11-3. Hypothesized model for understanding television's socializing impact. (From Stroman, 1991.)

Even though television is a "magnet," especially for young viewers, the physical design of the area where the set is placed can inhibit the amount of time spent viewing. This conclusion is supported by research on use of dormitory viewing areas in college (Preiser, 1970, cited in Ross, 1979), Young children engage in many other activities in the television area even if the television isn't in a desirable location for other activities (Rivlin, Wolfe & Beyda, 1973, cited in Ross, 1979).

Winn (1977) argues that the television should be put in an out-of-the-way area such as the basement in order to minimize its dominance. Others argue that the more centrally located the set, the more likely viewers will be influenced by other powerful variables such as coviewing.

One concept that could be used in research on placement is "household centrality."* Medrich et al. (1982, cited in Comstock & Paik, 1991) proposed that families can be classified on a dimension reflecting behavior and norms* that favor viewing. If there is high use by parents and children and there are few rules governing viewing, the household can be said to have "centrality" of television. Research is needed on the effect of placement of set(s) on centrality. Generally, if there is only one set, it is in a living or group recreational area. If there is a second set, it is usually placed in a bedroom (Leichter et al.,1985). The more central the location, the greater the likelihood that social interaction or coviewing will mediate the effects of television.

The majority of households in the United States have two or more sets, subscribe to cable, and own a VCR (Huston et al., 1992). Children in multiple-set homes tend to watch more television than those in single-set homes (Webster, Pearson & Webster, 1986). Christopher, Fabes, and Wilson (1989) found that parents who owned one television set tended to exert more control over their children's viewing than did parents owning multiple sets. They also found that parents who owned three or more sets were more positive about their children's watching television and spent twice as much time watching as those with fewer sets. Webster et al. (1986) cautioned that multiple sets could lead to decreased parent-child interactions.

Since additional sets are used to resolve conflicts over program choices, children may view more since they have more control over their own viewing. In sum, one obvious guideline is that young children should not have access to more sets than parents can monitor. The experience of resolving conflicts over who watches what can provide valuable lessons in sharing.

11.53.2. Availability of Toys. Children develop strategies for viewing, including strategies that allow for competing activities, such as playing with dolls (Levin & Anderson, 1976). Rapid television pacing has no effect on the number of toys used during a play period (Anderson, Levin & Lorch, 1977). Family rules govern the placement and use of toys during viewing. Some families forbid toys in the television room; others permit toys to be available during viewing (Leichter et al., 1985), Where the set is placed may affect the use of toys during viewing. If the set is in the living room where no toys are permitted, the use of toys as distractors or reinforcers during viewing will be less than if the set is in the playroom or recreation room where toys and games are available.

The availability of toys may distract young children from the television set. In a study by Lorch, Anderson, and Levin (1979), when attractive toys were available to 5-yearolds, attention to Sesame Street dropped from 87% to 44%. One of the methods employed in the earliest research on Sesame Street was to conduct formative evaluation by having children watch a sequence while seated at a table filled with toys. If the children played with the toys rather than watching, the sequence was deemed ineffective in holding attention. Among these now classic studies were studies by Lesser in 1972 and 1974, and by Lorch and his colleagues in the late 1970s. When Lorch, Anderson, and Levin (1979) showed a version of Sesame Street to two groups, one group of children surrounded by toys and one group with no toys in the environment, the children in the group without toys attended twice as much. However, there was no difference between the groups in comprehension of television content. Thus, toys may be seen as positive elements of the viewing environment in that they can reinforce viewing and provide a basis for interaction with others about television and other topics. On the other hand, toys can decrease attention, but this phenomenon does not seem to affect cognitive learning.

It is commonly believed that children learn about life through forms of play and social interaction (D. Winn, 1985). Although television can model prosocial forms of interaction, the time spent watching television results in less time for play, practice, and real interactions with other children or family members.

Television has no sign on it: "Frespassers will be prosecuted." Television is living made easy for our children. It is the shortest cut yet devised, the most accessible back door to the grown-up world. Television is never too busy to talk to our children. Television plays with them, shares its work with them. Television wants their attention, needs it, goes to any length to get it (Shayon, 1950, p. 9).

