16

Violent Content on US Television

A Historical Overview of the Research

Nancy Signorielli

Violence on television has been a critical societal concern since television entered our homes in the 1950s. Most of the attention and research on violence have focused on its portrayal and its effects, particularly in relation to children. This chapter will discuss the portrayal of violence on US television from a historical perspective and will examine what we know about television violence at the start of the second decade of the twenty-first century. It will discuss how violence has been defined in this research, how public policy has been an important component fueling research on television violence, as well as some theoretical perspectives relating to the portrayal of violence. The predominant focus is on television violent content in a US context.

As we progress in the twenty-first century, although we spend more and more time with “screens,” the screen we still watch the most is television (Blair, 2009). Today, however, the way we watch has changed from the decades when we only could watch the “big three” networks (ABC, CBS, NBC), perhaps PBS, and maybe a UHF station. Now broadcast network television as well as cable channels, while still designed to appeal to the general audience, seek more narrow audiences. Television has also become an “on demand” medium. We can purchase DVDs, whole seasons of our favorite programs, and can call up movies and programs whenever and, to some extent, wherever we want “on demand” (Gandossy, 2009). Consequently, we are no longer tied to the time-bound schedules of broadcast television or cable channels. In addition, we spend more and more time watching television on computers and the World Wide Web, mobile phones, gaming devices, and MP3 players that do more than just play music. Images and screens thus take up a substantial portion of our day.

Nielsen (2009) reports that viewing statistics are stable and have been rising for the past few years. At the end of 2008 the average person watched TV in their home for about 5 hours a day or more than 150 hours a month. Watching TV on a mobile phone could add another 3.5 hours a month, and watching on the Internet another 2.5 hours a month. Moreover, about 7 hours a month of viewing were “time-shifted” programs. Children's media use and television viewing remained high. The most recent report of the Kaiser Family Foundation, a survey of 2,000 youngsters between 8 and 18, found that children's media use averaged more than 7.5 hours a day, a figure that expanded to almost 11 hours a day when multitasking was taken into account (Rideout, Foehr, & Roberts, 2010). Children watched television for 4.5 hours a day, which included 24 minutes watching TV or movies on the Internet, about 15 minutes watching on their cell phones, and another 16 minutes watching on their iPods. The Nielsen report (2009) also found that viewing increased with age. Young adults between 18 and 24 watched almost 4 hours a day, those between 25 and 45 watched almost 5 hours a day, while those older than 45 watched more than 6 hours a day. Consequently, at the end of the first decade of the twenty-first century, few escaped exposure to television.

Television's role for the past 60 years as the central and most pervasive mass medium in US culture has not changed. Television still plays a distinctive and historically unprecedented role as our nation's most common, constant, and vivid learning environment. Television, whether we watch on a traditional set, on our mobile phone, or on the Internet, is still our primary storyteller, telling most of the stories to most of the people, most of the time. Stories are found in dramas, action-adventures, situation comedies, reality programs, the news, and even commercials. These stories, however, are the product of a few multinational and centralized commercial institutions. Television's stories show and tell us about life – its people, places, power, and fate – as well as how things work and how to solve problems. Characterizations represent the good and bad, the happy and sad, successes and failures, and show who's on the top and who's on the bottom of the economic ladder and/or pecking order. As such, characters do not live or die but are created or destroyed to tell the story.

Common themes cut across all programs and present aggregate and conventional images. Commercial constraints, however, limit the diversity of television content (Morgan, Shanahan, & Harris, 1990). Television thus cultivates a common world view and common stereotypes through a relatively restrictive set of programs, images, and messages. One such set of images, in particular, relates to violence.

This chapter will focus specifically upon the portrayals of violence on US television, most often discussing prime-time broadcast programming as this is where the lion's share of the research has been conducted. The review will discuss television violence from a historical perspective as well as discussing the existing research, focusing specifically upon research that has been conducted and published during the first decade of the twenty-first century. In addition, in order to provide the most up-to-date assessment of violence on prime-time television, the review will present and discuss data from the author's ongoing yearly content analyses of week-long samples of prime-time network programming broadcast each fall between 2000 and 2009. The data set consists of 783 programs and 3,268 major or leading characters. The methods used in this ongoing research can be reviewed in several recent publications by the author (see, e.g., Signorielli, 2003). Although important, this review will not discuss violence in videogames (since another chapter in this volume covers that topic), nor will it discuss effects of television violence.

The Policy Perspective

In the past 40 years US public concern about television violence has fluctuated, almost cyclically. Between 1968 and 1978 there was considerable public concern and numerous Congressional hearings about the amount of violence on television. Most, if not all, of these hearings did not result in substantive action (Hoerrner, 1999). Public debate subsided during the 1980s era of deregulation. Concern about television violence surfaced again in the early 1990s with the passing of the Television Violence Act (designed to protect the networks, for three years, from antitrust action if they joined to talk about ways to reduce violence on television). Toward the end of 1992, when it appeared as though little had changed in regard to television violence, the close expiration date of the Television Violence Act prompted its author, Senator Paul Simon (D-Illinois), to warn of harsher legislation. The result was a renewed promise by network executives that they would explore ways to reduce violence in prime time (Dustin, 1992). In 1993, for the first time since the late 1970s, Congressional hearings on television violence were held and a number of separate bills relating to television violence were introduced in Congress (Hoerrner, 1999). In response to Congressional concern, the television industry implemented parental “advisories” before those programs they designated as “violent.”

These advisories, however, did not adequately solve the problem and an amendment was added to the Telecommunications Act of 1996 that mandated all television sets 13” or larger, manufactured after 1999, be equipped with the V-chip, an electronic device that enables parents to screen and block the programs their children watch on television. The television industry was asked to develop a rating system to use with the V-chip to filter violent and sexually explicit programming (Federal Communications Commission [FCC], 2000). The result was the implementation of ratings (TV-G, TV-PG, TV-PG14, TV-M, TV-Y, TV-Y7, and TV-Y7-FV), similar to those used by the motion picture industry, supplemented by advisories for content (V-violence, S-sexual situations, D-suggestive dialogue, and L-language).

