Framing and Cultivating the Story of Crime

Abstract

The current study extended prior research by considering the effects of media, victimization, and network experiences on attitudes about crime and justice, drawing on the problem frame, cultivation, real-word, and interpersonal diffusion theses. Data were from a survey of Nebraska adults (n = 550) who were asked about their social networks; beliefs about media reliability; use of newspaper and news on TV, radio, and the Internet; and exposure to violence on TV, movies, and the Internet. Results indicated that viewing TV violence predicted worry and anger about crime. Believing the media are a reliable source of information about crime predicted more anger and more support for the justice system. Personal and network members’ victimization was also linked to attitudes. Other network contacts, including knowing police or correctional officers or knowing someone who had been arrested or incarcerated, had limited effects. The results support the problem frame and cultivation theses in that media framing and media consumption influence attitudes about crime, as do certain real-world experiences.

Keywords

media crime social networks

The role of media in the construction and proliferation of crime images has been illustrated across several types of media, particularly the news media. Research has generally demonstrated the effect of media consumption on public attitudes about crime and justice, such as misperceptions about the extent of crime (Lowry, Nio, & Leitner, 2003), fear (Chiricos, Padgett, & Gertz, 2000), anger (Johnson, 2009), and policy support (Rosenberger & Callanan, 2011). This research, however, has also revealed that the effects of media may be conditioned by factors like audience characteristics (Eschholz, Chiricos, & Gertz, 2003) or the genre of media examined (Kort-Butler & Sittner Hartshorn, 2011). Less research has given simultaneous attention to other personal factors that may influence an individual’s knowledge and attitudes about crime and justice issues. Individuals’ experiences with the criminal justice system, as well as the experiences and composition of their social networks, may also be influential (Pickett, Mancini, Mears, & Getz, 2015).

Accordingly, people’s knowledge or understanding about crime and justice comes from at least three sources: the media, personal experiences, and the experiences of others in their social networks. Despite this, the degree to which information derived from these sources impacts attitudes about crime and justice has not been fully explored. The current study provided insight into this issue by examining the effects of these three knowledge sources on attitudes about crime, including worry and anger about crime, and support for the criminal justice system.

In this article, we first consider how the media socially construct the crime problem and the implications of that imagery for people’s attitudes about crime and justice, highlighting the problem frame thesis (Altheide, 1997). Next, we outline the contributions of media consumption, personal experiences, and social network experiences to attitudes about crime. In doing so, we highlight the cultivation paradigm (Signorielli, Gerbner, & Morgan, 1995), the real-world thesis (Weitzer & Kubrin, 2004), the interpersonal diffusion thesis (Romer, Jamieson, & Aday, 2003), and the substitution and resonance theses (Eschholz et al., 2003). Using a data from a survey mailed to a random sample of Nebraska adults (analytic n = 550), the analyses explored how news consumption, exposure to media violence, personal victimization, victimization of social network members, employment of network members in the justice system, and involvement of network members in the justice system influenced respondents’ worries and anger about crime as well as their support for the criminal justice system. Results point to the complex yet overarching impact of media consumption on attitudes about crime and justice.

Media Images and the Crime Problem

Although empirical research is somewhat equivocal about whether the media react to or directly motivate public opinion about crime and justice (Frost, 2010), recent analyses demonstrate that on a broad level, reporting about crime and the tone of that reporting has influenced public punitiveness over time (Enns, 2016). The media are a means through which cultural images about crime are disseminated and reinforced as well as a means by which criminal justice policy debates are shaped (Barlow, Barlow, & Chiricos, 1995; Cavender, 2004). The media play a prominent role in socially constructing and shaping ideology about crime and its control in ways that generally uphold the status quo (Dotter, 2002; Surette, 1998). Most people have little to no direct experience with street crime; therefore, they must rely on other sources for information about crime, victimization, and the response of the justice system (Chermak, 1994; Surette, 2003).

The media are often considered the public’s primary frame of reference for issues of crime and control. The media focus public attention on certain types of criminal events and offer interpretations for how to understand them (Barak, 1994; Eschholz, 1997). The rarity and severity of crimes drive media presentations such that the characteristics of crime, criminals, and victims represented in the media are frequently the opposite of the pattern demonstrated in official crime statistics (Pollak & Kubrin, 2007) and tend to fit cultural stereotypes (Gruenewald, Chermak, & Pizarro, 2013). Furthermore, across various television genres, crime and justice issues are frequently framed by the ideologies supportive of the punitive crime control policies that emerged in the later part of the 20th century (Cavender & Fishman, 1998; Eschholz, Mallard, & Flynn, 2004; Welsh, Fleming, & Dowler, 2011).

The news media have also been central to the process of constructing the social reality of crime (Chiricos et al., 2000). Altheide (1997) argued that the process of problem framing in the news media acts as a primer on crime and justice by providing tools that consumers can use to interpret information and events. According to the problem frame thesis, the media inform the audience that some situation is undesirable, many people are affected by it, and the main contributing factors are identifiable. Further, the media employ “expert” commentaries who are most often political or criminal justice figures and rarely academic researchers (Buckler, Griffin, & Travis, 2008; Frost & Phillips, 2011). Such experts tend to reiterate existing narratives in detailing how the problem can be changed and what mechanisms exist to change it. They serve to remind the audience that we as a society already have agents and procedures in place to remedy the problem. This problem frame thus becomes a resource that the audience can use to interpret subsequent information about crime.

