Social Goals and the Media

views updated


Media effects researchers have tended to focus on negative rather than positive effects of watching television. However, given that the same processes of observation, learning, and imitation should be at work for both types of effects, it is plausible that there should be prosocial as well as antisocial outcomes of television exposure.

Prosocial Content on Television

During the 1970s, prosocial behaviors were reported to appear quite frequently on television. However, these behaviors typically occurred in a context of violence and hostility. As Bradley Greenberg and his associates reported in 1980, the favorite programs of a sample of grade-school children contained equal numbers of prosocial and antisocial acts. Marsha Liss and Lauri Reinhard (1980) found that even those cartoons that were considered by the researchers to have moral messages contained high levels of violence—the same amount as in standard cartoons that had no such moral lessons. Moreover, only some types of prosocial behaviors were shown. Rita Poulos, Eli Rubinstein, and Robert Liebert reported in 1975 that although there were an average of eleven altruistic acts and six sympathetic acts per hour, there were very few depictions of self-control (e.g., controlling aggressive impulses or resisting temptation).

Content analyses of prosocial behavior on television were scarce throughout the 1980s, but prosocial content received more attention again in the 1990s with the introduction of legislation aimed at improving the quality of television programming for children. The Federal Communications Commission (FCC) enacted a processing guideline known as the "three-hour rule" that went into effect in 1997. Under the three-hour rule, broadcasters that wish to have their license renewals expedited are required to air a minimum of three hours a week of educational and informational (E/I) television that meets the "cognitive/intellectual or social/emotional" needs of children. The E/I programs must be specifically designed for children who are sixteen years of age or under and must air between the hours of 7:00 A.M. AND 10:00 P.M. Broadcasters are required to place an on-air symbol at the beginning of E/I programs to indicate to the public that they are educational. This information must also be provided to listing services, such as the local newspaper and TV Guide.

In the late 1990s, the Annenberg Public Policy Center conducted a series of content analyses of all children's programs that were aired over the course of a composite week in Philadelphia, a large urban media market. Emory Woodard (1999) examined the frequency with which programs contained social lessons about how to live with oneself (i.e., intrapersonal skills such as understanding emotions, maintaining self-esteem, and overcoming fears) and how to live with others (i.e., interpersonal skills such as acceptance of diversity, altruism, and cooperation). In Woodard's sample, 50 percent of all children's shows contained at least one social lesson. These were mostly concentrated in programming for preschool children; 77 percent of preschool children's programming contained a social lesson. Public Broadcasting Service (PBS) programming had the highest overall rate with 72 percent of the children's programming containing social lessons. This was followed by premium cable channels, such as The Disney Channel and Home Box Office (HBO), with 59 percent of the children's programs containing social lessons. PBS also had the highest level of programs with traditionally academic lessons (i.e., reading, writing, and arithmetic), since 89 percent of the programs fell into this category.

Using the same sample of programs, Kelly Schmitt (1999) reported that of the subset of children's programs that had been designated as meeting the "educational and/or informational needs of children," 75 percent of those offered by the commercial broadcast networks were prosocial in nature. In this case, "prosocial" was broadly defined as promoting "learning to live with oneself and others." That is, broadcasters appeared to be meeting the three-hour rule by focusing on general prosocial messages rather than conveying traditionally academic information.

Although these content analyses suggest that children's television is full of material that has the broad social goal of teaching life skills, children may actually see relatively little of this potentially positive content. Woodard (1999) examined the content of the twenty shows in the sample that received the highest Nielsen ratings among children between two and seventeen years of age. Only four of the twenty contained social lessons in the episodes that were analyzed. Of those four, only two contained content that was related to specific prosocial outcomes such as positive social interactions, stereotype reduction, or altruism. Of these two, only one was designed explicitly for children. These findings highlight the fact that it is relatively rare for children to watch the prosocial programming that is designed for them.

Evidence of Prosocial Effects?

Marie-Louise Mares and Woodard (2000) conducted a meta-analysis to summarize the results of thirty-nine studies of prosocial effects of television. Meta-analysis involves averaging statistical information across studies on a particular topic, in order to estimate the overall consistency and strength of effects. Mares and Woodard reported that prosocial content had an overall weak to moderate effect. Effects were strongest for studies of altruism, largely because such studies were more likely to model behaviors that were identical to the behaviors that were subsequently observed in the children. Efforts to promote other prosocial behaviors such as positive interaction, aggression reduction, and stereotype reduction were less likely to use identical treatment and outcome measures. Effect sizes for these treatments were smaller and remarkably similar to each other.