In the late 1990s, a Kaiser Family Foundation study of programs on 10 major US broadcast and cable channels (Kunkel, Farinola, Farrar, Donnerstein, Biely, & Zwarun, 2002) found that the ratings did not adequately reflect content. For example, while 7 out of 10 PG-14-rated programs had violence, 1 out of 5 G-rated programs and more than half of the PG-rated programs also had violence. Moreover, many of the programs that had violence were not given the content-based “V” rating and many children's violent cartoons were not given the “FV” (fantasy violence) label. Similarly, Signorielli's (2005) analysis of nine week-long samples of prime-time network programs broadcast between 1997 and 2003 found that 9 out of 10 programs had an age-based rating while only 4 out of 10 had a content-based rating. While those programs with age- and particularly content-based ratings accurately reflected the degree of violence in the program, many programs with violent content were not given the “V” content-based rating. Unfortunately, while half of the parents surveyed by the Annenberg Public Policy Center at the University of Pennsylvania in June 2000 said that they were aware of the ratings, very few said that they actually used them (Woodward, 2000). Finally, there is evidence that ratings are most often used by those parents who are very much aware of the problems television viewing may pose to their children. In short, advisories, age- and content-based ratings, may be “preaching to the choir” (Abelman, 1999, p. 544).

The Theoretical Perspective

Numerous theories explain why the study of television violence is important and how it may affect viewers, especially children. Desensitization (see Potter, 1999) and social learning/social cognitive theory (Bandura, 2009), for example, examine the immediate and typically harmful effects of viewing violence. Desensitization or “conscience-numbing effects” (Bushman, Huesmann, & Whitaker, 2009, p. 364) suggests viewing media violence results in viewers becoming less sensitive or sympathetic to real-life violence; viewers may have less empathy and are less willing to help victims of violence. Priming is also an important theory. Berkowitz (1984), for example, posits that exposure to media violence may move viewers, especially those who may be frustrated or angry, to be poised to behave aggressively in certain situations. Similarly, Huesmann's (1988) social informational processing theory posits that viewing violence may lead to the development of scripts about violence that provide guidelines about what to do in certain situations or what kinds of consequences might result from certain aggressive behaviors.

Cultivation theory and the Cultural Indicators paradigm, on the other hand, look at viewing violence from a cumulative, long-term perspective and examine institutional policy perspectives, messages about violence on television, and what it means for viewers to live with violent television images. Cultivation theory argues that to understand the effects of viewing on attitudes, beliefs, and behaviors we must examine television as a collective symbolic environment with an underlying formulaic structure (Morgan, Shanahan, & Signorielli, 2009). Commercial constraints necessitate that common themes cut across all programs. These, in turn, cultivate common world views and stereotypes. Violence is one such theme and is especially important in the cultivation perspective because people are more likely to experience violence when they watch television (whether in news or entertainment programs) than in real life. Consequently, cultivation theory predicts that people's conceptions about violence are more likely to reflect the messages about violence they see, day in and day out, on television. Cultivation research has found that those who watch more television are more likely to view the world as a mean and scary place, to believe that crime and violence are more prevalent than they actually are, and to take precautions to protect themselves, their homes, and their families against crime (Gerbner, Gross, Morgan, & Signorielli, 1994).

Defining Violence

Research on television violence has, for the most part, focused on physical violence. There have been relatively few studies of psychological (emotional/verbal) violence (see, e.g., Mustonen & Pulkkinen, 1997) because psychological violence is more difficult to define, both conceptually and operationally, and to isolate consistently. In fact, measures of psychological violence often do not meet the discipline's standards for reliability.

Research conducted in the Cultural Indicators paradigm defines violence as “the overt expression of physical force (with or without a weapon, against self or other) compelling action against one's will on pain of being hurt or killed, or actually hurting or killing” (Signorielli, Gross, & Morgan, 1982, p. 163). This focus includes all plausible and credible violence. Comic violence in a humorous context is included because humorous violence may increase the risk of learning aggressive behaviors (Baron, 1978; Berkowitz, 1970). “Accidental” violence and violent “acts of nature” are also included because such actions are purposeful, claim victims, and demonstrate power. In fiction, writers add such scenes to programs to propel the story and perhaps to eliminate or incapacitate certain characters. Interestingly, gatekeepers of some recent reality programs (e.g., Survivor) have added physical violence between “contestants” or “participants” to maintain interest in the competition or day-to-day group activities.

The National Television Violence Study (NTVS), conducted during the mid-1990s, also focused on physical violence. This project defined violence 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” (Wilson, Colvin, & Smith, 2002, p. 41). The NTVS thus included credible threats of physical violence, rather than psychological or emotional violence, and also included “depictions of the harmful consequences of unseen violence” (Wilson et al., 2002, p. 41). In addition, the framework of the NTVS was sensitive to the context in which violence was presented (committed by heroes or villains, punished or rewarded, etc.) and brought context to the foreground as an important element in the portrayal of violence on television.

Historical Perspective of US Television Violence

Concerns about television violence have sparked intense debate since television's earliest days. There is general agreement that there is violence on television, but because of differences in the way violence is defined and measured, there is little agreement, and considerable controversy, about the degree or amount of violence (Lometti, 1995; Signorielli et al., 1982; Signorielli, Gerbner, & Morgan, 1995). But, even though these differences exist, there is ample evidence that violence is an ongoing and important element of television programming, particularly during the prime-time hours (Hetsroni, 2007; National Television Violence Study, 1997, 1998, 1999; Signorielli, 2003; Smith, Nathanson, & Wilson, 2002).

The earliest content analyses of television programming (Head, 1954; Smythe, 1954) coded programs “off the air” and found considerable violence in samples of New York City television programming. Violence appeared in two-thirds to three-quarters of all television plays at a rate of 6 to 10 incidents per hour in prime time and at rates three or four times as much in children's programming (mostly cartoons). Dominick (1973), in an analysis of all prime-time programs broadcast during February 1972, found that two-thirds of the programs contained some violence and that 60% of the violence could be categorized as assault, armed robbery, or murder. Unlike real life, violence by strangers was more frequent than violence by someone who was known to the victim. In another study of programming aired during the mid-1970s, Dominick (1978) noted that television presented violence from a law enforcement point of view, emphasized personal violence, ignored social aspects, did not present an adequate picture of the legal process, and did not provide accurate information about crime, criminals, and real-life violence. Similarly, Haney and Manzolati (1980) found that television crime and violence emphasized greed and other personal characteristics but rarely addressed underlying social conditions.

An analysis of television series with law enforcement or other violent themes broadcast between 1950 and 1976 (Taylor & Dozier, 1983) found that violence was systematically presented within a framework suggesting that people have an unquestioned moral and/or legal right to use violence, including deadly force, to protect the status quo. Finally, Greenberg, Edison, Korzenny, Fernandez-Collado, and Atkin's (1980) analysis of dramatic series in three seasons of programming in the mid-1970s found that violence (defined as “physical aggression”) occurred more than 9 times per hour between 8:00 and 9:00 p.m., more than 11 times per hour between 9:00 and 11:00 p.m., and more than 21 times per hour during Saturday morning children's programs.