The discourse embedded in the problem frame has come to center on the malevolent individual uninhibited by social rules and moral values with whom law-abiding people cannot identify (Cavender, 2004; Surette, 1998), while emphasizing innocent or vulnerable victims (Bjornstrom, Kaufman, Peterson, & Slater, 2010; Lundman, 2003). Consequently, the discourse that pervades the media promotes anxiety about violent crime, while encouraging public reliance on existing structures for formal social control (Altheide & Michalowski, 1999; Green, 2009). Indeed, research has demonstrated that media consumption influences how people think about issues related to crime. Media consumption, across several genres, is linked to support for more punitive policies and support for law enforcement (Callanan & Rosenberger, 2011; Kleck & Jackson, 2016; Pickett & Barber, 2014; Rosenberger & Callanan, 2011).

Media Consumption, Personal Experiences, and Social Networks

The problem frame thesis and other insights into the social construction of crime in the media describe how the story of crime is crafted in such way as to shape, even distort, how people think and feel about crime. When it comes to crime and justice issues, constructed as they are to overemphasize extraordinary street crimes, criminals, and victims, research has frequently focused on fear, anxiety, or worry. Given the high levels of violence in American television, the cultivation thesis asserts that heavy television viewing engenders fear, mistrust, and perceptions that the world is a dangerous place (Gerbner, 1970; Signorielli et al., 1995). Early work with the Cultural Indicators Project confirmed the thesis (Gerbner & Gross, 1976).

As research has evolved, the relationship between television consumption and attitudes about crime, particularly fear, remains equivocal (Callanan & Rosenberger, 2015). Some studies have documented an association between television viewing and feeling fearful about crime, and others have not (Dowler, 2003; Doyle, 2006; Heath & Gilbert, 1996). The wide variation in samples, the media genre under investigation, and operationalization of fear or worry about crime, as well as model misspecification, likely contribute to inconsistency in empirical results (Ditton, Chadee, Farrall, Gilchrist, & Bannister, 2004; Morgan & Shanahan, 2010).

These complexities in the cultivation effect stimulated research in which audience characteristics and program characteristics were taken into consideration when modeling viewership effects (Eschholz, 2003). For instance, the strength of this relationship between news programming and fear of crime varies by viewer characteristics like gender and race/ethnicity (Chiricos, Eschholz, & Gertz, 1997; Chiricos et al., 2000). Consistent with iterations of the cultivation thesis that emphasize portrayals of violence (Shanahan, 2004), research on program characteristics suggests that programs with high levels of violence and realism, and which convey a sense of proximity to the viewer, are more closely associated with fear (Eschholz et al., 2003; Kort-Butler & Sittner Hartshorn, 2011). General media consumption may be less relevant than engaging with media more directly illustrative of crime.

In contrast to the cultivation perspective, the real-world thesis suggests attitudes about crime and justice issues are a function of objective conditions such as personal victimization and neighborhood disorder (Weitzer & Kubrin, 2004). Taking media use into consideration, personal victimization has been linked not only to fear of crime but also to lack of support for the criminal justice system and to the perception that the local crime rate is increasing or decreasing (Kort-Butler & Sittner Hartshorn, 2011; Weitzer & Kubrin, 2004). Moreover, cultivation effects may be meaningful for certain outcomes, whereas real-world experiences may influence other decisions. Custers and Van den Bulck (2011), for example, found that television viewing was associated with people feeling frightened about their risk for specific types of victimization, whereas prior victimization was associated with people feeling worried about the risks of being in a certain area. Understandably, personal victimization offers a perspective on crime that the media cannot.

Given that individuals are enmeshed in social networks, the experiences of others may be another way by which people gain real-world information about crime, which they weigh against both objective conditions and the broader mediated reality. According to the interpersonal diffusion thesis, people draw on the experiences of others in their networks in formulating their feelings about crime and justice (Romer et al., 2003). For example, Johnson (2009) noted that people with a relative or close friend incarcerated were less punitive. Personal experiences, as well as the experiences of others in an individual’s social network, essentially may make crime issues more salient as well as provide for a firsthand or secondhand account of how the justice system functions. Taken together, the real-world thesis and interpersonal diffusion thesis suggest that the salience of crime for an individual, be it drawn from personal experience or experiences of others in one’s social network, may be more relevant than media consumption for feelings and attitudes about crime (Frost, 2010; Kleck & Jackson, 2016).

Few studies have examined media consumption and social network variables simultaneously. Rosenberger and Callanan (2011) observed, controlling for personal victimization and fear of crime, that having a household member arrested was tied to greater support for rehabilitative policies, whereas media consumption was associated with greater support for punitive policies. In a separate study, Callanan and Rosenberger (2011) found a relationship between having a household member arrested and having less confidence in the police and being less likely to believe the police treat people fairly. Theory and research thus suggest that personal and social network factors should be considered alongside media factors in order to understand people’s attitudes about justice issues more fully.