It is likely that children watching television in an environment rich with toys and the opportunity for other activities will not be as mesmerized by television programming. Opportunities for elaboration, interaction, and creativity that extend the effect of the television stimulus should be

richer in such an environment. However, research is not available at this time to support such suppositions.

11.5.3.3. Relationship to Other Activities. Television impacts other activities, and other activities impact television. A study on television's impact conducted by Johnson in 1967 (cited in Liebert & Sprafkin, 1988) showed that of those surveyed, 60% changed their sleep patterns, 55% altered meal times, and 78% used television as an electronic babysitter. Liebert and Sprafkin also cite a study by Robinson in 1972 that showed reductions in sleep, social gatherings away from home, leisure activities, conversation, household care, and newspaper reading.

Television is frequently secondary to other activities, or there is frequently another activity even when viewing is primary (Comstock, Chaffee, Katzman, McCombs & Roberts, 1978). Krugman and Johnson (1991) report that compared to traditional programming, VCR movie rental is associated with less time spent on other activities.

Parental mediation and the incorporation of other activities as adjuncts to the viewing process may be beneficial for children. Friedrich and Stein (1975) concluded that when adults provide discussion after viewing or read storybooks that summarize important concepts conveyed in programming, children increase their understanding of concepts and are able to generalize them to new situations better than children not provided with summaries. Singer, Singer, and Zuckerman (1981) reached the same conclusion when they had teachers lead discussions following viewing of prosocial programs.

Some families engage in orienting activities prior to viewing that lead to awareness of program options. According to Perse (1990), heavy viewers tend to use television guides and newspaper listings to select programs. They reevaluate during exposure by grazing* (quickly sampling a variety of programs using zapping* techniques with remote controls) while they are viewing.

Some studies have shown that television viewing reduces time devoted to other activities (Murray & Kippax, 1978; Williams, 1986). Murray and Kippax collected data from three towns in Australia: a no-television town, a low-television town, and a high-television town. The lowtelevision town was defined as one receiving television for only I year, and the high as one receiving television for 5 years. Comparisons between the no-television town and the low-television town showed a marked decrease in other activities for all age levels when television was available. Television led to a restructuring of children's time use (Murray & Kippax, 1978; Himmelweit, Oppenheim & Vince, 1958, cited in Comstock & Paik, 1994). The displacement theory discussed in the section on school achievement attempts to explain the relationship of other activities to television viewing in the family context.

11.5-3.4. Rules for Viewing. The National Center for Educational Statistics conducted the National Education Longitudinal Study (NELS) of 1988. The study surveyed 25,000 eighth-graders, their parents, principals, and teachers.

A follow-up study was undertaken in 19§0 when the same students were tenth-graders. Results of these surveys are given in two reports (National Center for Education Statistics, 1991; Office of Educational Research and Improvement, 1991). According to these reports, "69% of parents reported monitoring their eighth-grader's television viewing, 62% limited television viewing on school nights, and 84% restricted early or late viewing" (National Center for Educational Statistics, 1992, p. 1). These statistics are not as reassuring as one would hope.

Two-thirds of the parents reported they did enforce rules limiting television viewing, while the same number of students reported their parents did not limit their television viewing. In fact, these eighth-graders spent almost 4 times as much time watching television each week as they did doing their homework (Office of Educational Research and Improvement, Fall 1991, p. 5).

Generally, research does not support the myth that children watch more television because their parents are absent. Even parents who are present rarely restrict children's viewing. The older the child, the less influence the parents have (Pearl, 1982). This pattern is disturbing in light of evidence that heavy viewers (4 hours a day or more) do less well in school and have fewer hobbies and friends (Huston et al., 1992). Gadberry (1980) did an experimental study in which parents restricted 6-year-olds to about half their normal viewing amount. When compared with a control group whose viewing was not restricted, the treatment group improved in cognitive performance and time spent on reading.

Parents who are selective viewers are more likely to encourage or restrict viewing and to watch with their children. Parents who believe television is a positive influence watch more television with children (Dorr et al., 1989). The least-effective position for parents to take is a laissez-faire one, because children whose parents neither regulate or encourage viewing watch more adult entertainment television, usually without an adult present. This puts children more at risk from the negative effects of television (Wright, St. Peters & Huston, 1990).