During the 1970s, 1980s, and 1990s most of our knowledge of violence on television came from the Cultural Indicators (CI) Project. The NTVS contributed to our knowledge about television violence in the mid-1990s. The CI Project examined and measured the amount of physical violence on television by monitoring prime-time and weekend daytime network broadcast television programming and studying relationships between television viewing and conceptions of social reality (Gerbner, Morgan, & Signorielli, 1994; Gerbner, Gross, Morgan, & Signorielli, 1980a, 1980b, 1986; Gerbner & Signorielli, 1990; Morgan et al., 2009; Signorielli, 1990), periodically publishing the results as the violence profile.

One of the early reports of the CI Project looked at violence in prime-time and Saturday morning programs broadcast between 1967 and 1975 (Gerbner & Gross, 1976). This report found violence in more than 8 out of 10 programs with an average of 5 violent actions per program and 7 acts of violence per hour. Over half of the major characters were involved in violence, either committing violence, being a victim of violence, or both, and 1 in 10 major characters was involved in killing.

Another report (Gerbner et al., 1994) found, for samples of prime-time programs broadcast between 1973 and the fall of 1992, that violence appeared in 7 out of 10 programs at the rate of 5.3 incidents per hour and 4.6 incidents per program and that half of the major characters in these programs were involved in violence. Moreover, the figures reported for the samples of programs broadcast during the early 1990s were under the CI Project's 25-year averages. In the sample of programs from the 1992–1993 season, while 65% of prime-time fictional dramatic programs contained violence and 46% of the characters were involved in violence, the average frequency of violent acts was 2.9 per hour, about three-fifths of the 25-year average.

The NTVS (Smith et al., 1998; Wilson et al., 1997, 1998) examined physical violence in three yearly samples (1994–1995, 1995–1996, and 1996–1997) of three composite weeks of programming across 23 channels operating between 6:00 a.m. and 11:00 p.m. each day. The sample (N = 8,200) included broadcast (commercial networks, independent stations, and public television) and cable channels (basic and premium offerings). All genres except game shows, religious programs, “infomercials,” or home shopping channels, sports, instructional programs, and news were included (Smith et al., 1998).

The NTVS found similar levels of violence and no change in the prevalence of violence from the 1994–1995 to the 1996–1997 television seasons; 58% of the programs in the 1994–1995 sample, 61% of the programs in the 1995–1996 sample, and 61% of the programs in the 1996–1997 sample contained violence. There was, however, in prime time (8:00 p.m. to 11:00 p.m.), an 8% increase in the overall level of violence in cable and broadcast programs; 59% of the programs in the 1994–1995 sample, 66% of the programs in the 1995–1996 sample, and 67% of the programs in the 1996–1997 sample contained violence. The largest increase from 1994–1995 to 1996–1997 (14%) was found for those programs broadcast on the commercial networks. This study thus found in prime time, the time of day that draws the largest share of viewers, an increase in the percentage of programs with violence.

A more recent examination of the 1996–1997 NTVS data set (Smith et al., 2002) found that the amount of violence in prime-time programs was similar to that found in programs aired during other times of the day. Violence was found in 6 out of 10 programs and the rate of violent interactions in prime time was 6.6 per hour compared to 6.4 per hour during other day parts. This analysis found that 67% of broadcast network programs contained violence at a rate of 5.2 violent interactions per hour, figures similar to those for programs on basic cable. Premium cable programming, particularly the movie channels, had the most violence – 88% of the programs at 12.4 violent interactions per hour.

There were two other studies of television violence in the programming of the 1990s. An industry-funded study (ABC, CBS, NBC, Fox) monitored at least four episodes of every prime-time and Saturday morning (7:00 a.m. to noon) network program in each of three seasons (1994–1995, 1995–1996, and 1996–1997) and two weeks of programs on independent stations, public television, and pay cable (Cole, 1995, 1998). While this analysis did not provide overall measures of the level of violence in these samples between 1994–1995 and 1996–1997, there was a drop in the number of television series that raised frequent concerns about the way violence was presented. The only programs that raised more concerns at the end of this three-year period were reality shows (e.g., World's Most Dangerous Animals, World's Scariest Police Chases).

The Center for Media and Public Affairs (Lichter & Amundson, 1992) isolated physical violence on 10 channels (network, independent, and cable) during one day. Violence appeared most frequently during the afternoon (2:00 to 5:00 p.m.) with 19.1 acts per hour, in the early morning (6:00 to 9:00 a.m.) with 15.8 acts per hour, and during prime time (10.2 acts per hour). In an update, Lichter, Lichter, and Amundson (1999) isolated acts of violence in two randomly selected constructed weeks of prime-time network and cable fictional programs (N = 284) and 50 movies on cable and broadcast television during the 1998–1999 season. They found 12 acts of violence per episode (half were “serious”) in broadcast programs and 10 acts per episode (half were “serious”) in cable programs.

US Television Violence Before and After the Turn of the Century

The work of the CI Project has continued into the twenty-first century in the ongoing research of the author. One study (Signorielli, 2003) examined television violence in samples of fall network programming broadcast between 1993 and 2001. This analysis found that violence appeared in 61% of the programs at a rate of 4.5 acts of violence per program. The levels found in this analysis were similar to those found in earlier analyses in the CI perspective as well as those found in the NTVS. This analysis, however, found that fewer major characters were involved in violence. Compared to earlier CI analyses (see, e.g., Signorielli, 1990) when more than half of the characters were involved in violence, only one-third of all the characters at the turn of the century were involved in violence. Although more men than women were involved (38% of the men compared to 27% of the women), the proportions of Whites and minorities were almost equal.

Men were more likely to both commit violence and to be victimized – 27% of the men commit violence compared to 19% of the women and 27% of the men were victimized compared to 17% of the women. There were no differences in the percent of Whites and minorities who committed violence or who were victimized. Interestingly, during the 1990s the ratios of hurting to being hurt changed from the patterns seen in the 1970s and through the mid-1980s (Signorielli, 1990) for women but not for men. At the turn of the century, for every 10 male characters that hurt or killed other characters, 11 were victimized, the same ratio found in the earlier analysis. For women, however, instead of 16 women being victimized for each woman who hurt or killed, the odds were now even – women were equally likely to hurt or kill as be hurt or killed.

An analysis of violence in prime-time programs broadcast during the first decade of the twenty-first century was conducted for this chapter to bring our knowledge of the levels of violence in prime-time programming up to date. The methods used in this analysis are the same as those used in Signorielli (2003), extending that analysis to samples of network programming broadcast in the fall of each year between 2000 and 2009.¹ The variables used the same definitions in all of the CI-related research and meet all standards for reliability.