People also bring their personal dispositions to their media usage, which influences what crime-related content may mean for them (Sacco, 1995). According to the substitution thesis, the effects of media should be more robust for people who have little direct experience with crime or the justice system, in that the media act as a window to crime. In contrast, the resonance thesis suggests that media portrayals of crime resonate more with people who have experience with the criminal justice system and for whom crime is already a salient issue (Eschholz et al., 2003). For example, Chiricos, Padgett, and Gertz (2000) found the impact of local news was stronger for people with recent victimization experiences. In contrast, Pickett, Mancini, Mears, and Getz (2015) found that individuals who had social network contacts with criminal justice experiences were no more reliant on the media for information about crime than those without contacts. Research remains unsettled on whether viewers who have limited experience with crime in fact substitute mediated experiences when forming attitudes about crime or whether mediated experiences resonate more strongly with people who have actual experience with crime (Callanan & Rosenberger, 2015).

The Current Study

According to Morgan and Shanahan (2010), the cultivation perspective emphasizes how the media offer us social lessons about what “the world is like” (p. 343), yet Morgan and Shanahan acknowledge how other personal experiences and social influences shape individuals’ perceptions of that reality. For people who have limited non-media sources of information about crime, the media may be more relevant for defining the socially constructed reality of crime (Surette, 2003). On the other hand, the more salient crime is to an individual, be it through personal experiences or through connections to others in their social networks, media consumption may be less relevant for beliefs about crime and justice (Frost, 2010). Little research, however, has taken both perspectives into consideration (Kleck & Jackson, 2016). Accordingly, the current study extended prior research by examining the effects of three knowledge sources on attitudes about crime and justice.

The study focused on worry about crime, anger about crime, and support for the justice system. As described above, in research examining the media and real-world effects, fear or worry about crime and support for various aspects of the justice system are commonly examined outcomes. However, fear may not be the only, or even the most important, emotion people draw on when considering criminal justice issues (Hartnagel & Templeton, 2012). For example, anger at the thought of crime may be a more common reaction than fear (Ditton, Bannister, Gilchrist, & Farrall, 1999). Anger about crime has been linked to support for punitive policies, controlling for fear (Hartnagel & Templeton, 2012; Johnson, 2009). Anger may be expressive of insecurities brought on by social change or by perceived threats to the moral order posed by crime, both of which may be communicated via the media (Karstedt, 2002; Lyons & Scheingold, 2000). Anger may also be an understandable reaction to experiencing crime firsthand or vicariously via social networks. Research remains equivocal about the source of anger regarding crime.

First, relying on the problem frame thesis (Altheide, 1997), we hypothesized that perceiving the media as a reliable source of crime information would increase fear and anger about crime, while at the same time enhancing support for the justice system. Although people do not fully trust local and national news outlets, more than three quarters report trusting them some or a lot (Mitchell, Gottfried, Barthel, & Shearer, 2016). The degree to which people trust the media or find it to be reliable as a source of information may be tied to their acceptance of the problem frame.

Second, drawing on insights from the cultivation thesis (Morgan & Shanahan, 2010), we explored different aspects of media use. We examined the relationship between news consumption and attitudes about crime, hypothesizing that routine news consumption would increase worry and anger but decrease support for the justice system. We also expected higher levels of exposure to violent media content to increase worry and fear but negatively impact support for the justice system.

Although news consumption, in general, may be associated with anxieties about crime and control (Chiricos et al., 2000), research suggests that the types of programs may matter (Kort-Butler & Sittner Hartshorn, 2011) as well as the nature of the violent content (Eschholz et al., 2003). We considered several sources of news, including television, newspapers, Internet, and radio. We also assessed exposure to media violence across different sources, including television news, TV shows, movies, and the Internet. Although newspapers and television have frequently been examined, few studies to date have investigated the use of online sources (Britto & Noga-Styron, 2014; Roche, Pickett, & Gertz, 2016). A recent Pew Research Center report noted that about 57% of adults often get news from television, and more than one third often get news online, outpacing radio (25%), and print newspapers (20%; Mitchell et al., 2016). As Roche, Pickett, and Gertz (2016) argued, Internet news reaches an ever-wider audience, yet people can be selective in the news they view, be it through the filters of their social networking sites or through the news-oriented sites they visit. Unlike television, newspapers, and radio, which act to create and sustain the problem frame by dishing up only certain kinds of content, Internet news may allow users more latitude in framing their own stories about crime.

Third, drawing on insights from the real-world thesis (Weitzer & Kubrin, 2004) and the interpersonal diffusion thesis (Romer et al., 2003), we also considered victimization experiences and network characteristics (Kleck & Jackson, 2016). When it came to victimization experiences, we hypothesized that both personal and network victimization experiences would contribute to worry and anger about crime but decrease support for the justice system. Based on prior research, we posited that people who have criminally involved network members may be less supportive of the justice system. However, because they know someone who is criminally involved, they may be less likely to experience the worry or anger about crime borne of media stereotypes. We also posited that people who have network members employed in criminal justice would be more supportive of the justice system. Our analyses for worry and anger were exploratory. On the one hand, we might expect people who know a criminal justice employee to feel less worried or angry, because they are more knowledgeable about how the system operates. On the other hand, we might expect them to feel more worried or angry if they are concerned about the network member’s safety or if that network member relates frightening experiences.

