DB 7 Schools, Peers, and Media Influences on Childhood Behavior and Beliefs

Meanonscreen.pdf

Home >Psychology homework help >DB 7 Schools, Peers, and Media Influences on Childhood Behavior and Beliefs

Journal of Communication ISSN 0021-9916

ORIGINAL ARTICLE

Mean on the Screen: Social Aggression in Programs Popular With Children Nicole Martins1 & Barbara J. Wilson2

1 Department of Telecommunications, Indiana University, Bloomington, IN 47405, USA 2 Department of Communication, University of Illinois at Urbana-Champaign, Urbana, IL, USA

A content analysis was conducted to examine the portrayal of social aggression in the 50 most popular television programs among 2- to 11-year-old children. Results revealed that 92% of the programs in the sample contained some social aggression. On average, there were 14 different incidents of social aggression per hour in these shows, or one every 4 minutes. Compared to the portrayals of physical aggression, social aggression was more likely to be enacted by an attractive perpetrator, to be featured in a humorous context, and neither rewarded or punished. In these ways, social aggression on television poses more of a risk for imitation and learning than do portrayals of physical aggression. Findings are discussed in terms of social cognitive theory.

doi:10.1111/j.1460-2466.2011.01599.x

The popular movie Mean Girls grossed nearly $87 million at the box office in 2004, making it one of the top 30 films of the year. The plot features a young girl, played by actress Lindsey Lohan, who moves to a new high school and struggles to navigate the social dynamics of female popularity. The film is rife with female cliques that use covert tactics such as manipulation and gossip to bully other girls. One line in the movie captures perfectly how the girls interact with each other: ‘‘This is a girl world . . .all the fighting has to be sneaky.’’

Critics have argued that Mean Girls’ popularity is due to the fact that its themes resonate with the everyday experiences of young people in this country (Bennet, 2006; Thompson, 2004). Indeed, a recent national survey of 1,001 adolescents suggests that the behaviors featured in the movie are quite common in everyday life (Galinsky & Salmond, 2002). According to the survey, 50% of the adolescents engaged in gossiping about others on a monthly basis. Additionally, 66% of the adolescents reported that they were the target of cruel gossip or teasing at least once per month. In fact, when the teens were asked if they could make one change in their school to help stop violence, the most frequent response was that they would put an end to what the authors called ‘‘emotional violence,’’ or cruel gossip and putdowns.

Corresponding author: Nicole Martins; e-mail: [email protected]

Social Aggression in Programs Popular With Children N. Martins & B. J. Wilson

Where do children learn these types of behaviors? The purpose of this study is to explore the role of the media in children’s development of these nonphysical forms of aggression. A content analysis was conducted to examine the amount and nature of social aggression in the 50 most popular programs among children.

Social aggression Although the harmful effects of overt physical aggression on child development have been widely studied, little attention has been paid to aggressive behavior that is more subtle and relational in nature. Social aggression is a type of aggression that damages a target’s self-esteem or social standing (Galen & Underwood, 1997). Examples of social aggression include gossiping, social exclusion, giving dirty looks, and friendship manipulation. Social aggression includes both indirect (e.g., spreading a rumor) and direct (e.g., ignoring another) acts of aggression. Furthermore, social aggression may be delivered via verbal and nonverbal means, which distinguishes it from other conceptually related overlapping constructs such as indirect (Buss, 1961) or relational aggression (Crick & Grotpeter, 1995).

Research indicates that social aggression is more prevalent among girls. In one study, Crick, Bigbee, and Howes (1996) surveyed 496 adolescents and asked them ‘‘What do most girls/boys do when they are mad at someone?’’ Social aggression was the most frequently cited angry behavior for girls’ interactions, whereas physical aggression was the most frequently cited angry behavior for boys’ interactions. Scholars have argued that socially aggressive strategies are more often used in girls’ peer interactions because such tactics are particularly effective in harming social goals (e.g., establishing close relationships) that are most important to girls (Crick & Grotpeter, 1995).

Because social aggression does not result in physical injury, it may be tempting to conclude that it is less problematic than overt behavioral aggression. Yet the consequences of social aggression can be quite serious. Victims of social aggression typically are rejected by their peers and consequently experience sociopsychological adjustment problems (Crick & Grotpeter, 1995). Victims also report poor self–self concept and self-esteem (Paquette & Underwood, 1999). In extreme cases, social aggression has even been implicated in suicide (Baumeister, 1990). As a result of some of this evidence, the U.S. Department of Health and Human Services has recently identified ‘‘psychological bullying’’ (i.e., when a target is repeatedly the victim of social aggression) as a serious public health issue (What adults can do, 2006).

What causes this nonphysical form of aggression? Scholars have speculated that the same factors that instigate physical aggression in youth also promote social aggression (Galinksky & Salmond, 2002; Twenge, Baumeister, Tice, & Stucke, 2001). For example, family issues, such as exposure to marital conflict, have been linked to both physical (Cummings & Davies, 1994) and social aggression (Crick, 2003). Another environmental factor that may contribute to both social and physical aggression is media violence. In the short term, laboratory studies show that viewing televised aggressive models leads children to imitate aggressive behavior immediately

N. Martins & B. J. Wilson Social Aggression in Programs Popular With Children

after exposure (Bandura, Ross, & Ross, 1963). In the long run, longitudinal studies have found that exposure to television violence in childhood is associated with subsequent increases in adult aggression (Huesmann, 1986; Huesmann, Moise-Titus, Podolski, & Eron, 2003). In addition, several meta-analyses have found a medium- sized effect for the relationship between watching TV and physical aggression in children (Anderson & Bushman, 2001; Paik & Comstock, 1994). Given that children can learn physical aggression from the media, it seems reasonable to expect that they can learn social aggression as well. However, the relationship between media violence and social aggression remains relatively unexamined.