Lull (1990) describes the many roles television can play in family interaction. The roles are structural (time and activity cues) or relational (facilitation of either shared communication or avoidance of communication and demonstration of competence or authority). Thus, television is an important variable in how family members relate to each other. Using surveys, Bower (1988) has compared parents' use of rules for viewing in 1960, 1970, and 1980. The results indicated a trend toward an increase in the restrictions and prescriptions parents impose on viewing. This increase in rules about amount of viewing and hours for viewing was indicated for 4- to 6-year-olds and 7- to 9-year-olds. For younger children, this also included an increase in rules about changing the channel or "grazing." Bower found that the higher the educational level of parents, the more likely there were rules about viewing. This confirms the findings of Medrich et a]., who also found that the likelihood of rules increased with parental education for all households, but African-American households at every socioeconomic level were less restrictive about television viewing (Bower, 1985; Medrich, Roizen, Rubin & Buckley, 1982, cited in Comstock & Paik, 1991).

Several studies discuss the effects of new-technology, such as cable and VCRs, on parental restrictions. These studies are reviewed by Comstock and Paik (1991). Lin and Atkin (1989) found that several variables interact with rulemaking* for adolescent use of television and VCRs, including school grades, child media ownership, child age, and gender. They point out the difficulty in separating the research on rulemaking, parental mediation, and coviewing.

Within this realm of parental guidance, the relationship between mediation and rulemaking is, itself, worthy of separate consideration. Few researchers have considered mediation (e.g., encouraging, discouraging, discussing viewing) apart from the notion of rulemaking (established guidelines about acceptable and/or prohibited behaviors). Those making mediation-rulemaking distinctions (Brown & Linne, 1976; Reid, 1979; Bryce & Leichter, 1983) found a fair degree of correspondence between the two. Although these two concepts may appear as indicators of the same general process, we maintain that they should be theoretically distinguished. Actual mediation isn't necessarily contingent upon established rules. Clearly, one can have mediation without making explicit rules (and vice versa) (Lin & Atkin, 1989, p. 57).

Still Lin and Atkin found that mediation and rule making were predicted by each other.

There is also the question of whether information or training can increase parental involvement. Greenberg, Abelman, and Cohen provided television guides that reviewed programs to parents who did not use them. However, the children used them to find programs with the warning "parental discretion is advised" so that they could watch them (Greenberg, Abelman & Cohen, 1990, cited in Comstock & Paik, 1991). The jury is out, however, on whether training can help parents guide children in using television wisely. There are many books available for parents, including the Corporation for Public Broadcasting's Tips for Parents: Using Television to Help Your Child Learn (1988); the more recent American Psychological Association's (APA) "Suggestions for Parents" (Huston et al., 1992); Chen's The Smart Paren6 Guide to KIDS'TV (1994a); and the USOE Office of Educational Research and Improvement publication TV Viewing and Parental Guidance (1994).

There has been little training of parents and almost no research on the effectiveness of such training. There have been many materials for television awareness training, such as critical-viewing teaching materials, which have been evaluated formatively. These will be discussed later in this chapter.

11.5.3.5. Parental Attitude and Style. Several studies found that parents did not mediate or enforce rules about television viewing because they did not believe television was either a harinful or beneficial force (Mills & Watkins, 1982; Messaris, 1983; Messaris & KerT, 1983, cited in Sprafkin, Gadow & Abelman, 1992). There is some research that reports that a parent's positive attitude towards television is an important mediator (Brown & Linne, 1976; Bybee, Robinson & Turow, 1982; Doff, Kovaric & Doubleday, 1989). In 1991, St. Peters, Fitch, Huston, Wright, and Eakins concluded that attitudes about television were correlated with parents' regulation and encouragement of viewing. The next year, they reported that parents' negative attitudes about television were not sufficient to modify the effects of television viewing. To reach their conclusions, the researchers collected data from 326 children and their families through diaries, questionnaires, standardized - instruments, and one-way miffor experiments. This research led to a finer delineation of the variable "parental attitude" toward television:

Positive attitudes were positively associated with parents' encouragement of viewing certain types of programs. Negative attitudes were positively related to regulating children's television viewing. Those parents who both regulated and encouraged discriminating viewing had children who viewed less television than parents who were high on encouragement of viewing. However, the present analysis shows that while parents appear to criticize and regulate television's content because of its negative influence and coview violent programming (news and cartoons) with their children, parents may not be taking advantage of the opportunity to discuss the programs they watch with their children and moderate the effects of content either directly or indirectly. Parents' education and attitudes about television were not associated with children's social behavior towards others (St. Peters, Huston & Wright, 1989, p. 12).