Finally, both the substitution thesis and the resonance thesis emphasize potential interactions between media use and people’s crime-related experiences, albeit in different directions. The nature and direction of those interactions remains unresolved (Callanan & Rosenberger, 2015). Our analyses regarding these interaction effects were therefore exploratory. If the effects of media use on attitudes are stronger for people who did not have experience with crime or the justice system, this could be evidence for substitution. If media effects are stronger for people who do have experience, it could be evidence for resonance.

Method

Sample

Data were drawn from the Nebraska Community Survey conducted by mail over the spring of 2014 (Habecker, Dombrowski, & Khan, 2015). The survey was a random selection of 2,000 Nebraska households. The list of addresses was drawn from the U.S. Postal Services Delivery Sequence File, which offers high rates of coverage for household populations (Iannacchione, 2011). The mailing schedule adopted practices outlined by Dillman, Smyth, and Christian (2014). The “next birthday” method was used to select the person over 19 in the household to complete the survey. The final response rate was 31%, yielding a sample size of 618. Eliminating surveys that were largely incomplete and following listwise deletion, the analytic sample was 550. The sample was 59% female, 9% non-White, with an average age of about 54. Half of the sample lived in an area defined as urban, and 45% had at least a bachelor’s degree (see Table 1).

Table 1.
Descriptive Statistics.

Variables Mean Standard Deviation Range

Age 54.08 16.81 20/92

Female 0.59 0.49 0/1

Non-White 0.09 0.29 0/1

Education 0.45 0.50 0/1

Urban 0.50 0.50 0/1

Conservative 2.26 0.91 0/4

Media reliability 2.27 0.76 1/4

TV news viewing 0.67 0.47 0/1

Newspaper 0.48 0.50 0/1

Web news 0.49 0.50 0/1

Radio news 0.43 0.50 0/1

Violence on TV news 0.63 0.48 0/1

Violence on TV shows 0.57 0.50 0/1

Violence in movies 0.72 0.45 0/1

Violence on the Internet 0.41 0.49 0/1

Employee in network 2.63 2.83 0/10

Involvement in network 0.69 1.24 0/5

Personal victimization 0.08 0.27 0/1

Victimization in network 0.45 0.50 0/1

Worry about crime 2.49 0.78 1/5

Angry about crime 2.64 0.87 1/4

Support for the justice system 2.27 0.57 1/4

Measures

Dependent variables

Worry about crime was assessed with six questions regarding how often respondents personally worried about victimization, including worry about walking alone at night, getting robbed, having their residence broken into, being sexually assaulted, getting murdered, and having someone in their family become a victim of a crime. Response options ranged from 1 (never) to 5 (almost always). Responses to the 6 items were averaged into an overall indicator of worry. Cronbach’s α equaled .83. This scale is similar to that used in Gallup polls and other research (Ditton et al., 2004).

Anger about crime was measured with 2 items that asked respondents to rate how angry they felt about crime in the country and in their communities (Johnson, 2009). Response options ranged from 1 (not at all angry) to 4 (very angry). The items were highly correlated (r = .75, p < .001; see Appendix Table A1), so the items were averaged to create one measure of anger.

Finally, support for the justice system was assessed with two dimensions measuring attitudes about the justice system (Kort-Butler & Sittner Hartshorn, 2011). The first dimension, confidence, was measured with 4 items. Respondents were asked to rate how confident they were that the criminal justice system could reduce crime and drug use, and how confident they were that the police could protect society from violent and property crimes. Response options ranged from 1 (not at all confident) to 4 (very confident). The second dimension, fairness, was measured with 3 items that asked respondents to rate how fair the justice system was in its treatment of offenders, treatment of victims, and its use of the death penalty. Responses ranged from 1 (not at all fair) to 4 (very fair). Exploratory factor analysis indicated that these 7 items loaded onto a single factor. An overall indicator of support for the criminal justice system was created by averaging responses to the items, with a Cronbach’s α of .71.

Independent variables

The survey asked about three aspects of media use. The problem frame thesis suggests the extent to which viewers trust the information transmitted by the media may influence how they interpret that information (Kort-Butler & Sittner Hartshorn, 2011). Media reliability was based on one survey question, “How reliable is the media as a source of information about crime?” The response categories ranged from 1 (not at all reliable) to 4 (very reliable).

News consumption, a central component of both the problem frame thesis and the cultivation thesis, was based on a series of questions that asked how many days in the past 7 days the respondent read a print newspaper, read or watched news on the Internet, listened to news on the radio, watched local TV news, watched national network news, and watched cable TV news (ranging from 0 to 7 days). The distributions suggested a cutoff point that could be considered frequent but not daily consumption across the different news sources. Each variable—newspaper, Internet, radio, and television—was then constructed as a dichotomous indicator in which the category represented accessing the news source on 5 or more days. The television items were correlated (r = .35, p < .001), suggesting that people who watched local news frequently also watched national news of both genres frequently, so these were collapsed.