How do these effect sizes compare with the effects of violent content? Hae-Jung Paik and George Comstock (1994) looked at studies of the effects of television violence on viewer aggression. They reported an overall moderate negative effect of violent television content. Therefore, it is reasonable to suggest at this point that the effects of violent and prosocial content are reasonably close in magnitude.

Investigating Prosocial Effects

Research strategies for investigating television effects have evolved over the years, which is true of research on prosocial effects as well as in other areas. Early studies of prosocial effects were often simple one-shot experimental tests of modeling. These simple tests of modeling generally found quite strong, positive effects when comparing a group that watched explicit depictions of prosocial actions with a group that did not see the content.

A second strategy was to conduct a field experiment that typically involved repeated exposure to real television content in relatively uncontrolled environments. The major field experiments in this area looked at the effect of using a particular prosocial program such as Sesame Street, Mister Rogers' Neighborhood, or Barney & Friends as part of the school or preschool experience. Children generally watched whole episodes every week for a number of weeks and were then evaluated, often by observation over several days or weeks, rather than by assessing their performance on a single task. These studies often found that prosocial content could be effective, but chiefly when the content was combined with other forms of teacher intervention.

A third strategy was to conduct a survey to find out how much of an effect could be observed when children simply self-selected to watch prosocial programming at home. That is, these studies examined the effects of everyday viewing rather than special interventions. The results of such correlational studies are more easily and reliably applied to general audiences than are the other types of studies. However, as in all correlational studies, the question of whether prosocial programming causes prosocial outcomes is plagued by issues of causal direction and spuriousness. Maybe children who are already tolerant, friendly, caring people are attracted to prosocial programming. Maybe prosocial behavior and prosocial viewing are both caused by other variables such as parental style, gender, and so on. Overall, once these possible third variables are statistically controlled, most correlational studies find very weak effects of prosocial viewing.

Effects on Positive Social Interactions

In one example of a field experiment, Lynette Friedrich-Cofer and her associates (1979) focused on the effects of Mister Rogers' Neighborhood on "urban poor children." They had children who were in Headstart programs watch twenty episodes of the program over a period of eight weeks. Comparisons were then made between children from four different groups: (1) those who simply watched Mister Rogers' Neighborhood at the Headstart center without any additional prosocial materials, (2) those who watched it and had access to prosocial books, games, and so on, (3) those who watched it, had access to the prosocial materials, and had follow-up activities such as verbal labeling of the prosocial behaviors and role playing, and (4) those who were part of a control group that simply saw neutral films.

Friedrich-Cofer and her associates found that Mister Rogers' Neighborhood alone produced relatively few behavioral changes. Children who watched the program and had access to the prosocial materials became more active overall—they had a greater number of positive interactions, but they also had more aggressive interactions. The most successful group, in terms of prosocial behavior, was the one that watched the program and received training in role playing and verbal labeling. That group showed significant increases in positive social interactions without any increases in aggression.

Jerome and Dorothy Singer (1998) examined the effects of repeated exposure to Barney & Friends in preschool and daycare settings, and they found similar results. There were minimal effects of exposure when children simply watched the program without further adult elaboration of the content, but there were significant positive effects when viewing was combined with adult commentary and related activities.

Maurice Elias (1983) examined whether videos could be used as one component of treatment for boys who had serious emotional and educational disturbances. The ten videos, which were shown twice a week for five weeks, portrayed realistic scenarios of common problematic situations such as teasing and bullying, dealing with peer pressure, learning how to express feelings, and coping with new social situations. After watching each video, the boys were encouraged to discuss what they had seen and how they felt about it. The boys were measured (both for three months before and two months after the video series) on a variety of behavioral and emotional responses that were related to interactions. Compared to control children who did not see the videos, participants in the experiment were rated by their counselors as being less emotionally detached and less isolated, as having improved in their ability to delay gratification, and as having decreased in overall personality problems. These effects were still evident two months after the intervention.