This new analysis shows that while some elements of the portrayal of violence have escalated in the first decade of the twenty-first century, others have declined. The percent of programs with violence, for example, increased from 61% of the prime-time programs broadcast in the 1990s to 69% of the programs in the decade after the turn of the century, a difference that is statistically significant (F = 4.21, df = 1,1395, p = .04). Violence, moreover, increased during the 2000s. There was a statistically significant increasing linear trend in the percent of programs with violence from the fall of 2000 to the fall of 2009 (F = 19.4, df = 8,774, p = .000). The percent of programs with violence rose from a low of 53% in the fall of 2000 to a high of 85% in the fall of 2008 but dropped to 79% in the fall of 2009. The number of violent actions, however, remained steady with an average of 4.6 acts of violence per program and no evidence of any increasing or decreasing trends.

There were differences in the involvement of characters in violence. Data from the first decade of the twenty-first century show that there were no statistically significant differences from the 1990s in the overall percent of characters involved in violence (36% of all characters). Again, proportionately more of the men (42%) than the women (28%) were involved in violence but, as seen in the 1990s, the percent of Whites and people of color involved in violence was essentially the same (about 25%). Although more men than women were involved in violence, both men and women were equally likely to hurt or kill others as be hurt or killed themselves. Interestingly, after the turn of the century, involvement in killing increased significantly for both men and women. In the last decade of the twentieth century 9% of the men and 5% of the women were involved in killing. By comparison, in the first decade of the twenty-first century, 12% of the men and 7% of the women were involved in killing, differences that are statistically significant (χ² = 20.54, df = 1, p = .00001 (N = 7,256)).

The Context of Violence

The NTVS elevated the importance of the context of violence in understanding how violence is seen on television. Two contextual elements in the early research of the CI Project were the appearance of humorous violence and the significance of violence in the plot or storyline. Some of the earliest research in the observational learning/social cognitive perspective (Bandura, 1986, 1990; Bandura, Ross, & Ross, 1961, 1963) along with other early studies of symbolic aggression (Lovas, 1961) found that youngsters often learn and reproduce violent behaviors seen in cartoons, i.e., humorous violence. At the same time, cultivation theory posits that humorous violence is an effective way to convey lessons of power. Although comic violence may appear less threatening, the actions are often mean-spirited with few realistic consequences, a sure formula for influence. These lessons may then translate to viewers developing conceptions of living in a mean and dangerous world and overestimating chances of being involved in violence by those who watch more TV (Gerbner et al., 1980a).

Cultivation theory also explains the significance of violence in the plot reflected in the complex social scenario illustrated by the patterns of committing violence and victimization in characterizations as well as the frequency of violence in the program. These elements ultimately cultivate a sense of fear, intimidation, and vulnerability reflected in the positive relationships between television viewing and scores on both the Mean World Index and the Index of Alienation and Gloom (Signorielli, 1990). In short, the more violence there is and the more important it is for the storyline, the more likely viewers believe that they live in a mean and dangerous world. Similarly, those who watch more television, particularly if they have been to college, are more likely to feel more bored, depressed, and lonely (Morgan, 1984).

Our knowledge about the context of violence comes from early studies of the CI Project that examined humor and program genre. During the 1970s slightly more than a quarter of prime-time programs were comic in nature and less than half (45%) included violence at a rate of 2.0 incidents per program and 3.6 incidents per hour. In addition, close to half of the network prime-time programs were action-adventures, an exceptionally violent genre, with 94% containing some violence at a rate of 7.8 incidents per program and 6.8 incidents per hour (Gerbner et al., 1980b). Similarly, Signorielli (1990) in an analysis through the fall 1985 sample found that only one in five prime-time programs had humorous violence.

The examination of graphic, immoral, intentional, and justified violence as well as the portrayal of the physical consequences of violence are also critical to the understanding of the context of television violence. Each of these elements is supported by several theories, including desensitization (Potter, 1999) and social learning/social cognitive theory (Bandura, 1986). Cultivation theory is also relevant because it explores how these elements relate to how different groups of viewers perceive their own vulnerability. Specifically, heavy viewers are more likely to believe they will be victims of violence and consequently more likely to buy more guns and locks as well as have watchdogs for protection (Gerbner, Gross, Morgan, Signorielli, & Jackson-Beeck, 1979).

The NTVS advanced our understanding of the contextual elements in the portrayal of violence on television in the mid-1990s (Smith et al., 1998; Wilson et al., 1997, 1998). The NTVS examined the consequences of violence, whether or not humor was involved, the graphic nature of the violence, whether or not weapons were used, and the degree of realism. The analysis of data from the 1994–1995 sample found that the context in which violence is presented posed risks for viewers (NTVS, Executive Summary, 1994–1995). In particular, three-quarters of the violent scenes had unpunished perpetrators, negative consequences of violence were rarely presented, one-quarter of the violent incidents involved the use of a handgun, and fewer than 1 in 20 programs emphasized antiviolence themes. Yet, television violence was not particularly graphic. While the analysis found that broadcast network programs had less violence than cable channels, the context of violence on both broadcast and cable was similar.

The NTVS also examined year-to-year changes in the portrayal of the contextual elements of violence. Looking specifically at violent broadcast programming, only 35% of the prime-time programs in 1994–1995, 23% of the programs in 1995–1996, and 24% in 1996–1997 had any long-term negative consequences of violence. At the same time, there was no display of remorse, regret, or sanctions in 6 out of 10 of the violent scenes in these samples. Similarly, while two-thirds of the violent interactions in the 1994–1995 sample did not show any pain as a result of violence, the proportion dropped to slightly more than half of the violent interactions (54% in the 1995–1996 sample and 53% in the 1996–1997 sample). This analysis indicated that violence on television, examined at the program, scene, and interaction level, was antiseptic and devoid of pain and suffering.

Smith and colleagues (2002) found that prime-time broadcast network programming and basic cable programming were less likely than premium cable programming to include violent interactions that depicted pain or harm. The violent interactions in premium cable programs also were more likely to show long-term consequences of violence than the network broadcast programs or the basic cable programs. Moreover, prime-time network broadcast programs were relatively devoid of pain and suffering.

Potter and Smith (2000), in an analysis of data from the second year of the NTVS (1995–1996), examined graphic portrayals of violence. This analysis found that most violence presented a low level of graphicness and that the violence in fantasy programs rarely exhibited graphicness. High levels were found in only 1 out of 10 violent actions. Moreover, it was presented with a high degree of realism. That is, the violence in live-action programs (recreated reality programs) had human targets and perpetrators. The use of guns and knives (shooting or stabbing a victim) was also related to higher levels of graphic violence.