Extensions of the cultivation thesis emphasize the impact of violent program content. Media violence exposure was based on a series of questions that asked how often in the past 7 days the respondent saw violent acts in the news, on a TV program other than news, in a movie, and on the Internet. Response categories ranged from 0 (never) to 3 (every day). Few respondents reported never seeing violence on TV news or programs. These 2 items were dichotomized so that 0 = never or some days and 1 = almost every day or every day. The movie and Internet items were distributed differently, with a larger proportion of people reporting “never,” likely reflecting the more selective nature of these sources. These 2 items were coded so that 0 = never and 1 = at least some days.

The survey inquired about several aspects of people’s social networks, a key element of the real-world thesis and the interpersonal diffusion hypothesis. On the survey, “knowing” a person was defined as “you know them and they know you by sight or name, that you could contact them, and that there has been some contact (either in person, by telephone, mail, or web) in the past 2 years” (McCarty, Killworth, Bernard, Johnsen, & Shelley, 2001). This approach captures the outermost layer of personal contacts, which is considered an individual’s “active network” characterized by strong and weak ties (Roberts, Dunbra, Pollet, & Kuppens, 2009). Weak ties, although not always important, are indispensable to an individual’s opportunities and integration into communities (Granovetter, 1983). The measurement of networks in this study does not imply close personal connections to all ties; rather, it suggests that there are relationships with the potential for meaningful associations with how respondents view their worlds.

Respondents were asked to report how many people they knew in Nebraska who were police officers and how many they knew who were correctional officers. Criminal justice employee in network was a summed variable that added the number of police officers and number of correctional officers. Given the skewed distribution, the resulting variable was collapsed, so that 0 = 0, 1 = 1, and so on, then 8 = 8 through 10, 9 = 11 through 15, and 10 = 16 or more people the respondent knew who worked in the justice system

Respondents were also asked about a variety of criminal behaviors and justice involvement among people they knew. This variable focused on arrests and convictions in the previous year. Respondents were asked how many people they knew who had been arrested and how many they knew who had been convicted of a crime. Criminal justice involvement in network was a summed variable that added the number of arrests and the number of convictions. Again, the resultant variable was collapsed, so that 0 = 0, 1 = 1, 2 = 2, 3 = 3 through 4, 4 = 5 through 10, and 5 = 11 or more people the respondent knew who were involved with the justice system.

Another network factor considered in the analysis was others’ victimization experiences. Respondents reported whether they knew someone who had been a victim of burglary, assault, robbery, sexual assault, or homicide in the past year. Victimization in network was a dichotomous variable where knowing no one victimized was coded 0 and knowing at least one person victimized by one of these crimes was coded 1. Personal victimization was also considered as an element of the real-world thesis. Personal victimization was a dichotomous variable in which respondents answered yes or no to a question about whether they had been a victim of crime in the last year. The variable was coded so that 1 equaled yes.

Control variables

The analyses controlled for several factors that may influence perceptions of media reports and attitudes about crime. Audience characteristics may affect the relationship between media viewing and attitudes about crime and justice (Chiricos et al., 1997). Accordingly, the analyses controlled for respondents’ demographic characteristics, including age, education, race, sex, residential location, and political ideology. Age was reported as a continuous variable. Education was a dummy variable with a bachelor’s degree and higher coded as 1 and less than a bachelor’s degree as the reference category. Race was a dummy variable with non-White coded as 1 and White as the reference category.¹ Sex was a dummy variable with female coded as 1. Respondents were asked to indicate their zip codes. The dummy variable urban was coded so that 1 equaled residence in an area with more than 50,000 people; residence in an area with a smaller population was the reference category. The respondents’ political conservatism was assessed with an item that asked respondents to rate themselves politically using a 5-point scale. Following prior research on public attitudes toward the criminal justice system (Unnever, Benson, & Cullen, 2008), the variable was coded, so that 0 was very liberal and 4 was very conservative.

Analysis Plan

To determine the effects of each knowledge source on attitudes about crime and justice, the media variables, network variables, and victimization variables were entered into separate ordinary least squares (OLS) regression models for each independent variable, following similar research (Callanan & Rosenberger, 2015; Roche et al., 2016). Then, to fully assess the study hypotheses about the effects of each knowledge source, all variables were entered into the final models for worry about crime, anger about crime, and support for the justice system. Because the substitution and resonance hypotheses suggest that media knowledge of crime may be less or more relevant for people who have network or personal encounters with crime, interaction terms between the media variables and the other substantive variables were also calculated and entered into the full regression models.

Results

Table 1 displays the descriptive statistics for the sample. On average, respondents found the media only somewhat reliable in providing information on crime. About two thirds of the sample viewed television news at least 5 times per week. Of the other sources, 47.5% of respondents read a newspaper, 49% accessed web-based news, and 43.5% listened to radio news at least 5 days a week. Across media platforms, people frequently saw violence. About 63% of people reported seeing violence almost or every day on TV news and about 57% reported seeing violence routinely on TV shows. Nearly three quarters of respondents reported seeing violence in movies on at least some days, whereas only 40.5% reported seeing violence on the Internet.

With regard to their social networks, 70% of people knew at least one person employed as a police or corrections officer in the state.² On average, respondents had about three people in their networks who were employed in the justice system. Nearly 30% of respondents knew someone who had been arrested or convicted in the prior year; on average, respondents knew about one person. In terms of victimization, 45% of people knew someone who had been victimized by a serious crime in the prior year, whereas only 8% of respondents had been personally victimized.