Social aggression on television To date, only two studies have examined the prevalence of social aggression on television. Coyne and Archer (2004) content analyzed 29 programs that were reported by British adolescents as their favorite. The researchers coded the programs for physical aggression (e.g., hitting, punching, kicking) and indirect aggression. The researchers defined indirect aggression as acts of social exclusion (e.g., ignoring others), malicious humor (e.g., prank phone calls), and/or guilt induction (e.g., putting undue pressure on another character). They found that indirect aggression was portrayed in 92% of the episodes. Furthermore, indirect aggression (92%) was more common than physical aggression (55%) in the programs sampled. In terms of characters, females were more likely to be perpetrators of indirect aggression than were males. In contrast, male characters were more likely to be shown as physically aggressive perpetrators. Lastly, indirect aggression was more likely to be committed by an attractive perpetrator, whereas physical aggression was more likely to be committed by an unattractive perpetrator.

In the second study, Feshbach (2005) analyzed 12 programs (6 half-hour sitcoms, 6 one-hour dramas) reported by high school seniors and college juniors as their favorite. She coded the 12 programs for acts of physical and indirect aggression. In this study, indirect aggression was defined as depictions of excluding, ignoring, or rejecting another character. She found that 32% of the male characters and 50% of the female characters displayed some form of indirect aggression in the dramas. In the comedies, almost every female (93%) engaged in indirect aggression compared to fewer than half of the males (40%). Feshbach concluded that the use of indirect aggression is the norm for female comedians.

Although these two content analyses are an important first step in documenting nonphysical forms of aggression on television, they are limited in several ways. First, both studies analyzed small samples of programming. Feshbach’s sample consisted only of 12 shows and Coyne and Archer’s analysis only contained 29 programs. Furthermore, half of Coyne and Archer’s sample was comprised of programs that were not from the United States, and therefore may not be generalizable to the kinds of content that American children watch. The shows that were American in origin were in syndication for several years, such as Tom and Jerry and Star Trek.

Social Aggression in Programs Popular With Children N. Martins & B. J. Wilson

Second, both studies focused on programs that preteens and teens watch. Yet research suggests that social learning from TV begins quite early in development (Bandura, 1986). Indeed, some of the strongest effects of media violence on physical aggression are found among preschoolers (Paik & Comstock, 1994). Moreover, social aggression, especially nonverbal forms of it, seems to emerge as early as the preschool years (Crick, Casas, & Mosher, 1997). Thus, it is crucial to also examine socially aggressive portrayals in programs viewed by younger children.

A final limitation is that the two studies looked primarily at the amount of aggression on television rather than the way in which social aggression was portrayed. Research shows that violent portrayals differ in how much risk they pose to viewers depending upon the context of the aggressive behavior (for review, see Wilson et al., 1998). Coyne and Archer acknowledged the importance of context when they looked at whether the aggression was justified, and whether it was rewarded or punished. However, there are a host of contextual variables that have been identified as factors that encourage the learning of aggression (Wilson et al., 1998).

The present study The present study was designed to address these limitations. We conducted a content analysis of the portrayal of social aggression in the top 50 programs popular with children. The analysis focused on the most watched programs by children ages 2 through 11 as reported by Nielsen Media Research (2005). We analyzed three episodes of each program (150 programs total) to ensure that the sample was as representative as possible. We also focused on a range of contextual variables so that we could fully describe how social aggression on television is portrayed. Finally, we coded physical aggression as well so that we could compare the two forms of aggression.

One of the first issues examined was how often social aggression was portrayed on television. Coyne and Archer (2004) found that indirect aggression occurred in 92% of the episodes sampled. However, their sample size was limited and consisted of programs that were popular with adolescents. Thus, the first research question asked:

RQ1: How often is social aggression portrayed in programs popular with children?

If children can learn social aggression from television, what are they learning? As suggested above, social aggression can come in a variety of different forms. For example, it can be verbal or nonverbal in nature. Similarly, it can be perpetrated directly at a target (e.g., insults, name calling) or it can be perpetrated indirectly or behind a target’s back (e.g., spreading a rumor). Thus, the second research question asked:

RQ2: What does social aggression look like in programs popular with children?

Research demonstrates that physical aggression is pervasive on television, par- ticularly in programs targeted to children (Wilson et al., 1998). Coyne and Archer

N. Martins & B. J. Wilson Social Aggression in Programs Popular With Children

(2004) and the Feshbach (2005) studies found that social aggression occurred more frequently than physical aggression in programs popular with adolescents. On the basis of these studies, the following hypothesis was posed:

H1: Social aggression will occur more frequently than physical aggression in programs popular with children.

Several major reviews of the media violence literature have concluded that certain depictions are more likely than others to pose risks to viewers (Comstock & Paik, 1991; Wilson et al., 1998). One contextual factor to consider is the nature of the perpetrator. According to social cognitive theory, an attractive perpetrator can be a potent role model, and thus increases the likelihood that viewers will learn aggression (Bandura, 1986). One factor related to character attractiveness is the perpetrator’s perceived similarity to the viewer (Eyal & Rubin, 2003). For example, children report liking same-sex more than opposite-sex characters (Jose & Brewer, 1994), and children seem to strongly identify with characters of a similar ethnicity or race (Anderson & Williams, 1983). Children also respond favorably to characters who are perceived as physically attractive. Evidence indicates that a character’s physical appearance has considerable potential to affect impressions because it is generally the first attribute that comes to the viewer’s attention (see Hoffner & Cantor, 1991 for review). A final factor related to character attractiveness is the character’s behavior in the storyline. Research shows that children assign more positive ratings to characters who act benevolently than to characters who are cruel (Hoffner & Cantor, 1985).

A second contextual feature that may influence whether a child will learn social aggression is the perpetrator’s motives or reasons for engaging in violence. Research indicates that exposure to justified violence increases the probability of aggressive behavior (Jo & Berkowitz, 1994). In contrast, violence that is unjustified or purely malevolent decreases the risk subsequent aggressive behavior among viewers (Brown & Tedeschi, 1976). Coyne and Archer (2004) measured justification in their content analysis. The researchers found that indirect aggression was significantly more likely to be justified than unjustified. Conversely, physical aggression was more likely to be portrayed as unjustified.