Despite these encouraging results, researchers generally find much weaker outcomes when they use surveys to look at the effects of normal, everyday viewing. Joyce Sprafkin and Eli Rubinstein (1979) studied children who were seven, eight, and nine years of age and lived in middle-class communities. The children reported how often they watched each of fifty-five television series that were then rated for levels of prosocial and antisocial content. The children's prosocial behavior was measured by teacher and classroom peer reports.

Sprafkin and Rubinstein found that the strongest predictors of prosocial behavior were background variables. Children who were high academic achievers or whose parents were well educated received more reports of prosocial behavior. Girls were also rated as being more prosocial than boys. Compared to these effects, television viewing was only weakly related to prosocial behavior. The partial correlation between prosocial viewing and behavior (controlling for background variables) reflects very minimal differences between heavy viewers of prosocial content and light viewers of prosocial content.

Oene Wiegman, Margot Kuttschreuter, and Ben Baarda (1992) studied second-and third-grade children in The Netherlands for three years (until the children were in fifth and sixth grade). Children were measured once a year on a number of variables, including exposure to prosocial television content and levels of prosocial behavior. As in the Sprafkin and Rubinstein (1979) study, prosocial and antisocial exposure were measured by the frequency of viewing specific programs, and prosocial behavior was assessed by peer nominations. Wiegman and his colleagues found no relationship between prosocial viewing and prosocial behavior, despite the considerable power granted by their large sample size (i.e., 466 children). If anything, the relationship tended to be very weakly negative, rather than positive. Why was this the case? The researchers noted that watching prosocial content was very highly correlated with watching antisocial content—children who saw the most prosocial content were simply heavy television viewers who were exposed to numerous antisocial models as well.

Effects on Altruism

In a well-known study of altruism (i.e., generosity), Poulos, Rubinstein, and Liebert (1975) randomly assigned first-grade children to one of three viewing conditions: (1) a prosocial episode of Lassie, in which the protagonist, Jeff, risked his life to save a puppy, (2) a neutral episode of Lassie, or (3) a neutral episode of The Brady Bunch. After viewing the episode, the children were told how to play a "game" in which they could accrue points by pressing a button. The more points they earned, the larger the prize they would win. At the same time, they were asked to listen to puppies in a distant kennel and to push a help button to call the researcher if the puppies seemed distressed. As children played the game, the recorded puppy sounds grew increasingly loud and intense. The researchers compared the average number of seconds children spent pushing the help button (and thereby sacrificing points in the game) in each of the conditions. Children who saw the prosocial episode pushed the help button nearly twice as long as children in the other two conditions.

In a more typical study of altruism, conducted by James Bryan and Nancy Walbek (1970), children were brought into the laboratory and told that they would learn how to play a new game by watching a video. The children watched one of several versions of the video, in which the model played the game, was rewarded by tokens that could be used to win a prize, and then immediately behaved either altruistically (giving some of the tokens to charity or to another child) or selfishly (cashing in all the tokens for a big prize). The children then played the game, won a fixed number of tokens, and were given the opportunity to donate some tokens. Children who saw the altruistic model donated more tokens than those who had seen the selfish model. As in this example, studies of altruism have generally involved explicit modeling of very specific behaviors immediately after observing the model.

Effects on Tolerance and Stereotype Reduction

In one of the largest and most impressive studies of stereotype reduction, Jerome Johnston and James Ettema (1982) conducted a field experiment that involved more than seven thousand children. Fourth-to sixth-grade classrooms in seven sites across the United States were randomly assigned to watch twenty-six episodes of Freestyle, a public television program that was designed to reduce stereotypes about gender roles. Children were assigned (1) to watch the program at school and to engage in teacher-led discussions about the material, (2) to watch the program at school without any such discussions, or (3) to watch the program at home. The students completed extensive questionnaires before seeing any episodes and again after the exposure period. Compared to a control group that had not viewed the program, there were significant positive changes in the students' perceptions of personal ability and their interest in various types of jobs, and there were reductions in the stereotypes that students had about gender roles in employment. These effects were strongest when the program was viewed in the classroom and accompanied by teacher-led discussions. Much smaller effects were observed among students who simply watched the program in school or at home.