Abelman found that parents who were more concerned with cognitive effects were more likely to discuss and criticize television content, whereas parents who were more concerned about behavioral effects were more likely to mediate by restricting viewing (Abelman, 1990, cited in Sprafkin, Gadow & Abelman, 1992). Earlier, Abelman and Rogers (1987) presented findings that compared the television mediation of parents of exceptional. children. Parents of nonlabeled (no disability* identified) children were restrictive in style; parents of gifted children were evaluative in style; and parents of emotionally disturbed, learningdisabled, or mentally retarded children were unfocused in style. The actions of parents with restrictive styles included forbidding certain programs, restricting viewing, specifying viewing time, specifying programs to watch, and switching channels on objectionable programs. Parents with an evaluative style explained programs and advertising, evaluated character roles, and discussed character motivations and plot/story lines. Parents with an unfocused style were characterized by one or two of these actions: (a) coviewed with the child, (b) encouraged the use of a television guide, (c) used television as reward or punishment, and (d) talked about characters (Abelman & Rogers, 1987).

Singer and Singer and their colleagues have studied parental communication style as it interacts with television viewing and affects comprehension of television (Desmond et aL, 1985, 1990; Singer, Singer & Rapaczynski, 1985, cited in Sprafkin et al., 1992). In a summary of these research findings, Desmond et al. (1990) suggest that "general family communication style may have been more critical than specific television rules and discipline for enhancing a range of cognitive skills, including television comprehension" (p. 302). Children are helped by an atmosphere that promotes explanation about issues instead of just comments on people and events. Similarly, Korzenny et al. conducted a study at Michigan State University to deterniine under what conditions children's modeling of antisocial portrayals on television was strongest. They found, that parents who disciplined by reasoning and explanation had children who were less affected by antisocial content than children whose parents disciplined through power (Korzenny et al., 1979, cited in Sprafkin et al., 1992).

11. 5.4 Coviewing as a Variable

Coviewing refers to viewing in a group of two or more, such as a child and parent or three adolescent peers. Since discussion has been shown in many studies to be an important variable in learning from television (Buerkel-Rothfuss, Greenberg, Atkin & Neuendorf, 1982, cited in Comstock & Paik, 1991; Desmond, Singer & Singer, 1990), one would expect coviewing to be a significant variable in the homeviewing context. Unfortunately, studies suggest that although coviewing is an important variable, there are few effects due to coviewing. The reasons for this conclusion will be explained in this section. Three categories will be discussed: the nature and frequency of interaction, the effect of attitudes, and the effects of age and roles.

11-5.4.1. Nature and Frequency of Interaction. Based on a review of several articles, Comstock and Paik (1991) speculate that the time adolescents and adults spend coviewing is declining. The greatest concern in the literature is that most parents don't spend time coviewing, and when parents do coview, their level of involvement is usually low. It is not just the amount of time spent coviewing; the type of interaction during coviewing is critical. Most conversation during coviewing is about the television medium itself, the plots, characters, and quality of programs (Neuman, 1982, cited in Comstock & Paik, 1991). These conversations help educate young viewers and make them more critical. According to Comstock and Paik, however, they are not as crucial as conversations that deal with the reality of the program or the rightness or wrongness of the behavior portrayed.

The evidence suggests that parental mediation-when it employs critical discussions and interpretations of what is depicted and sets some guidelines on television use-can increase the understanding of television, improve judgments about reality and fantasy, and reduce total viewing (Comstock & Paik, 199 1, p. 45).