A third contextual factor to consider is whether the televised violence is rewarded or punished. Social learning theory predicts that violence that is rewarded or not overtly punished fosters the learning of aggressive attitudes and behavior among viewers (Bandura et al., 1963). In contrast, portrayals of punished violence can serve to inhibit or reduce the learning of aggression (Bandura, 1986).

Humor may also facilitate children’s learning of social aggression. There is evidence to support the idea that when humor is linked to physical violence, inhibitions toward committing aggressive acts in viewers are lowered (Baron, 1978). One reason why humor may foster the learning of aggression is because humor increases viewer arousal and heightened arousal has been shown to increase viewer aggression (Zillmann, 1979). Humor may also trivialize the seriousness of violence and its consequences which can increase the probability of aggression.

Social Aggression in Programs Popular With Children N. Martins & B. J. Wilson

A final contextual feature of media violence concerns the consequences of aggression. Research demonstrates that when pain and harm to the victim of physical violence are shown, children are less likely to imitate the aggressive act (Wotring & Greenberg, 1973). The assumption is that pain cues inhibit aggression by eliciting sympathy and reminding the viewer about social norms against violence. The absence of physical harm or emotional suffering in an aggressive scene may fail to trigger such inhibitory reactions and thus increase the learning of aggression.

This study incorporated these contextual variables into its content analysis. This study also compared the amount and context of socially aggressive portrayals with portrayals of physical aggression. Thus, the next two research question asked:

RQ3a: How are socially aggressive portrayals contextualized in programs popular with children?

RQ3b: How does the context of socially aggressive portrayals compare with the context of physically aggressive portrayals in programs popular with children?

Method

Sample The 50 most popular programs among children 2–11 years of age according to Nielsen Media Research (2005) data were selected for inclusion in this study. To assure that subsequent findings were characteristic of the programs in general and not typical of just one program in particular, three episodes of each show were recorded resulting in a total sample of 150 shows. The 150 shows were recorded from December 2006 through March of 2007. Programs aired during sweeps weeks were not used.

Definitions of social aggression and physical aggression On the basis of Galen and Underwood’s (1997) definition, we operationalized social aggression as follows:

Actions directed at damaging another’s self-esteem, social status, or both, and includes behaviors such as facial expressions of disdain, cruel gossiping, and the manipulation of friendship patterns (p. 589).

Physical aggression was also coded so that comparisons between each form of aggression could be made. The National Television Violence Study’s (Wilson et al., 1998) definition of violence was used in this study and was operationalized as:

Any overt depiction of a credible threat of physical force or the actual use of such force intended to physically harm an animate being or group of beings. Violence will also include certain depictions of physically harmful consequences against an animate being(s) that result from unseen violent means (p. 20).

N. Martins & B. J. Wilson Social Aggression in Programs Popular With Children

Units of analysis Social and physical aggression was measured at three units of analysis: the PAT, the aggressive scene, and aggressive program (Wilson et al., 1998). An aggressive PAT was defined as an interaction that occurs between a perpetrator (P) and a target (T) using a particular type of aggressive act (A). Any time a perpetrator, target or type of aggression changed, a new PAT is created. The second level of analysis is the aggressive scene. A scene was defined as interrelated series of PATs that occurred without a significant break in the flow of actual or impending violence (Wilson et al., 1998). Finally, social and physical aggression were examined at the program level. By examining aggression across these levels, features of aggression at different points in the program could be assessed.

Three measures examined the amount of aggression in programs popular with children: prevalence, concentration (i.e., density of aggression in a program), and rate per hour. In addition to these variables, the context of aggression was also measured. Each contextual variable is defined below according to the level at which it was coded.

PAT-level variables Several character attributes, all of which have relevance to viewer identification, were assessed at the incident or PAT level. Coders classified perpetrators according to their type (human, supernatural creature, anthropomorphized character, or other), and sex (male and female). Next, coders assessed the physical attractiveness of each perpetrator. Perpetrators were coded as being very attractive, attractive, not attractive, or neutral (i.e., neither attractive nor unattractive). Finally, the perpetrator’s behavior was coded as good (i.e., benevolent), bad (i.e., malevolent), blended (i.e., good and bad), or neutral (i.e., not featured long enough to assess nature of behavior).

Two variables dealing with the aggression itself were assessed at the PAT level: (a) justification for the aggression—whether the aggression was portrayed as ‘‘morally correct,’’ ‘‘right,’’ or ‘‘just’’ given the circumstances of the plot; and (b) consequences of aggression. The consequences of aggression were assessed by coding the amount of emotional or psychological pain the target experienced as a result of the socially aggressive act. Pain was coded as no pain, mild pain, moderate pain, or extreme pain.

Scene-level variables Several variables were assessed after taking the entire aggressive scene into account. First, the positive and negative reinforcements (i.e., verbal and/or nonverbal approval of aggression) for aggression were coded. The presence or absence of four types of rewards was coded at the scene level: self-praise, praise from others, praise from the audience, or material rewards. In terms of negative reinforcements (i.e., verbal and/or nonverbal disapproval of aggression), the presence or absence of four kinds of punishments were assessed disapproval from the perpetrator, disapproval from others, disapproval from the audience, or loss of material rewards. Finally, scenes were coded for the presence or absence of humor. Humor was defined as the use of speech, actions, and/or behavior that a character engaged in that was intended to amuse the self, another character(s), and/or the viewer (Wilson et al., 1998).

Social Aggression in Programs Popular With Children N. Martins & B. J. Wilson

Program-level variables Coders ascertained the duration negative consequences. Coders assessed the extent of pain depicted across the entire program, indicating whether such consequences generally were: not shown at all, short-term in nature (limited to within a few aggressive scenes), or long-term in nature (displayed throughout the entire program).