There was a high degree of correlation among the news consumption variables (see Appendix Table A1), particularly TV news and the other sources, suggesting that people who routinely access news from one source were likely to do so from another. In order to isolate the direct effect of each news source, each source and the control variables were regressed on each outcome. With the control variables considered, the only significant relationship occurred between TV news use and worry about crime (see Appendix Table A2). The relationship between TV news and anger was nearly significant (b = 0.17, p = .052). No other news source—newspapers, radio, and Internet—had a significant relationship with worry, anger, or support for the justice system. For the remaining analyses, only TV news viewing was included in the models.

In the next stage of the analysis, worry about crime, anger about crime, and support for the justice system were regressed on the knowledge sources and the control variables. Table 2 presents the regressions on worry. In Model 1, people who frequently saw violence on TV news and on the Internet expressed more worry about crime. TV news consumption and media reliability were not significant. In Model 2, neither network employment nor network involvement in the criminal justice system was significant. In Model 3, both network and personal victimization were positively related to worry. Finally, Model 4 included all knowledge sources. Violence on TV news, network victimization, and personal victimization remained significant predictors of worry. Violence on the Internet was marginally significant (p = .07). Among the control variables, female respondents, urban dwellers, and political conservatives were more worried about crime.

Table 2.
Ordinary Least Squares Regression on Worry about Crime.

Model 1 Model 2 Model 3 Model 4

β (SE) β (SE) β (SE) β (SE)

Media reliability .00 (.04) — — .01 (.04)

TV news viewing .05 (.07) — — .05 (.07)

Violence on TV news .15 (.07)* — — .14 (.07)*

Violence on TV shows .07 (.07) — — .06 (.07)

Violence in movies .03 (.07) — — .03 (.07)

Violence on the Internet .09 (.07)* — — .08 (.07)

Employee in network — .05 (.01) — .01 (.01)

Involvement in network — .06 (.03) — −.02 (.03)

Personal victimization — — .14 (.11)* .12 (.11)

Victimization in network — — .13 (.06)** .10 (.07)*

Age .00 (.00) −.02 (.00) .01 (.00) .02 (.00)

Female .39 (.06)* .41 (.06)* .41 (.06)* .40 (.06)*

Non-White −.02 (.11) −.03 (.11) −.03 (.10) −.03 (.11)

Education −.05 (.06) −.05 (.06) −.05 (.06) −.04 (.06)

Urban .08 (.06) .11 (.07)** .09 (.06)* .08 (.07)*

Conservative .11 (.03)** .10 (.03)* .10 (.03) .11 (.03)

F 12.54* 14.86* 18.60* 10.83*

Adj. R ² .21 .17 .20 .23

Note. Standardized coefficients presented. SE = standard error.

p < .05. p < .01. p ≤ .001.

Table 3 displays the regressions on anger about crime. In Model 1, media reliability, seeing violence on TV news, and seeing violence on TV shows were positively associated with anger. Internet violence was marginally significant (p = .06). News consumption was not significant. In Model 2, having criminal justice employees in one’s network positively predicted anger, but knowing someone involved with the justice system did not. In Model 3, knowing someone victimized by crime was positively associated with anger; personal victimization was not significant. Finally, Model 4 included all knowledge sources. Considering the media a reliable source of information about crime was positively related to anger about crime, as was exposure violence on TV news and in TV shows, and having criminal justice employees in one’s network. Internet violence was marginally significant (p = .07), but the effect of network victimization was not significant once all other knowledge sources were considered. In addition, conservatism was positively related to anger about crime, and female respondents, respondents with less education, and urban dwellers reported more anger.

Table 3.
Ordinary Least Squares Regression on Anger About Crime.

Model 1 Model 2 Model 3 Model 4

β (SE) β (SE) β (SE) β (SE)

Media reliability .12 (.05) — — .12 (.05)

TV news viewing .03 (.09) — — .02 (.09)

Violence on TV news .13 (.08) — — .12 (.08)

Violence on TV shows .11 (.08)* — — .09 (.08)*

Violence in movies .06 (.09) — — .05 (.08)

Violence on the Internet .09 (.08) — — .08 (.08)

Employee in network — .16 (.01)* — .13 (.01)

Involvement in network — .07 (.03) — .01 (.03)

Personal victimization — — .01 (.14) −.03 (.13)

Victimization in network — — .12 (.08) .06 (.08)

Age .14 (.00) .12 (.00) .12 (.00) .15 (.00)

Female .10 (.07)* .12 (.07) .11 (.07) .11 (.07)**

Non-White −.04 (.13) −.04 (.13) −.04 (.13) −.05 (.13)

Education −.10 (.08)* −.10 (.08)* −.10 (.08)* −.10 (.08)*

Urban .09 (.08)* .12 (.08) .06 (.08) .12 (.08)

Conservative .15 (.04)*** .10 (.04)* .12 (.04) .14 (.04)*

F 6.86* 6.33* 4.70* 6.11*

Adj. R ² .12 .07 .05 .13

Note. Standardized coefficients presented. SE = standard error.

p < .05. p < .01. p ≤ .001.