Nevertheless, parental coviewing is not always a positive influence. Parents can give implicit approval to violence, prejudice, or dangerous behavior (Desmond, Singer & Singer, 1990, cited in Comstock & Paik, 1991). After surveying 400 second-, sixth-, and tenth-graders, Dorr, Kovaric, and Doubleday (1989, cited in Comstock & Paik, 1991) found that coviewing basically reflected habits and preferences, rather than parental mediation or conversational involve ment. In 1989, Dorr et aL reported only weak evidence for positive consequences from coviewing. They concluded that coviewing is an imperfect indicator of parental mediation of children's viewing. In their review, they identify several methodological problems that make it difficult to use the literature, including differing definitions of coview ing, overestimates by parents, and the assumption that coviewing is motivated by parents' desire to b e responsible mediators of childrens' interactions with television. They report that coviewing with young children is infrequent (Hopkins & Mullins, 1985, cited in Dorr et al., 1989). Moreover, several studies have found that parent-child coviewing decreases as the number of sets in the house increase (Lull, 1982; McDonald, 1986, cited in Doff et al., 1989). Dorr and her colleagues investigated several hypothe ses about coviewing using data from seven paper-and-pencil instruments given to both parents and children. Their subjects included 460 middle-class second-, sixth-, and tenth-grade children and one parent for each of 372 of these children. The results indicated that coviewing by itself had little relationship to children's judgment of reality. It did predict satisfaction with farnily viewing.

Thus, research shows that most coviewing takes place because parents and children have similar viewing interests and tastes. Little of the coviewing has been planned by the parent to aid with the child's understanding and comprehension of the show (MacDonald, 1985, 1986, cited in Doff et al., 1989; Wand, 1968). Nevertheless, it is possible that coviewing may help parents deal with difficult issues. Through viewing scenarios on television, the child may discuss the television character's dilemma with a parent, or the child may simply accept the television portrayal as the appropriate solution.

11.5.4.2. Effect of Attitudes. Dorr and her colleagues also found that parental attitudes toward television were predictors of coviewing. Parents who were more positive coviewed with children more frequently. Coviewing also correlated moderately with parents' belief that children can learn from television and with parents' encouragement of viewing. They concluded that it has a greater effect when motivated by parents' determination to mediate television experiences. This is an important finding, because coviewing occurs least with those who need it most, young children. Children are willing to discuss television content with their parents. Gantz and Weaver (1984) found that children initiate discussions of what they view with their parents; however, children did not initiate discussions about programs unless the programs were coviewed.

11.5.4.3. Effect of Age and Roles. Coviewing is usually described in terms of whether the viewers are children, adolescents, or adults, and whether the social.group is of mixed age or not. The usual roles referred to are siblings, peers, and parents.

Haefher and Wartella (1987) used an experimental design to test hypotheses about coviewing with siblings. By analyzing verbal interactions in coviewing situations, they determined that relatively little of the interaction helped younger children interpret the content. Some teaching by older siblings did occur but was limited to identifying characters, objects, words, and filmic conventions. The result was that older siblings influenced evaluation of characters and programs in general, rather than interpretation of content. Haefiter and Wartella (1987) noted that other variables needed to be accounted for, such as gender, birth order, viewing style, and attitude, because they could affect differences in learning from siblings. Pinon, Huston, and Wright (1989) conducted a longitudinal study of family viewing of Sesame Street using interviews, testing sessions, and diaries with 326 children from ages 3 to 5 and 5 to 7. The presence of older children wai found to reduce viewing, the presence of younger children to increase it. Alexander, Ryan, and Munoz (1984, cited in Pinon et a]., 1989) found that younger children imitated the preferences of older children and that coviewing with older siblings promoted elaboration of program elements.

Salomon (1977) conducted an experimental study on mothers who coviewed Sesame Street with their 5-yearolds. He found:

Hill and Stafford (1980) investigated the effect of working on the time mothers devote to activities such as childcare, leisure television viewing, and housework. The addition of one child increased the time devoted to housework by 6 to 7 hours a week. Mothers who worked took this time from personal care time, including sleep and television watching. Because early childhood may be an important time for the establishment of long-term patterns of television use, it becomes essential that parental patterns of viewing continue to include coviewing with children, even when family routine mandates changes.

Collett (1986) used a recording device to study coviewing. The device, a C-Box,* consisted of a television set and video camera that recorded the viewing area in front of the television. In addition, subjects were asked to complete a diary. He points out that:

It is a sad fact that almost everything we know about television has come from asking people questions about their viewing habits and opinions, or from running them through experiments. The problem with asking people questions is that they may not be able to describe their actions reliably, or they may choose to offer accounts which they deem to be acceptable to the investigator (p. 9).