Training and reliability Six female undergraduates at a large university in the Midwestern United States performed the coding for this project. The coders received approximately 20 hours of classroom training and 15 hours of laboratory practice to learn the coding scheme. Once trained, the coders performed their work individually in a video laboratory. The entire sample took roughly 10 weeks to code. Each coder was randomly assigned to code 11% of the sample (approximately 16 cases each). Each coder was also assigned a randomly determined overlap of 33% of the sample to allow for an assessment of intercoder reliability. For these cases, any coding disagreements between the coders were resolved by N.M. Intercoder reliability was assessed using Krippendorf’s alpha (Krippendorf, 1980). The coefficients reported below are the median alpha’s for the entire 10-week coding period, with social aggression coefficients listed first and the physical aggression coefficients listed second: type of aggression (.99, 1.0), sex of perpetrator (.99, .97), physical attractiveness (.76, .78), nature of behavior (.87, .87), justification (.88, .89), consequences (.82, .74), self-praise (.88, .92), praise from others (.87, .90), laughter from audience (1.0, 1.0), material reward (.88, .91), self- condemnation (.88, .91), condemnation from others (.91, .91), condemnation from audience (1.0, .90), humor (.86, .89), and duration of negative consequences (.78, .84).

Results

Analysis plan To test our hypothesis and answer some of the research questions, portrayals of social aggression were compared with those involving physical aggression. For these comparisons, a series of chi-square tests (p < .05) were computed and post hoc comparisons were performed using the chi-square analog to the Scheffé procedure (Marascuilo & Busk, 1987). However, because the sample size in this study was unusually large, small differences between percentages (3–4%) were statistically significant but not very meaningful. Therefore, we stipulated that there be at least a 10% difference in magnitude between two percentages in order to establish practical significance (see Wilson et al., 2002). In all cases, then, percentages having no subscripts in common are both statistically (p < .05) and practically (10% difference) different.

Amount of social aggression in programs popular with children Research question 1 The first research question concerns the amount of social aggression that appears in programs popular with children. The amount of social aggression was assessed in

N. Martins & B. J. Wilson Social Aggression in Programs Popular With Children

two ways. First, the proportion of programs that featured some social aggression was examined. The analysis revealed that a full 92% of these shows contained some social aggression. Second, the rate of socially aggressive interactions per hour was assessed. To calculate the rate per hour, the number of aggressive interactions was divided by the number of program hours. There was a total of 85.5 hours of programming in the sample and we coded 1,234 separate PATs of social aggression. By this criterion, there was an average of 14.4 incidents of social aggression per hour, or one incident of social aggression every 4 minutes.

Research question 2 The second research question concerns what social aggression looks like in programs popular with children. The vast majority of socially aggressive incidents (78%) were verbal. In other words, perpetrators used words to hurt the self-esteem or social standing of the target. The most common types of verbal social aggression were insults and name calling (Table 1).

Only about 20% of socially aggressive incidents were nonverbal in nature. When social aggression was nonverbal, perpetrators typically used a mean face or laughter to hurt the self-esteem or social status of the target (Table 2). It was very uncommon for social aggression to be both verbal and nonverbal in the same exchange (3%).

We also coded whether social aggression was directly perpetrated at the target (e.g., making a mean face) or indirectly perpetrated behind the target’s back (e.g., spreading

Table 1 Categories of Verbal Socially Aggressive Incidents

Category Percent

Insult 52 Name calling 25 Teasing 10 Sarcastic remark 9 Other 4

Table 2 Categories of Nonverbal Socially Aggressive Incidents

Category Percent

Mean/disgusted face 36 Laughing/giggling 31 Eyes rolling 8 Finger pointing 3 Ignoring 3 Sticking out tongue 2 Staring 2 Other 13

Social Aggression in Programs Popular With Children N. Martins & B. J. Wilson

a rumor). The vast majority of socially aggressive incidents (86%) were enacted directly at the target. Rarely were socially aggressive incidents (14%) perpetrated behind the target’s back.

Hypothesis 1 The first hypothesis predicted that social aggression would occur more frequently than physical aggression in programs popular with children. To test this hypothesis, both the prevalence and rate of physical aggression were calculated. The prevalence analysis revealed that 81% of the programs in the sample contained some physical aggression. A McNemar test revealed that these programs were significantly more likely to feature social than physical aggression, McNemar’s χ2(1, N = 150) = 11.56, p = .001.

The frequency of physical aggression was examined by computing the rate within the programs. In 81.5 hours of programming, 1,261 separate PATs of physical aggression were coded. This computed to an average rate of 15.4 PATs of physical aggression per hour, or one incident of physical aggression every 5 minutes. Thus, Hypothesis 1 was partially supported. Social aggression was more prevalent across programs popular with children than was physical aggression, but the rates of the two types of aggression within programs were nearly equal.

Context of aggression Research questions 3a and b The third research question concerned the (a) context of social aggression in programs popular with children and (b) how the context of socially aggressive portrayals compared to the context of physically aggressive portrayals. Table 3 shows the comparisons of social and physical aggression as a function of the five contextual features of violence. In terms of the perpetrators of aggression, the vast majority of socially aggressive incidents (86%) were perpetrated by human characters. A much smaller proportion of incidents (11%) were perpetrated by anthropomorphized characters. Rarely were socially aggressive incidents (2%) enacted by supernatural creatures. A chi-square analysis comparing social and physical aggression by form of the perpetrator was statistically significant, χ2(2, N = 2,475) = 126.55, p < .001, V∗ = .22. Post hoc comparisons revealed that socially aggressive incidents were more likely to be perpetrated by humans than were physically aggressive incidents (Table 3).

We also coded the sex of the perpetrator. Results revealed that nearly two-thirds of socially aggressive incidents (62%) were perpetrated by males, whereas only one-third of such incidents (38%) were perpetrated by females. Although most aggression was perpetrated by males, females were slightly more likely to perpetrate socially aggressive incidents than physically aggressive incidents whereas males were more likely to perpetrate physical aggression, χ2(1, N = 2,284) = 17.73, p < .001, V∗ = .29. However, this difference fell just short of practical significance.