Table 4 shows the regressions on support for the justice system. In Model 1, considering the media a reliable source of crime information was positively associated with support for the justice system, whereas frequent viewing of violence on TV shows was negatively associated with support. In Model 2, knowing people involved with the justice system was negatively associated with support, but knowing criminal justice employees was not significant. In Model 3, personal victimization was negatively associated with support, but network victimization was not significant. Finally, Model 4 included all knowledge sources. Media reliability was positively related to support and personal victimization was negatively related to support. Controlling for all knowledge sources, violence on TV news (p = .08) and network involvement (p = .08) were reduced to marginal levels of significance. Additionally, age was inversely associated with support, and education level and political conservatism were positively related to support for the justice system.

Table 4.
Ordinary Least Squares Regression on Support for Criminal Justice System.

Model 1 Model 2 Model 3 Model 4

β (SE) β (SE) β (SE) β (SE)

Media reliability .43 (.03)* — — .43 (.03)*

TV news viewing .05 (.05) — — .05 (.05)

Violence on TV news −.01 (.05) — — .00 (.05)

Violence on TV shows −.09 (.05)* — — −.07 (.05)

Violence in movies −.03 (.05) — — −.03 (.05)

Violence on the Internet −.04 (.05) — — −.02 (.05)

Employee in network — .01 (.01) — .03 (.01)

Involvement in network — −.12 (.02) — −.08 (.02)

Personal victimization — — −.11 (.09) −.13 (.08)**

Victimization in network — — −.11 (.05)* −.04 (.05)

Age −.10 (.00)* .01 (.00) −.01 (.00) −.13 (.00)

Female −.02 (.05) −.05 (.05) −.05 (.05) −.02 (.05)

Non-White .00 (.08) .02 (.09) .01 (.08) .02 (.08)

Education .12 (.05) .09 (.05) .09 (.05)* .11 (.05)*

Urban −.02 (.05) −.02 (.05) −.01 (.05) −.02 (.05)

Conservative .13 (.02)*** .10 (.03)* .10 (.03)* .13 (.02)

F 11.89* 2.48* 3.78* 10.26*

Adj. R ² .20 .02 .04 .22

Note. Standardized coefficients presented. SE = standard error.

p < .05. p < .01. p ≤ .001.

For the last stage of the analysis, several multiplicative interaction terms were created between the media variables (i.e., TV news, sources of violent media, and media reliability) and each of the other knowledge sources. Then, these interaction terms were entered as blocks into the full model for each outcome. Models with significant interaction terms are presented in Table 5, with the caveat that of the many interaction terms that were tested, only a handful were significant. Model 1 includes interaction terms between TV news viewing and the other knowledge sources regressed on worry about crime. Only the TV news consumption by network involvement term is significant, indicating that knowing someone who was involved in the criminal justice system amplified the effect of TV news consumption on worry about crime. Model 2 includes interaction terms between violence on TV shows and the other knowledge sources regressed on anger about crime; two terms were significant. First, having network members involved in the criminal justice system muted the effect of TV show violence exposure on anger about crime. Second, and in contrast, the effect of TV show violence exposure on anger about crime was more relevant for people who had a crime victim in their network. Model 3 includes interaction terms between violence on TV shows and the other knowledge sources regressed on support for the justice system; interaction terms with victimization were significant. Both personal and network victimization experiences muted the impact of seeing violence on TV shows on support for the justice system.

Table 5.
Ordinary Least Squares Regression on Outcomes With Interaction Effects.

Worry Angry Support for Criminal Justice System

b (SE) b (SE) b (SE)

Media reliability .01 (.04) .13 (.05) .32 (.03)*

TV news viewing −.08 (.11) .02 (.09) .06 (.05)

Violence on TV news .23 (.07)* .23 (.08) .00 (.05)

Violence on TV shows .10 (.07) .18 (.11) −.01 (.07)

Violence in movies .04 (.07) .09 (.08) −.05 (.05)

Violence on the Internet .12 (.07) .16 (.08) −.01 (.05)

Employee in network .01 (.02) .07 (.02)** .00 (.01)

Involvement in network −.10 (.05)* .12 (.05)* .01 (.03)

Personal victimization .44 (.18)* .01 (.23) −.53 (.14)***

Victimization in network .06 (.11) −.15 (.12) .100 (.08)

TV News × Employed .00 (.02) — —

TV News × Involvement .14 (.06)* — —

TV News × Personal Victimization −.17 (.23) — —

TV News × Network Victimization .16 (.14) — —

Violent Shows × Employed — −.04 (.03) .01 (.02)

Violent Shows × Involvement — −.16 (.07)* −.07 (.04)

Violent Shows × Personal Victimization — −.12 (.28) .42 (.17)*

Violent Shows × Network Victimization — .40 (.16)* −.23 (.10)*

F 9.31* 5.54* 9.134***

Adj. R ² .24 .15 .24

Note. Unstandardized coefficients presented. All models include controls. Only models with significant interaction terms shown. SE = standard error.

p < .05. p < .01. p ≤ .001.

Discussion

Media representations are the primary way in which many Americans learn about and interpret issues of crime and justice, even as the cultural line between lived experiences and virtual landscapes continues to move (Dotter, 2002). However, people may also come to understand crime through their own experiences and through the experiences of others in their social networks. The current study examined the influence of these different knowledge sources on attitudes about crime and justice. Taking all knowledge sources into consideration, media variables and victimization were the most robust predictors of attitudes about crime and justice.