In 1988, Anderson and Collins examined the research literature on the relationship between coviewing by parents and critical-viewing skills programs, school achievement, and learning outcomes. The review concluded that there was little support for most of the beliefs about the negative influence of television on children. This opinion contrasts to some extent with conclusions of Haeftier and Wartella (1987) and Winn (1977). Anderson and Collins concluded that adults can be helpful to children's comprehension through coviewing, but that it is not clear that interactions are common.

11.5.5 Viewing Habits

Another factor in the family viewing context is the viewing habits or patterns of the household. Because television viewing is often a social as well as a personal act, viewing habits both effect and are affected by other fairdly variables. The factors that seem to emerge from research on viewing habits are the amount of viewing, viewing patterns, and audience involvement.

11.5.5.1. The Amount of Viewing. So far research related to this variable centers around the effects of heavy viewing. Estimates for the typical number of hours television is watched in the American home each day vary from 7 hours (Who are the biggest couch potatoes?, 1993) to 21 hours (Would you give up TV for a million bucks?, 1992). Those over age 55 watched the most; teenage girls, who averaged 3 hours a day, watched the least (Who are the biggest couch potatoes?, 1993). If heavy viewing is defined as more than 3 to 4 hours a day, many Americans are heavy viewers, which makes it difficult to research and draw conclusions about heavy viewing. Research does indicate that heavy viewing is associated with more negative feelings about life. Adults who watch television 3 or more hours daily are twice as likely to have high cholesterol levels as those who watch less than an hour daily, according to Larry Tucker, director of health promotion at Brigham Young University, who examined the viewing habits of 12,000 adults.

Children who are heavy viewers often have parents who are heavy viewers. Such parents are usually less educated and enforce fewer family rules about appropriate programs (Roderick & Jackson, 1985). 71be amount of viewing changes over a life span. Teenagers are relatively light viewers when compared with children and adults (Comstock & Paik, 1987). Some studies reported that children of mothers who work outside the home watch no more or less television than children of mothers at home (Webster, Pearson & Webster, 1986; Brown, Childers, Bauman & Koch, 1990); yet Atkin, Greenberg, and Baldwin (1991) summarized research that concluded that children view more in homes where the father is absent (Brown, Bauman, Lenz & Koch, 1987, cited in Atkin, Greenberg & Baldwin, 1991) and where the mother works (Medrich, Rozien, Rubin & Buckley, 1982, cited in Atkin, Greenberg & Baldwin, 199 1).

Using a questionnaire, Roderick and Jackson (1985) identified differences in television viewing habits between gifted and nongifted viewers. More nongifted students were found to have their own television sets, which may account for the heavier viewing habits of nongifted students. Gifted students preferred different programs (educational, documentaries) from nongifted students (sitcoms, soaps, game shows). Gifted students were more likely to have VCRs in their home. They did not engage in the wishful thinking or fantasizing about television characters that was common with nongifted students. Roderick and Jackson had nongifted students respond in their classrooms and gifted students respond at home, which may have introduced bias.

The CPB participated in the 1993 Yankelovich Youth Monitor in order to answer some questions about viewing patterns in the 1990s (Corporation for Public Broadcasting, 1993). The Youth Monitor survey studied 1,200 children ages 6 to 17 with an in-home interview in randomly selected households. Today 50% of children have a television set in their bedroom. They watch 3 hours per weekday and 4 hours per weekend day. Less than 20% watch an hour or less per day. Viewing decreases as income increases. AfricanAmerican and Hispanic children view the most. Television viewing is the number-I activity in the hours between school and dinner time. Nearly half the children reported viewing television with their family each evening. This is especially true for children who watch public television.

11.5.5.2. Viewing Patterns. "Viewing patterns" refers to content preferences, but content does not dictate viewing, because, with few exceptions, other variables have more effect on preferences. This concept can be misleading, because, although there are few discernable patterns of preferences by program types, viewers would be unlikely to watch test patterns or the scrolling of stock market reports. Research supports the conclusion that viewers are relatively content indifferent.* Huston, Wright, Rice, Kerkman, and St. Peters (1990) conducted a longitudinal investigation of the development of television viewing patterns in early childhood, focusing on types and amounts of viewing from ages 3 to 7. They were interested in developmental changes resulting from maturation or cognitive development, individual and environmental variables affecting viewing patterns, and the stability of individual differences in viewing patterris over time. Viewing was measured from diaries kept by parents who were instructed to record as a viewer anyone who was present for more than one-half of a 15-minute interval when the television was on. While there were many individual differences, these differences tended to be stable over time. As they grew older, children watched programs that required more cognition, such as programs with less redundancy and increasing complexity. Nevertheless, the researchers concluded that family patterns and external variables are more important determinants of viewing than individual or developmental differences. They also found that boys watched more cartoons, action-adventure, and sports programs than did girls. Boys watched more television overall. Viewers of humorous children's programs evolve into viewers of comedy at a later age. Viewers of adventure stories become viewers of action-adventure by age 7. In comparison to this study, Lyle and Hoffman (1972, cited in Comstock & Paik, 199 1) documented through questionnaires that preferences change with age.