Beyond demographics, we also examined two qualities of the perpetrator that reflected how attractive they were in the program: physical appearance and behavior in the storyline. In terms of appearance, results revealed that roughly one-fourth of

N. Martins & B. J. Wilson Social Aggression in Programs Popular With Children

Table 3 Context of Social Aggression Versus Context of Physical Aggression

Contextual Variable Social Aggression Physical Aggression

Nature of the perpetrator Human (PAT) 86%a 72%b Female (PAT) 38 29 Physically attractive (PAT) 26b 17a Blended characters (PAT) 36b 14a

Justified aggression 32b 13a Rewards/punishments

Immediate rewards (Scene) 17 12 Immediate punishments (Scene) 4 6

Humor (Scene) 76b 61a Consequences of social aggression

Victim shows no pain (PAT) 71 70 Depicts long-term suffering (Program) 19 5

Note: Within rows, percentages having no subscripts in common are both statistically (p < .05) and practically (10%) different.

socially aggressive incidents (27%) were enacted by physically attractive perpetrators. Far fewer incidents (5%) were committed by unattractive perpetrators. By far, most of the socially aggressive incidents (76%) in these programs were enacted by characters who were coded as neutral in appearance. Although the majority of perpetrators were neutral in appearance, social aggression was more likely than physical aggression to be committed by a physically attractive perpetrator, χ2(2, N = 2,154) = 33.49, p < .001, V∗ = .12.

In terms of the perpetrators’ behavior in the story line, results revealed that only 20% of socially aggressive incidents were perpetrated by characters who behaved benevolently. Roughly one-third of such incidents (32%) were perpetrated by char- acters who behaved meanly and one-third (36%) were perpetrated by characters who displayed both good and bad behaviors. A small proportion of socially aggressive inci- dents (12%) were committed by neutral characters. A chi-square analysis comparing type of aggression by nature of the perpetrators’ behavior was statistically significant, χ2(3, N = 2,527) = 168.26, p < .001, V∗ = .25. Socially aggressive incidents were more likely than physically aggressive incidents to be committed by a character that displayed both good and bad behaviors (Table 3).

Another contextual feature concerned how often social aggression is portrayed as morally justified in the plot. Results revealed that less than 20% of socially aggressive incidents were portrayed as justified in the plot. Moreover, socially aggressive incidents were less likely than physically aggressive incidents to be portrayed as justified, χ2(1, N = 2,522) = 340.53, p < .001, V∗ = .36.

In terms of reinforcements, social aggression was neither rewarded nor punished in the majority of scenes (78%). When reinforcements were portrayed, only a small

Social Aggression in Programs Popular With Children N. Martins & B. J. Wilson

proportion of scenes (17%) presented social aggression as being rewarded. Far fewer of the scenes (4%) presented social aggression as being punished. Virtually none of the scenes (2%) featured social aggression as being both rewarded and punished (Table 3).

Although rewards were rarely shown in socially aggressive scenes, the most common rewards were self-praise. In particular, the perpetrator expressed personal satisfaction for social aggression in 60% of rewarded scenes. A much smaller percentage of rewarded scenes (28%) featured other characters who expressed approval of the aggression. Far fewer of the scenes involving rewards (9%) portrayed the perpetrator receiving material goods for social aggression. Only 3% of the rewarded scenes portrayed the perpetrator receiving praise from the audience (i.e., audience laughter, cheers).

Just as with rewards, punishments did not occur very often in the immediate context of social aggression. When scenes with punishments were featured, the most common form involved condemnation expressed from characters other than the perpetrator (74%). Some of the scenes involving punishment featured negative audience reactions (e.g., ‘‘oohs’’) for social aggression (13%). Far fewer of the scenes (10%) with punishment showed a perpetrator feeling remorse for social aggression. Only 3% of the scenes with punishment featured the perpetrator losing material rewards as a result of social aggression.

A chi-square analysis comparing type of aggression (social vs. physical) by reinforcement (rewarded, punished, neither or both) was statistically significant, but none of these differences was practically significant (Table 3).

Another contextual factor concerned the presence or absence of humor. The findings revealed that three-fourths of socially aggressive scenes contained some form of humor. In addition, socially aggressive scenes were significantly more likely than physically aggressive scenes to be presented in a humorous context, χ2(1, N = 1,394) = 67.66, p < .001, V∗ = .22.

The final contextual factor concerns whether negative consequences of social aggression are portrayed on television. The findings revealed that the majority of socially aggressive incidents (71%) portrayed the target experiencing no pain whatsoever. A much smaller proportion of socially aggressive incidents (28%) showed the target experiencing mild pain. Virtually none of the incidents featured the target experiencing moderate pain (1%) or extreme pain (0%). Although a chi-square analysis comparing social and physical aggression by the amount of pain was statistically significant, these differences were not practically significant.

The consequences of social aggression were also coded at the program level. The findings indicate that about one-third of the programs (31%) in the sample showed virtually no negative consequences of social aggression. About one-half of the programs (49%) portrayed the short-term consequences of social aggression. Far fewer of the programs (19%) depicted the long-term negative consequences of social aggression. No significant difference emerged between social and physical aggression in long-term consequences (Table 3).

N. Martins & B. J. Wilson Social Aggression in Programs Popular With Children

Discussion

The results of this study indicate that socially aggressive role models are prevalent on programs popular with children. The findings also suggest that some of the ways in which social aggression is contextualized make these depictions particularly problematic for young viewers. In some cases, social aggression on television may pose more of a risk than portrayals of physical aggression do.

In terms of specific findings, the first research question concerned how often social aggression was portrayed in programs popular with children. We found that a full 92% of the programs in the sample contained some social aggression. Moreover, we also examined the rate, a variable not measured in the two prior studies of social aggression on television. We found an average of 14 incidents of social aggression per hour, or one incident every 4 minutes.