Potter and colleagues' (1995) analysis of a composite week (6:00 p.m. to midnight) of programs broadcast on ABC, CBS, Fox, and NBC in the spring of 1994 also found that aggressive acts were typically presented in a context-less venue; fewer than one in six acts was depicted with any major consequences, only one in six acts was punished, while one-third were actually rewarded. While Potter and Ware (1987) found that only 1 in 10 acts of violence was punished and that heroes and villains were equally likely to commit antisocial acts, violence, at least from the perspective of the perpetrator, was seen as justified. Similarly, Lichter, Lichter, and Amundson (1999) found that most television violence did not have either psychological or physical consequences and occurred in a moral vacuum because heroes saw the violence they committed as justified (in self-defense, in a law enforcement context, etc.).

Signorielli's (2003) analysis of programming broadcast between 1993 and 2001 found that 39% of the programs did not have violence, that violence was an important element of the plot (significant or major) in 34% of the programs, and that 29% of the programs had violence that was not important to the story (minor to the plot). On the other hand, there were changes in the presentation of violence in the last 10 years. The analysis of data from 2000 to 2009 found that the percent of programs without violence decreased to 31% of the programs, while 43% of the programs had violence that was significant to the story and 26% had violence that was not particularly important to the story (minor plot element). Interestingly, there was a significant linear trend in the percent of programs with violence – in the fall of 2000 half of the programs (53%) had violence while in the fall of 2009 almost 8 out of 10 programs (79%) had violence.

Prime time is also more likely to have violence that is serious rather than comic in nature. At the end of the twentieth century, Signorielli (2003) found serious violence in 37% of the programs and comic violence in 18% of the programs. Once again, there were no trends in the appearance of this element of the context of violence. There was, however, some change in the portrayal of comic and serious violence at the start of the twenty-first century. Almost half of the programs broadcast after 2000 had violence judged to be serious in nature while only 13% had comic violence. In addition, while the percent of programs with comic violence did not change between the fall of 2000 and the fall of 2009, the percent of programs with serious violence increased during this time period from 40% of the programs at the first part of this decade to 60% of the programs at the end of the decade. (There is a statistically significant linear trend. F = 11.65, df = 8,774, p = .0007.)

Similar to the findings of the NTVS, Signorielli (2003) found that contextual elements did not appear very frequently during prime-time broadcast programs. More than 4 out of 10 programs had no context of violence. Violence that was intentional was found in 4 out of 10 programs, while justified violence, immoral violence, and showing the consequences of violence were found in about a third of the programs. Gratuitous and graphic violence occurred least often, appearing in a quarter of the programs. Overall, the average number of contexts was 1.8 per program, while programs with violence had an average of 2.9 contexts of violence per program. Moreover, there was a statistically significant linear trend. In short, the number of programs that included a context for violence in the story increased during the 1990s from 1.2 contexts per program in the spring of 1993 to 3.8 contexts per program in the fall of 2000.

The numbers of programs including contexts of violence in the storyline continued to increase during the first decade of the twenty-first century. Overall, only 34% of the programs in this decade compared to 44% of the programs in the last decade of the twentieth century had no context of violence. Interestingly, all of the contexts of violence (immoral violence, intentional violence, graphic violence, justified violence, and consequences of violence) except gratuitous violence increased. Specifically, almost half of the prime-time network programs broadcast after 2000 showed the context of violence compared to about a third of the prime-time network programs broadcast before 2000. Moreover, the average number of contexts per program increased from two contexts per program in 2000 to almost three contexts per program in the fall of 2009 (F = 9.71, df = 8,774, p = .002).

Signorielli's (2003) analysis of characters in programs broadcast during prime time in the 1990s found little information about the context in which characters commit violence. Only 13% of the men, 5% of the women, 9% of the Whites, and 9% of the characters of color who committed violence were either rewarded and/or punished for their behaviors. Similarly, 14% of the men, 11% of the women, 13% of the Whites, and 12% of the characters of color were presented as committing violence that was justified. Remorse was rarely found, exhibited by only 4% of the men, 3% of the women, 3% of the Whites, and 4% of the characters of color. And finally, only 9% of the men, 5% of the women, 7% of the Whites, and 7% of the characters of color were portrayed as having committed immoral violence. Overall, gender differences were statistically significant but race differences were not.

Between 2000 and 2009, although the percentages of characters shown in relation to a context of violence increased slightly, there was still little information about the context in which characters committed violence. Only 15% of the males, 8% of the females, 11% of the Whites, and 12% of the characters of color committed violence that was rewarded or punished. Men were a little more likely (17%) than women (12%) to commit justified violence while Whites and people of color were equally likely (14%) to commit justified violence. Again, few characters committed immoral violence (10% of the men, 5% of the women, 8% of the Whites, and 7% of the people of color), and even fewer characters (5% of the men, 3% of the women, and 4% of both Whites and people of color) exhibited any remorse for the violence they committed.

Violence in US Children's Programs

Children's cartoons have a considerable amount of violent content. One of the first studies of children's television found that over half of the children's programs monitored during late October 1960 contained violence that was an important part of the story (Schramm, Lyle, & Parker, 1961). Poulos, Harvey, and Liebert (1976), in an analysis of programs on two Saturday mornings during the 1974–1975 television season, found that three-quarters of the programs contained violence. Liss and Reinhardt (1980) found that Saturday morning network prosocial programs in the late 1970s were just as violent as regular cartoons. Similarly, Greenberg and colleagues (1980), in a study of Saturday morning programs broadcast between 1975 and 1978, found that cartoons contained more acts of violence than programs broadcast during the prime-time hours.

Studies in the CI tradition consistently found high levels of violence in samples of children's weekend-daytime programs broadcast in the fall of each year. For example, between 1967 and 1978, violence was found in 94% of these programs at a rate of 6 acts of violence per program, 18 acts of violence per hour, and involved 80% of the characters (Gerbner et al., 1979). The last CI analysis of children's weekend-daytime programs examined samples broadcast between the fall of 1973 and the fall of 1992 (Gerbner et al., 1994) and found violence in 92% of these programs at a rate of 7 acts per program, 23 acts per hour, and involved 80% of the characters.