Two studies of attempts to counteract gender stereotypes suggest that the same content may have positive effects on some groups but cause a backlash in other groups. Suzanne Pingree (1978) showed commercials to children who were in the third and eighth grades. The commercials featured women in either traditional or nontraditional roles, and Pingree found that, among most children, stereotyping was reduced when the children viewed the commercials that showed the nontraditional behaviors. This result was particularly strong when children were told that the advertisements depicted real people. However, among eighth-grade boys, there appeared to be something of a contrary reaction because stereotyping was significantly higher for this group when commercials featured the nontraditional condition than when they featured the traditional condition.

Shirley O'Bryant and Charles Corder-Bolz (1978), over a period of one month, showed nine half-hour cartoons to children who were between five to ten years of age. Embedded in each cartoon were commercials for a fruit-juice drink. In the traditional condition, the commercials featured a female telephone operator, fashion model, file clerk, and manicurist. In the nontraditional condition, the commercials featured a female pharmacist, welder, butcher, and laborer. Over the course of the month, children saw many repetitions of the commercials. Comparisons between pre-and post-test scores for occupational stereotyping found that children who were exposed to the non-traditional commercials were significantly more likely to say that a traditionally male job was also appropriate for a woman. Moreover, girls who viewed the nontraditional condition gave higher ratings to traditionally male jobs when asked how much they would like to have that job in the future. Boys' ratings of future interest in traditionally male jobs were lower for those who had viewed the nontraditional condition—apparently seeing women in those roles was a deterrent. This effect is consistent with the finding by Pingree (1978) of a "backlash" among boys, and it underlines the point that seemingly prosocial content can have unintended effects on certain subgroups.

Gerald Gorn, Marvin Goldberg, and Rabindra Kanungo (1976) assessed the effects of Sesame Street on children's tolerance for playmates of different ethnic and racial backgrounds. They assigned white, English-Canadian children who were 3.5 to 5.5 years of age to see twelve minutes of Sesame Street programming—either with multi-cultural inserts or without the inserts. The children were then shown two sets of four photographs that were taken from the inserts. One set featured Caucasian children, and the other set featured children from other ethnic and racial backgrounds. The participants in the study then chose which of the photographed children should be brought to the nursery school the next day. The control group that did not see the inserts showed a marked preference for playing with the Caucasian children (67%). Among children who saw the multiracial inserts, this was reversed, with a marked preference for the non-Caucasian playmates (71%) over the Caucasian playmates (29%). However, in one of the very few prosocial projects that involved delayed testing, Goldberg and Gorn (1979) expanded their earlier study and found that children who were tested a day after viewing the multicultural inserts were no longer significantly more willing to play with non-Caucasian playmates than those who had not seen the inserts.

Early studies of Sesame Street confirmed that the message of tolerance in Sesame Street took time to extract. Gerry Ann Bogatz and Samuel Ball (1971) conducted longitudinal studies of children who were exposed to Sesame Street. They reported that viewing Sesame Street was positively related to tolerant racial attitudes, but only after two years of exposure. Measures at the end of the first year had found no such relationship.

Mixing Prosocial and Antisocial Content

A final point worth noting is the particularly harmful effect of combining prosocial and antisocial content. As Wiegman, Kuttschreuter, and Baarda (1992) pointed out, not only do heavy prosocial viewers also tend to be heavy viewers of violence, but many of the prosocial acts shown on television are actually presented in the context of violence, as when a "good" group of people fights a "bad" group. When Mares and Woodard (2000) conducted their meta-analysis, they found stronger negative effects of aggressive-prosocial content than of aggressive content that was unadulterated by any prosocial themes. Perhaps having antisocial acts within the context of positive behaviors actually lends legitimacy to the antisocial acts.

The research suggests that prosocial content can have positive effects that are as strong as the negative effects of antisocial content. However, prosocial content may require repetition as well as related adult commentary and activities for long-term positive effects to occur. Moreover, prosocial content may have different effects on different populations, occasionally causing a backlash among groups that feel threatened by themes such as gender equality.

See also:Advertising Effects; Arousal Processes and Media Effects; Children and Advertising; Children's Attention to Television; Children's Comprehension of Television; Children's Preferences for Media Content; Gender and the Media; Interpersonal Communication; Parental Mediation of Media Effects; Public Broadcasting; Public Health Campaigns; Sesame Street; Social Change and the Media; Society and the Media; Television, Educational; Violence in the Media, Attraction to; Violence in the Media, History of Research on.