The second research question concerned what social aggression looks like in children’s favorite shows. This study found that the vast majority of socially aggressive incidents were verbal in nature, where the perpetrator used words to hurt the self- esteem or social standing of the target. The most common forms of verbal social aggression were insults and name calling. In addition, socially aggressive incidents were nearly always enacted directly at the target. Rarely were socially aggressive incidents indirectly perpetrated.

This form of televised social aggression is not consistent with what we know about social aggression in real life. Research has demonstrated that gossip is one of the most common forms of social aggression (e.g., Galinsky & Salmond, 2002). Yet in this study, gossip was rarely portrayed on television. One possibility why gossip was seldom portrayed is because television is made of conflict where main characters go head-to-head in direct confrontations. Gossip, on the other hand, is an indirect act that does not involve a direct exchange between a perpetrator and a target. Given that young children are more likely to attend to conflict that is concrete and perceptually salient (Van Evra, 2004), gossip may elude young audiences because it is too subtle to capture their attention. Thus, gossip is seldom used in these shows as a way to advance the plot. Future research should examine whether gossip is more common among programs that older children watch.

In addition to examining the amount of social aggression in children’s favorite shows, we also examined the way in which social aggression was portrayed (Research Question 3). The results of this study suggest that the perpetrators of social aggression can serve as potent role models for children. The fact that this study found that the preponderance of socially aggressive incidents was perpetrated by humans suggests that children are likely to attend to socially aggressive perpetrators. Human characters are presumably easier to identify with because they are similar to the self (von Feilitzen & Linne, 1975), thereby increasing the likelihood that the child viewer will attend to these characters. In addition, socially aggressive incidents tended to be perpetrated by females. Several studies have shown that viewers attend more closely to the actions of same-sex characters and remember more content concerning those characters (see

Social Aggression in Programs Popular With Children N. Martins & B. J. Wilson

Hofner & Cantor, 1991). This could result in young girls, in particular, attending to these behaviors because female characters are modeling them. This finding seems to reinforce the notion that such behaviors are typically perpetrated by girls in real life (Crick et al., 1996).

In addition to demographics, we also examined two qualities of the perpetrator that reflected how attractive they were in the program: physical appearance and behavior in the storyline. In terms of appearance, we found that the majority of aggressive incidents were perpetrated by characters who were coded as neutral (e.g., neither attractive nor unattractive). Even so, socially aggressive incidents were significantly more likely to be committed by an attractive perpetrator. Evidence indicates that a character’s physical appearance, particularly attractiveness, has considerable potential to affect impressions because it is generally the first attribute that comes to the viewer’s attention (see Hoffner & Cantor, 1991 for review). Given that this study found socially aggressive incidents were often committed by an attractive character, it seems reasonable to assume that a child will attend to socially aggressive perpetrators on TV.

When behavior in the storyline was examined, we found that socially aggressive incidents were committed by blended characters that displayed both good and bad qualities. This finding suggests that the socially aggressive perpetrators in this sample are not the traditional ‘‘good’’ or heroic characters we are accustomed to seeing on television. Research demonstrates that preschoolers, as well as older viewers, assign more positive ratings to characters who engage in kind, helpful behavior than to those who behave cruelly or unkindly to another character (Hoffner & Cantor, 1985). What kind of rating, then, will children assign to characters who display both good and bad behaviors? Unfortunately, there is no evidence available to answer this question. Future research should examine children’s interpretation of characters who exhibit both good and bad motives.

We also examined how often social aggression was portrayed as morally justified in the plot. We found that fewer than 15% of socially aggressive incidents were portrayed as justified. There is evidence to indicate that acts that appear to be justified or morally defensible are likely to facilitate viewer aggression, whereas unjustified violence can actually diminish the risk of subsequent aggression (Brown & Tedeschi, 1976; Jo & Berkowitz, 1994). Thus, the fact that the majority of socially aggressive incidents were unjustified may actually inhibit children’s learning of such behaviors.

In terms of the reinforcements delivered for aggression, we found that in the majority of aggressive scenes (78%), social aggression was neither rewarded nor punished. When reinforcements were portrayed, only a small proportion of scenes (17%) presented social aggression as being rewarded. Far fewer of the scenes (4%) presented social aggression as being punished. Social learning theory predicts that children will imitate a model if a reward is delivered for performing the behavior (Bandura, 1965). Moreover, behaviors that are not overtly punished can also foster imitation because the absence of punishment serves as a tacit reward or sanction for such behavior (Bandura, 1965). In contrast, portrayals of punished violence can serve

N. Martins & B. J. Wilson Social Aggression in Programs Popular With Children

to inhibit or reduce the learning of aggression (Bandura, 1986). Taken together, then, 95% of socially aggressive scenes in programs popular with children are portrayed in such a way that fosters children’s imitation of social aggression. Only a small percentage of socially aggressive scenes in children’s favorite shows serve to diminish the learning of such behaviors.

This study found that humor is featured in conjunction with social aggression in one in four socially aggressive scenes. Although humor is the least understood of all contextual features, there is some evidence to suggest that humor can increase viewer aggression (Baron, 1978). Thus, it appears that the serious nature of social aggression is often trivialized and heightens the probability that children will learn such behavior from viewing.

Finally, we also assessed whether the negative consequences of social aggression were portrayed in programs that children watch. As noted earlier, research has shown that the depiction of a victim’s pain can inhibit the learning of aggression among viewers (Wotring & Greenberg, 1973). Yet, across two measures, the negative consequences of social aggression were rarely portrayed.

There are at least two reasons why the consequences of social aggression may be lacking in such programs. One possibility is that the forms of social aggression in programs popular with children are simply less serious or benign. However, the type of socially aggressive behaviors typically depicted in these programs—name calling, insults, making mean faces—are known to cause serious psychological harm to victims in the real world (Paquette & Underwood, 1999). Therefore, the sanitation of social aggression is not due to a difference in the seriousness of the aggression. A second possibility for the lack of consequences is that social aggression often occurs in humorous scenes. Given that one in four socially aggressive scenes were couched in humor, it would be difficult to introduce the consequences to the victim in these scenes while still keeping the scene funny.