The NTVS examined children's programming aired between October 1995 and June 1996 (Wilson, Smith et al., 2002). The analysis found violence in 69% of the programs targeted to children under the age of 13. Moreover, the violence was more concentrated in children's compared to regular programming. Violent incidents, interactions between perpetrators and targets (PATs), occurred 14 times an hour compared to 6 times an hour in non-children's programs. In addition, there were 7 scenes of violence per hour in children's programs compared to 3 per hour in non-children's programs. The amount of time devoted to violence, however, was similar for both children's and non-children's programs.

The NTVS analysis of prime-time programs also examined the context of violence in children's programs. Wilson, Smith, and colleagues (2002) found that violence was more sanitized and trivialized in children's programs. While more than half of the violent incidents involved lethal violence, the violence was rarely graphic. At the same time, a third of these violent scenes were rewarded and 8 out of 10 scenes had no type of punishment. Similarly, about two-thirds of the violent incidents showed no consequences of violence or physical injury to the victim, but three-quarters of the scenes presented violence in a humorous context. Moreover, the degree of violence was related to the type of children's program. Slapstick (Tom &Jerry), Superhero (Power Rangers), and adventure/mystery (Scooby Doo) programs had considerably more violence than programs focusing on social relationships (Care Bears) or magazine formats (Barney). Overall, Wilson, Smith, and colleagues (2002) concluded that violence in children's programs proposed considerable risks for young viewers.

There is not much information about the characters who are involved in violence in children's programs. Poulos and colleagues (1976) found that half of the violence committed in children's programs was by White males and half by animal (typically anthropomorphized) characters. Gerbner and colleagues (1994) found that more of the male characters than female characters in children's programs were involved in violence. Overall, more of the characters in children's programs were involved in violence and they were more likely to be victims of violence than to commit violence. Leaper and colleagues (2002) found that males exhibited more physical aggression and were victims of aggressive behavior in adventure and comic genres, while females were shown as fearful.

Similarly, the NTVS found that perpetrators in children's programs often were animals or other anthropomorphized creatures and that humans involved in violence were likely to be males. In addition, those who committed violence in these programs often were praised or even rewarded with material goods. At the same time, the victims were not portrayed realistically. Violence was sanitized so that victims did not exhibit pain or suffering and there were few signs of the types of physical injuries that would realistically be expected from the violence they suffered (Wilson, Smith et al., 2002).

A recent analysis of children's programs conducted in the CI tradition (Signorielli, 2008) examined 147 Saturday morning children's programs aired between 8:30 a.m. and 1:00 p.m. on ABC, CBS, NBC, Fox, CW, Cartoon Network, Nickelodeon, and Disney during the month of February 2007. Violence appeared in 86% of these programs at an overall rate of 7.0 acts of violence per program and 16.7 acts of violence per hour. Adventure programs (e.g., Teenage Mutant Ninja Turtles) were the most violent, with 98% having some violence and close to two-thirds with violence that was a significant part of the story. While 7 out of 10 teachy-preachy programs (e.g., Veggie Tales) were violent, most of this violence was not an integral part of the story. Half of the teen-scene programs (e.g., That's So Raven) had some violence but it rarely was an important part of the story. Once again, hourly rates of violence were high. While the entire sample had 17 acts of violence per hour, there were differences by genre. Adventure programs had 23 acts of violence per hour, teachy-preachy programs had 8 acts per hour, and the teen-scene programs had 4 acts per hour.

The violence in these children's programs was somewhat sanitized in that there was little, if any, graphic violence. At the same time, the violence was often presented as intentional and immoral. Characters' involvement in violence remained at high levels. The only major difference between today's programs and those of 30 years ago was that in 2007 the female characters had a greater likelihood of committing violence and were less likely portrayed as victims of violence.

Conclusion

Overall, we have not seen much change in the levels of violence in US prime-time network broadcast programs and in children's programs. From the 1950s to the first decade of the twenty-first century, violence has been an important and significant element of television programs. While, the overall amount of violence varies by genre (crime and action programs have more violence), it is still relatively difficult to escape watching violence. While the more diverse programming of the twenty-first century gives viewers opportunities to find nonviolent programs through “on demand” technologies, in order to totally avoid television violence a viewer must be adept in using these new technologies and take the time to implement them.

In short, the overall levels of violence have not changed and, if anything, have increased since the turn of the century. The overall numbers of major characters actually seen committing violence or being victims of violence, however, have diminished. Some of this may be due to the very popular crime-scene programs such as Law and Order or CSI, which focus on a crime that has already been committed and show, through the use of technologies (some of which may not really exist) (Houck, 2006), how crimes are solved and, for the most part, the perpetrators brought to justice.

At the same time, the portrayal of violence in programs of the first decade of the twenty-first century continues to show gender differences but few race-related differences. More men than women are involved in violence, primarily because men still outnumber women during prime time (Signorielli & Bacue, 1999). Moreover, within-gender distributions show that the proportion of men involved in violence (about 4 in 10) is considerably larger than the proportion of women involved in violence (about 1 in 4). Both men and women, however, are equally likely to commit violence (hurt or kill others) as to be victims of violence (be hurt or killed). The findings for gender distributions are similar to those of the NTVS (Smith et al., 1998) but smaller than Gerbner and colleagues' (1980a, 1980b, 1986, 1994) studies of programs from the 1970s and 1980s.

Another interesting difference is that the presentation of the context of violence has become somewhat more important. Through the 1990s and into the first decade of the twenty-first century, programs have begun to show that violence may have consequences, may be intentional in nature, and may be presented as justified in relation to the story. There is, however, little graphic or gratuitous violence, which, in turn, may reduce the amount of visual gore and thus show less realism. This has serious implications because less realistic images may result in viewers learning scripts of aggressive behavior and storing them for possible use in the future (Potter, 1999).

Similarly, even though today's programs may give more information about the context of violence overall, television may still not adequately support or reinforce the lesson that “crime does not pay.” Indeed, the context in which characters commit violence still does not appear very frequently and the characters who commit violence typically are not sorry for their actions, and may not be punished for their transgressions. From a social learning perspective these messages may result in viewers being more likely to learn and even accept aggressive behaviors. Thus, the environment of violent entertainment in which many people, particularly children, spend most of their free time may be potentially harmful for viewers (Smith et al., 1998). Moreover, as Potter and colleagues (1995) conclude, television's lack of realistic contexts for violence may signal that aggression and violence are acceptable. Thus, the long-term effects of watching television violence may result in a vicious cycle. Eron (1982), for example, has postulated that viewing leads to aggressive behavior and those who are more aggressive typically watch more violence.