Plomin, Corley, DeFries, and Fulker (1990) conducted a longitudinal study of 220 adopted children from age 3 to 5. Evidence for both significant genetic and environmental influences on television viewing patterns was found. Neither intelligence nor temperament was responsible for this genetic influence.

McDonald and Glynn (1986) examined adult opinion about how appropriate it is for children to view certain kinds of content. Telephone interviews were conducted with 285 respondents. Adults did not approve of crimedetective and adult-oriented programming for children.

Over 4 years, Frank and Greenberg (1979) conducted personal interviews with 2,476 people aged 13 years or older. They found support for their thesis that viewing audiences are more diverse than usually assumed. From the information collected, they constructed profiles of 14 segments of the television audience. Their study is an example of research that clusters variables. More of such research is needed, because so many variables interact in the television environment.

11.5.5.3. Audience Involvement. Research shows that selectivity and viewing motives can affect viewing involvement (Perse, 1990). Using factor analysis techniques with data generated from questionnaires, Perse investigated viewing motives classified as ritualistic* (watching for gratification) or instrumental* (watching for information). The study included four indications of audience involvement: (a) intentionality, or anticipating television viewing; (b) attention, or focused cognitive effort; (c) elaboration, or thinking about program content; and (d) engaging in distractions while viewing. Ritualistic television use, which indicates watching a broad variety of programs, is marked by higher selectivity before watching but lower levels of involvement while viewing. The study confirms the value of the Levy-Windahl Audience-Activity Typology (Levy & Windahl, 1985, cited in Perse, 1990).

The Experience Sampling Method* was used to study media habits and experiences of 483 subjects aged 9 to 15 years (Kubey & Larson, 1990). Respondents carried electronic paging devices, and whenever contacted, they reported on their activities and subjective experiences. The utilization of three new forms of video entertainment (music videos, video games, and videocassettes) and traditional television was subsequently analyzed. Traditional television viewing remains the dominant video media form for preadolescents and adolescents. New video media are a relatively small part of their lives. However, the percentage of time spent alone with the new media is growing, perhaps because they offer chances for adolescents to be more independent of the family. Boys had more positive attitudes towards the new media. There could be many reasons for this, including gender differences or the content of the new media.

11.5.6 Summary and Recommendations

In 1978, Wright, Atkins, and Huston-Stein listed some characteristics of the setting in which a child views television:

• Presence of others who are better informed or who can answer questions raised by a child
• Behavior of others, who through well-timed comments and questions model elaboration of content
• Preparation of the child through previous reading, viewing, or discussion
• Opportunity to enact or rehearse, role play plots, characters, and situations viewed
• Distractions in the environment

Much is known today about each of these aspects of the family viewing context. In addition, new variables and interactions have been identified such as rulemaking, parental communication style, socioeconomic level, and ethnicity.

Nevertheless, many gaps exist in the research literature, especially about interactions. The well-supported conclu sion that learning from television increases when an adult intervenes to guide and support learning even if the program is. an entertainment (Johnston, 1987) suggests that much more needs to be done to relate the findings of mass-media research and research from instructional television and message design. Therefore, it is essential to relate findings about learning from television with findings about the family context for viewing* in order to design interventions that will ensure the positive benefits of television. Findings need to be related theoretically in order to develop recommendations for interventions.

St. Peters et al. (1991) summarize the situation:

Whatever the effects of parental coviewing, encouragement, and regulation, it is clear that the family context is central to the socialization of young children's television use. Families determine not only the amount of television available to children, but the types of programs, and the quality of the viewing experience (p. 1422).