How do these findings compare with previous research? The only two prior content analyses on social aggression have found, like this study, that social aggression is prevalent on the programs that children watch. Yet, when this study is compared to the Coyne and Archer (2004) study, it is somewhat surprising that the prevalence figures for social aggression are identical (92%) given that we used the more expansive definition of social aggression and a much larger sample. Nonetheless, this study sets new benchmarks from which future content analyses of social aggression in programs popular with children can be compared because of the expansive definition of social aggression used, the size of the sample analyzed, and the examination of some contextual features associated with socially aggressive portrayals.

In terms of what social aggression looks like, assessing how this study compares to the other two studies on social aggression is more difficult. We cannot directly compare this study to Feshbach’s content analysis because she did not report what indirect aggression looked like in the programs in her sample. Coyne and Archer, however, did report the frequency with which different forms of social aggression occurred, and reported that gossiping accounted for nearly 17% of the

Social Aggression in Programs Popular With Children N. Martins & B. J. Wilson

total aggressive acts in their analysis whereas name calling only accounted for 9% of the total aggressive acts. Yet drawing comparisons is challenging because Coyne and Archer examined social aggression in programs popular with adolescents. There is evidence to support that older children have the cognitive capacity to understand more nuanced behaviors such as gossip (Hill & Pillow, 2006); thus offering a possible explanation as to why gossip was used more in the storylines in their sample.

An additional goal of this study was to compare social aggression to physical aggression. Although we found social aggression (92%) was significantly more likely to occur in programs that children watch than was physical aggression (81%), the rate of physical aggression in these programs was very similar to the rate of social aggression. Thus, children who watch a typical hour of their favorite programs are likely to witness more programs that feature social than physical aggression. Within each show, however, the rate of aggressive incidents is similar.

There were some other important differences in the way both types of aggression were portrayed. In this study, we found that socially aggressive incidents were sig- nificantly more likely to be perpetrated by humans than were physically aggressive incidents. In contrast, physically aggressive incidents were more likely to be perpe- trated by supernatural creatures than were socially aggressive incidents. This finding suggests that children may be more likely to attend to socially aggressive perpetrators than physically aggressive perpetrators because human characters are more similar to the self and presumably easier to identify with, an important factor associated with imitation (von Feilitzen & Linne, 1975). Another contextual feature examined was the perpetrator’s physical appearance. We found that socially aggressive inci- dents were more likely than physically aggressive incidents to be committed by an attractive perpetrator. As mentioned earlier, a character’s physical attractiveness has considerable potential to affect impressions (see Hoffner & Cantor, 1991 for review). Thus, at least in terms of this one contextual feature, socially aggressive portrayals that feature physically attractive human characters may pose more of a risk for the child viewer than portrayals of physical aggression do.

Another important difference between social and physical aggression found in this study concerns the contextual feature of humor. We found that socially aggressive scenes were significantly more likely than physically aggressive scenes to be presented in a humorous context. Here, too, is another example where portrayals of social aggression may pose more of a risk to the child viewer than portrayals of physical aggression would. Because there is evidence to suggest that humor may increase viewer aggression (Baron, 1978), it seems reasonable to assume that viewers may be more likely to learn socially aggressive behaviors than physically aggressive behaviors if couched in humor.

To summarize, this study demonstrates that socially aggressive role models are prevalent on programs popular with children. The findings also suggest that some of the ways in which social aggression is contextualized poses more of a danger than portrayals of physical aggression do. Given the large body of research that supports that heavy exposure to media violence leads to increased physical aggression in

N. Martins & B. J. Wilson Social Aggression in Programs Popular With Children

children (Anderson & Bushman, 2001; Paik & Comstock, 1994), it seems plausible that children may be at risk for learning socially aggressive behaviors; behaviors that are more prevalent and just as concentrated as physical aggression in children’s favorite shows.

Practically, these findings should help parents and educators recognize that there are socially aggressive behaviors on programs children watch. Parents should not assume that a program is okay for their child to watch simply because it does not contain physical violence. Parents should be more aware of portrayals that may not be explicitly violent in a physical sense but are nonetheless antisocial in nature. Such content may be encouraging children to engage in behavior that is destructive and cruel.

Of course, we cannot make firm claims about what types of effects exposure to these portrayals may have on young viewers. The next step is to ascertain whether viewing these types of acts is associated with an increase in aggression that is more subtle and more relational in orientation.

Acknowledgments

This research was funded by a grant awarded to N.M. from the Fred Rogers Scholarship Memorial Fund. We thank Brandi Cooke, Jill Dressen, Rebekah Pure, and Amy Snyder for their help in program coding. We also thank Dale E. Brashers, Kristen Harrison, David Tewksbury, and two anonymous reviewers for their insightful contributions.

References

Anderson, C. A., & Bushman, B. J. (2001). Effects of violent video games on aggressive behavior, aggressive cognition, aggressive affect, physiological arousal, and prosocial behavior: A meta-analytic review of the scientific literature. Psychological Science, 12, 353–359.

Anderson, W. H., & Williams, B. M. (1983). TV and the Black child: What Black children say about the shows that they watch. Journal of Black Psychology, 9, 27–42.

Bandura, A. (1965). Influence of models’ reinforcement contingencies on the acquisition of imitative responses. Journal of Abnormal and Social Psychology, 66, 575–582.

Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory. Englewood Cliffs, NJ: Prentice Hall.

Bandura, A., Ross, D., & Ross, S. A. (1963). Imitation of film-mediated aggressive models. Journal of Abnormal and Social Psychology, 66, 3–11.

Baron, R. A. (1978). The influence of hostile and nonhostile humor upon physical aggression. Personality and Social Psychology Bulletin, 4, 77–80.