Finally, as cultivation theory postulates, the ultimate long-term effects of watching television violence may post threats for civil liberties and freedom. Cultivation studies have found that those who watch more television, compared to those who watch less, are more likely to overestimate their chances of being involved in violence, believe that fear of crime is an important personal problem, and assume that crime is rising (when in actuality it is not). Those who spend more time watching television tend to believe that they are living in a mean and dangerous world as well as express feelings of alienation and gloom (Morgan et al., 2009; Signorielli, 1990). The problem is that today, even though there are more viewing choices, violent images are difficult to avoid; as a result, those who watch more television may become more fearful and alienated. The end result is that they may express sentiments of dependency and be willing to accept deceptively simple, strong, and hard-line political and religious postures if these beliefs seem to promise to relieve existing insecurities and anxieties. From the perspective of cultivation theory, the overall long-term effects of television violence may be the ready acceptance of repressive political and social environments that could translate into a loss of personal liberties.

NOTE

1 The analysis was not conducted in the fall of 2007 because the author was on sabbatical and did not teach the class in which this data collection/analysis has been conducted since 1993.

REFERENCES

Abelman, R. (1999). Preaching to the choir: Profiling TV advisory ratings users. Journal of Broadcasting and Electronic Media, 43, 529–550.

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

Bandura, A. (1990). Selective activation and disengagement of moral control. Journal of Social Issues, 46, 27–46.

Bandura, A. (2009). Social cognitive theory of mass communication. In J. Bryant & M. B. Oliver (Eds.), Media effects: Advances in theory and research (3rd ed., pp. 94–124). New York, NY: Routledge.

Bandura, A., Ross, D., & Ross, S. A. (1961). Transmission of aggression through imitation of aggressive models. Journal of Abnormal and Social Psychology, 63, 575–582.

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.

Berkowitz, L. (1970). Aggressive humor as a stimulus to aggressive responses. Journal of Personality and Social Psychology, 16, 710–717.

Berkowitz, L. (1984). Some effects of thoughts on anti- and prosocial influences of media events: A cognitive-neoassociation analysis. Psychological Bulletin, 95, 410–427.

Blair, E. (2009, December 31). In this decade, every room is a screening room. NPR. Retrieved from http://www.npr.org/templates/story/story.php?storyId=122068037

Bushman, B. J., Huesmann, L. R., & Whitaker, J. L. (2009). Violent media effects. In R. L. Nabi & M. B. Oliver (Eds.), The SAGE handbook of media processes and effects (pp. 361–376). Los Angeles, CA: Sage.

Cole, J. (1995). The UCLA television violence monitoring report. Los Angeles, CA: UCLA Center for Communication Policy.

Cole, J. (1998). The UCLA television violence monitoring report, year 3. Los Angeles, CA: UCLA Center for Communication Policy.

Dominick, J. R. (1973). Crime and law enforcement in prime-time television. Public Opinion Quarterly, 37(2), 241–250.

Dominick, J. R. (1978). Crime and law enforcement in the mass media. In C. Winick (Ed.), Deviance and mass media (pp. 105–128). Beverly Hills, CA: Sage.

Dustin, D. (1992, December 12). 3 networks agree to standards governing violence on television. Philadelphia Inquirer.

Eron, L. D. (1982). Parent–child interaction, television violence and aggression of children. American Psychologist, 37, 197–211.

Federal Communications Commission (FCC). (2000, September 15). Viewing television responsibly: The V-chip [V-chip home page]. Retrieved from http://www.fcc.gov/vchip/

Gandossy, T. (2009). TV viewing at “all-time high,” Nielsen says. CNN.com/Entertainment. Retrieved from http://www.cnn.com/2009/SHOWBIZ/TV/02/24/us.video.nielsen/

Gerbner, G., & Gross, L. (1976). Living with television: The violence profile. Journal of Communication, 26(1), 172–199.

Gerbner, G., Gross, L., Morgan, M., & Signorielli, N. (1980a). The “mainstreaming” of America: Violence profile no. 11. Journal of Communication, 30(3), 10–29.

Gerbner, G., Gross, L., Morgan, M., & Signorielli, N. (1980b). Violence profile no. 11: Trends in network television drama and viewer conceptions of social reality, 1967–1979. Philadelphia, PA: The Annenberg School of Communications, University of Pennsylvania.

Gerbner, G., Gross, L., Morgan, M., & Signorielli, N. (1986). Television's mean world: Violence profile no. 14–15. Philadelphia, PA: The Annenberg School of Communications, University of Pennsylvania.

Gerbner, G., Gross, L., Morgan, M., & Signorielli, N. (1994). Growing up with television: The cultivation perspective. In J. Bryant & D. Zillmann (Eds.), Media effects: Advances in theory and research (pp. 17–48). Hillsdale, NJ: Lawrence Erlbaum Associates.

Gerbner, G., Gross, L., Morgan, M., Signorielli, N., & Jackson-Beeck, M. (1979). The demonstration of power: Violence profile no. 10. Journal of Communication, 29(3), 177–196.

Gerbner, G., Morgan, M., & Signorielli, N. (1994). Television violence profile no. 14: The turning point. Philadelphia, PA: The Annenberg School of Communications, University of Pennsylvania.

Gerbner, G., & Signorielli, N. (1990). Violence profile 1967 through 1988–89: Enduring patterns. Newark, DE: Department of Communication, University of Delaware.

Greenberg, B. S., Edison, N., Korzenny, F., Fernandez-Collado, C., & Atkin, C. K. (1980). Antisocial and prosocial behaviors on television. In B. S. Greenberg (Ed.), Life on television: Content analyses of US TV drama (pp. 99–128). Norwood, NJ: Ablex.

Haney, C., & Manzolati, J. (1980). Television criminology: Network illusions of criminal justice realities. In E. Aronson (Ed.), Readings about the social animal. San Francisco, CA: W. H. Freeman.

Head, S. W. (1954). Content analysis of television drama programs. Quarterly of Film, Radio, and Television, 9, 175–194.

Hetsroni, A. (2007). Four decades of violent content on prime-time network programming: A longitudinal meta-analytic review. Journal of Communication, 57, 759–784.

Hoerrner, K. L. (1999). Symbolic politics: Congressional interest in television violence from 1950 to 1996. Journalism and Mass Communication Quarterly, 76(4), 684–698.

Houck, M. M. (2006). CSI: The reality. Scientific American, 295(1). Retrieved March 7, 2012, from http://www.scientificamerican.com/article.cfm?id=csi-the-reality

Huesmann, L. R. (1988). An information processing model for the development of aggression. Aggressive Behavior, 14, 13–24.

Kunkel, D., Farinola, W. J. M., Farrar, K., Donnerstein, E., Biely, E., & Zwarun, I. (2002). Deciphering the V-chip: An examination of the television industry's program rating judgments. Journal of Communication, 52(1), 112–138.