Bogatz, Gerry Ann, and Ball, Samuel. (1971). The Second Year of Sesame Street: A Continuing Evaluation. Princeton, NJ: Educational Testing Service.

Bryan, James H., and Walbek, Nancy H. (1970). "The Impact of Words and Deeds Concerning Altruism Upon Children." Child Development 41:488-500.

Elias, Maurice J. (1983). "Improving Coping Skills of Emotionally Disturbed Boys Through Television-Based Social Problem Solving." American Journal of Orthopsychiatry 53:61-72.

Friedrich-Cofer, Lynette K.; Huston-Stein, Aletha; Kipnis, Dorothy M.; Susman, Elizabeth J.; and Clewett, Ann S. (1979). "Environmental Enhancement of Prosocial Television Content: Effect on Interpersonal Behavior, Imaginative Play, and Self-Regulation in a Natural Setting." Developmental Psychology 15:637-646.

Goldberg, Marvin E., and Gorn, Gerald J. (1979). "Television's Impact on Preferences of Non-White Playmates: Canadian 'Sesame Street' Inserts." Journal of Broadcasting 23:27-32.

Gorn, Gerald J.; Goldberg, Marvin E.; and Kanungo, Rabindra N. (1976). "The Role of Educational Television in Changing the Intergroup Attitudes of Children." Child Development 42:277-280.

Greenberg, Bradley; Atkin, Charles K.; Edison, Nadyne G.; and Korzenny, Felipe. (1980). "Antisocial and Prosocial Behaviors on Television." In Life on Television: Content Analysis of U.S. TV Drama, ed. Bradley S. Greenberg. Norwood, NJ: Ablex.

Johnston, Jerome, and Ettema, James S. (1982). Positive Images: Breaking Stereotypes with Children's Television. Beverly Hills, CA: Sage Publications.

Liss, Marsha B., and Reinhardt, Lauri C. (1980). "Aggression on Prosocial Television Programs." Psychological Reports 46:1065-1066.

Mares, Marie-Louise, and Woodard, Emory H. (2000). "Prosocial Effects on Children's Social Interactions." In Handbook of Children and the Media, eds. Dorothy G. Singer and Jerome L. Singer. Thousand Oaks, CA: Sage Publications.

O'Bryant, Shirley L., and Corder-Bolz, Charles R. (1978). "The Effects of Television on Children's Stereotyping of Women's Work Roles." Journal of Vocational Behavior 12:233-244.

Paik, Hae-Jung, and Comstock, George. (1994). "The Effects of Television Violence on Antisocial Behavior: A Meta-Analysis." Communication Research 21:516-546.

Pingree, Suzanne. (1978). "The Effects of Nonsexist Television Commercials and Perceptions of Reality on Children's Attitudes about Women." Psychology of Women Quarterly 2:262-277.

Poulos, Rita W.; Rubinstein, Eli A.; and Liebert, Robert M. (1975). "Positive Social Learning." Journal of Communication 25:90-97.

Schmitt, Kelly L. (1999). The Three Hour Rule: Is it Living up to Expectations? (Report No. 30). Philadelphia: Annenberg Public Policy Center, University of Pennsylvania.

Singer, Jerome L., and Singer, Dorothy G. (1998). "Barney & Friends as Entertainment and Education." In Research Paradigms, Television, and Social Behavior, eds. Joy Keiko Asamen and Gordon L. Berry. Thousand Oaks, CA: Sage Publications.

Sprafkin, Joyce N.; Liebert, Robert M.; and Poulos, Rita W. (1975). "Effects of a Prosocial Televised Example on Children's Helping." Journal of Experimental Child Psychology 20:119-126.

Sprafkin, Joyce N., and Rubinstein, Eli A. (1979). "Children's Television Viewing Habits and Prosocial Behavior: A Field Correlational Study." Journal of Broadcasting 23:265-276.

Wiegman, Oene; Kuttschreuter, Margot; and Baarda, Ben. (1992). "A Longitudinal Study of the Effects of Television Viewing on Aggressive and Prosocial Behaviors." British Journal of Social Psychology 31:147-164.

Woodard, Emory H. (1999). The 1999 State of Children's Television Report: Programming for Children Over Broadcast and Cable Television. Report No. 28. Philadelphia: Annenberg Public Policy Center, University of Pennsylvania.

Marie-Louise Mares