Baumeister, R. F. (1990). Suicide as an escape from self. Psychological Review, 97, 90–113. Bennet, K. (2006). Mean girls. Retrieved from

www.movies.go.com/parentpreviews/review?rid=1189.htm Brown, R. C., & Tedeschi, J. T. (1976). Determinants of perceived aggression. Journal of

Social Psychology, 100, 77–87.

Social Aggression in Programs Popular With Children N. Martins & B. J. Wilson

Buss, A. H. (1961). The psychology of aggression. New York, NY: Wiley. Comstock, G., & Paik, H. (1991). Television and the American child. New York, NY:

Academic Press. Coyne, S. M., & Archer, J. (2004). Indirect aggression in the media: A content analysis of

British television programs. Aggressive Behavior, 30, 254–271. Crick, N. R. (2003). A gender-balanced approach to the study of childhood aggression and

reciprocal family influences. In: A. C. Crouter & A. Booth (Eds.), Children’s influence on family dynamics: The neglected side of family relationships. (pp. 229–235). Mahwah, NJ: Erlbaum.

Crick, N., Bigbee, M. A., & Howes, C. (1996). Gender differences in children’s normative beliefs about aggression: How do I hurt thee? Let me count the ways. Child Development, 67, 1003–1014.

Crick, N. R., Casas, J. F., & Mosher, M. (1997). Relational and overt aggression in preschool. Developmental Psychology, 33, 579–588.

Crick, N. R., & Grotpeter, J. K. (1995). Relational aggression, gender, and social-psychological adjustment. Child Development, 66, 710–722.

Cummings, E. M., & Davies, P. (1994). Children and marital conflict: The impact of family dispute and resolution. New York, NY: Guilford Press.

Eyal, K., & Rubin, A. (2003). Viewer aggression and homophily, identification, and parasocial relationships with television characters. Journal of Broadcasting & Electronic Media, 47, 77–98.

Feshbach, N. D. (2005). Gender and the portrayal of direct and indirect aggression on television. In E. Cole & J. H. Daniel (Eds.), Featuring females: Feminist analyses of media (pp. 155–166). Washington, DC: American Psychological Association.

Galen, B. R., & Underwood, M. K. (1997). A developmental investigation of social aggression among children. Developmental Psychology, 33, 589–600.

Galinsky, E., & Salmond, K. (2002). Youth & violence: Students speak out for a more civil society, summary and discussion guide. Retrieved from http://www.familiesandwork.org

Hill, V., & Pillow, B. H. (2006). Children’s understanding of reputations. The Journal of Genetic Psychology, 167, 137–157.

Hoffner, C., & Cantor, J. (1985). Developmental differences in responses to a television character’s appearance and behavior. Developmental Psychology, 21, 1065–1074.

Hoffner, C., & Cantor, J. (1991). Perceiving and responding to mass media characters. In J. Bryant & D. Zillmann (Eds.), Responding to the screen (pp. 63–101). Hillsdale, NJ: Erlbaum.

Huesmann, L. R. (1986). Psychological processes promoting the relation between exposure to media violence and aggressive behavior by the viewer. Journal of Social Issues, 42, 125–139.

Huesmann, L. R., Moise-Titus, J., Podolski, C. L., & Eron, L. D. (2003). Longitudinal relations between children’s television exposure to TV violence and their aggressive and violent behavior in young adulthood: 1977–1992. Developmental Psychology, 39, 201–221.

Jo, E., & Berkowitz, L. (1994). A priming effect analysis of media influences: An update. In J. Bryant & D. Zillmann (Eds.), Media effects (pp. 43–60). Hillsdale, NJ: Erlbaum.

Krippendorf, K. (1980). Content analysis: An introduction to its methodology. New York, NY: Sage.

N. Martins & B. J. Wilson Social Aggression in Programs Popular With Children

Marascuilo, J. M., & Busk, P. L. (1987). Loglinear models: A way to study main effects and interactions for multidimensional contingency tables with categorical data. Journal of Counseling Psychology, 26, 443–455.

Nielsen Media Research (2005). 2005 report on television. New York, NY: Author. Paik, H. J., & Comstock, G. (1994). The effects of television violence on antisocial behavior:

A meta-analysis. Communication Research, 21, 516–546. Paquette, J. A., & Underwood, M. K. (1999). Gender differences in young adolescents’

experiences of peer victimization: Social and physical aggression. Merrill-Palmer Quarterly, 45, 242–266.

Thompson, G. (2004). Mean girls. Retrieved from www.philly.com/mld/dailynewsliving/ 8555723.htm

Twenge, J. M., Baumeister, R. F., Tice, D. M., & Stucke, T. S. (2001). If you can’t join them, beat them: Effects of social exclusion on aggressive behavior. Journal of Personality and Social Psychology, 81, 1058–1069.

Van Evra, J. (2004). Television and child development (3rd ed.). Mahwah, NJ: Erlbaum. von Feilitzen, C., & Linne, O. (1975). The effects of television on children and adolescents

identifying with television characters. Journal of Communication, 4, 31–55. What adults can do (2006). Retrieved from http://www.stopbullyingnow.hrsa.gov Wilson, B. J., Kunkel, D., Linz, D., Potter, W. J., Donnerstein, E., Smith, S. L., et al. (1998).

Violence in television programming overall: University of California, Santa Barbara study. In National television violence study (Vol. 2, pp. 3–204). Thousand Oaks, CA: Sage.

Wilson, B. J., Smith, S. L., Potter, W. J., Kunkel, D., Linz, D., Colvin, C. M., et al. (2002). Violence in children’s television programming: Assessing the risks. Journal of Communication, 52(1), 5–35.

Wotring, C. E., & Greenberg, B. S. (1973). Experiments in televised violence and verbal aggression: Two exploratory studies. Journal of Communication, 23, 446–460.

Zillmann, D. (1979). Hostility and aggression. Hillsdale, NJ: Erlbaum.

or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission.

However, users may print, download, or email articles for individual use.