Leaper, C., Breed, L., Hoffman, L., & Perlman, C. A. (2002). Variations in the gender-stereotyped content of children's television cartoons across genres. Journal of Applied Social Psychology, 32(8), 1653–1662.

Lichter, S. R., & Amundson, D. (1992). A day of television violence. Washington, DC: Center for Media and Public Affairs.

Lichter, S. R., Lichter, L. S., & Amundson, D. R. (1999). Merchandizing mayhem: Violence in popular entertainment 1998–1999. Washington, DC: Center for Media and Public Affairs.

Liss, M. B., & Reinhardt, L. C. (1980). Aggression on prosocial television programs. Psychological Reports, 46, 1065–1066.

Lometti, G. E. (1995). The measurement of television violence. Journal of Broadcasting and Electronic Media, 39, 292–295.

Lovas, O. I. (1961). Effects of exposure to symbolic aggression on aggressive behavior. Child Development, 32, 37–44.

Morgan, M. (1984). Heavy television viewing and perceived quality of life. Journalism Quarterly, 61(3), 499–504, 740.

Morgan, M., Shanahan, J., & Harris, C. (1990). VCRs and the effects of television: New diversity or more of the same? In J. R. Dobrow (Ed.), Social and cultural aspects of VCR use (pp. 107–123). Hillsdale, NJ: Lawrence Erlbaum Associates.

Morgan, M., Shanahan, J., & Signorielli, N. (2009). Growing up with television: Cultivation processes. In J. Bryant & M. B. Oliver (Eds.), Media effects: Advances in theory and research (3rd ed., pp. 34–49). New York, NY: Routledge.

Mustonen, A., & Pulkkinen, L. (1997). Television violence: Development of a coding scheme. Journal of Broadcasting and Electronic Media, 41, 168–189.

National Television Violence Study, Executive Summary. (1994–1995). Studio City, CA: Mediascope.

National Television Violence Study 1. (1997). Thousand Oaks, CA: Sage.

National Television Violence Study 2. (1998). Thousand Oaks, CA: Sage.

National Television Violence Study 3. (1999). Thousand Oaks, CA: Sage.

Nielsen. (2009). Television, Internet and mobile usage in the US: A2/M2 three screen report, 4th quarter 2008. Retrieved from http://i.cdn.turner.com/cnn/2009/images/02/24/screen.press.b.pdf

Potter, W. J. (1999). On media violence. Thousand Oaks, CA: Sage.

Potter, W. J., & Smith, S. (2000). The context of graphic portrayals of television violence. Journal of Broadcasting and Electronic Media, 44(2), 301–323.

Potter, W. J., Vaughan, M. W., Warren, R., Howley, K., Land, A., & Hagemeyer, J. C. (1995). How real is the portrayal of aggression in television entertainment programming? Journal of Broadcasting and Electronic Media, 39(4), 496–516.

Potter, W. J., & Ware, W. (1987). An analysis of the contexts of antisocial acts on prime-time television. Communication Research, 14, 116–124.

Poulos, R. W., Harvey, S. E., & Liebert, R. M. (1976). Saturday morning television: A profile of the 1974–75 children's season. Psychological Reports, 42, 1047–1057.

Rideout, V. J., Foehr, U. G., & Roberts, D. F. (2010). Generation M²: Media in the lives of 8- to 18-year-olds. Menlo Park, CA: Kaiser Family Foundation.

Schramm, W., Lyle, J., & Parker, E. B. (1961). Television in the lives of our children. Stanford, CA: Stanford University Press.

Signorielli, N. (1990). Television's mean and dangerous world: A continuation of the Cultural Indicators perspective. In N. Signorielli & M. Morgan (Eds.), Cultivation analysis: New directions in media effects research (pp. 85–106). Newbury Park, CA: Sage.

Signorielli, N. (2003). Prime-time violence 1993–2001: Has the picture really changed? Journal of Broadcasting and Electronic Media, 47(1), 36–57.

Signorielli, N. (2005). Age-based ratings, content designations, and television content: Is there a problem? Mass Communication and Society, 8(4), 277–298.

Signorielli, N. (2008). Children' programs 2007: Basic demography and violence. Paper presented at the annual conference of the National Communication Association, San Diego, CA.

Signorielli, N., & Bacue, A. (1999). Recognition and respect: A content analysis of prime-time television characters across three decades. Sex Roles, 40(7/8), 527–544.

Signorielli, N., Gerbner, G., & Morgan, M. (1995). Violence on television: The Cultural Indicators Project. Journal of Broadcasting and Electronic Media, 39(2), 278–283.

Signorielli, N., Gross, L., & Morgan, M. (1982). Violence in television programs: Ten years later. In D. Pearl, L. Bouthilet, & J. Lazar (Eds.), Television and social behavior: Ten years of scientific progress and implications for the eighties (pp. 158–173). Rockville, MD: National Institute of Mental Health.

Smith, S. L., Nathanson, A. I., & Wilson, B. J. (2002). Prime-time television: Assessing violence during the most popular viewing hours. Journal of Communication, 52(1), 84–111.

Smith, S. L., Wilson, B. J., Kunkel, D., Linz, D., Potter, J., Colvin, C. M., & Donnerstein, E. (1998). Violence in television programming overall: University of California, Santa Barbara study. In National Television Violence Study 3 (pp. 5–195). Thousand Oaks, CA: Sage.

Smythe, D. W. (1954). Reality as presented on television. Public Opinion Quarterly, 18, 143–156.

Taylor, H., & Dozier, C. (1983). Television violence, African-Americans, and social control: 1950–1976. Journal of Black Studies, 14(2), 107–136.

Wilson, B. J., Colvin, C. M., & Smith, S. (2002). Engaging in violence on American television: A comparison of child, teen, and adult perpetrators. Journal of Communication, 52(1), 36–60.

Wilson, B. J., Kunkel, D., Linz, D., Potter, J., Donnerstein, E., Smith, S. L., . . . Berry, M. (1998). Violence in television programming overall: University of California, Santa Barbara study. In National Television Violence Study 2 (pp. 3–180). Thousand Oaks, CA; Sage.

Wilson, B. J., Kunkel, D., Linz, D., Potter, J., Donnerstein, E., Smith, S. L., . . . & Gray, T. (1997). Violence in television programming overall: University of California, Santa Barbara study. In National Television Violence Study 1 (pp. 3–158). Thousand Oaks, CA: Sage.

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

Woodward, F. H., IV (2000). Media in the home. Philadelphia, PA: The Annenberg Public Policy Center of the University of Pennsylvania.