First, perceived media reliability had a substantial impact on anger and support for the justice system. The more reliable people thought the media were as a source of information about crime, the angrier they felt about crime, yet they expressed support for the justice system. This result is consistent with the problem frame thesis: if viewers “buy in” to the media’s skewed presentation of crime, it elicits an emotional response while still leading them to endorse the status quo when it comes to the justice system (Altheide & Michalowski, 1999; Dotter, 2002). By comparison, frequent news consumption itself, regardless of source, was unrelated to negative emotions about crime and support for the justice system when all media variables were considered. What is reported in the news may be less relevant than people’s trust that the information reported is reliable, and how it is characterized.

Second, exposure to violence on television news and shows promoted worry and anger about crime, consistent with our expectations derived from the version of the cultivation thesis that highlights program characteristics (Eschholz et al., 2003). Addressing a key extension to the cultivation paradigm suggested by Morgan and Shanahan (2010), in this study, exposure to violence was considered across several media genres, including TV news, TV programs other than news, movies, and the Internet. Movie viewing had no effect, but Internet viewing did have marginally significant effects. It is worth noting here that the survey items did not specifically distinguish between broadcast TV shows, streaming services (e.g., Netflix), and web other than streaming programs (e.g., YouTube videos), so respondents may have some overlap in their viewing experiences. As our opportunities for viewing violence in the media evolve, this study indicates the cultivation paradigm continues to be relevant for understanding the complex relationship between exposure to media violence and negative emotions about crime, including anger.

Interestingly, exposure to media violence had limited direct effects on support for the justice system. One explanation for this inconsistency may itself be tied to cultivation. Whereas fear and anger in response to media violence tap into symbolic concerns about violence and general feelings of insecurity, consumption of such materials may not generate rationalized reactions, such as support for existing justice paradigms (Elchardus, De Groof, & Smits, 2008). Research has also demonstrated variation in the portrayal of the justice system across media types, with some genres reflecting justice agents as achieving their goals of stopping criminals and other genres emphasizing corruption or ineptitude (Cavender & Deutch, 2007; Kort-Butler & Sittner Hartshorn, 2011; Phillips & Strobl, 2006). The media violence measures, which covered a range of genres, may capture viewers’ overall exposure to a more complex—and less reassuring—model of justice. As a result, viewing media violence intensified respondents’ worry and anger about crime but did not directly shape their feelings about the justice system.

Taken together, these results suggest multifaceted effects of media on attitudes about crime and justice. Exposure to media violence is more consistent with cultivation effects, whereas reliance on the media for information about crime is more consistent with the problem frame thesis, which focuses largely on news and the information presented therein. In short, the impact of media on attitudes about crime and justice likely includes elements of both the cultivation paradigm and the problem frame thesis.

Third, beyond the direct effects of media, victimization experiences also played a role in attitudes about crime and justice. Personal victimization increased worry about crime and decreased support for the justice system. The results are consistent with our expectations derived from the real-world hypothesis and with prior research that considers media exposure and personal victimization (Custers & Van den Bulck, 2011; Kort-Butler & Sittner Hartshorn, 2011; Weitzer & Kubrin, 2004).

Fourth, there was limited support for the interpersonal diffusion thesis. In contrast to prior research (Johnson, 2009; Pickett et al., 2015; Rosenberger & Callanan, 2011), the social network characteristics—knowing people employed by or involved in the justice system—had little direct impact on attitudes about crime and justice. Similar to Buckler, Wilson, Hartley, and Davila (2011), network involvement in the justice system was generally unrelated to support for the justice system. Network employment directly contributed to an increase in anger, and network victimization increased worry about crime. Such network experiences may further enhance crime salience for individuals, heightening their personal concerns about crime (Frost, 2010), but additional research is warranted.

Finally, the substitution and resonance theses posit that personal experiences (or lack thereof) interact with media consumption in ways to either enhance or diminish the effect of media on attitudes, respectively. There was limited evidence for either of these perspectives, as only a handful of interaction terms were statistically significant. The most consistent interactions were between viewing violence on TV shows and the victimization measures, offering tentative evidence for the resonance thesis. The resonance thesis suggests that media use is more relevant for the attitudes of people who have more experience with crime because they are more attentive to media-based information about crime (Pickett et al., 2015). For people who knew someone victimized by crime, seeing violence on TV shows amplified their anger about crime but diminished their support for the justice system. Likewise, for respondents who had been victimized, viewing violence on TV shows further weakened their support for the justice system. Further research is needed, but this pattern suggests media portrayals of violence may resonate with those who have intimate knowledge of victimization.

In addition, this study was among the few studies that explicitly examine anger about crime as a dependent variable. To extent that the media construct the world as a violent place where criminals are out for themselves (Morgan and Shanahan, 2010) and victims are sympathetically displayed (Gruenewald et al., 2013), and to the extent people accept that frame, they may see crime as a signal to the loss of moral principles (Karstedt, 2002). The sense of moral disgust engendered by the media may also be stimulated when network members who work in the justice system recount “what went wrong at work today” or when someone they know is victimized. Anger about crime is an expression of that disgust, and it may be channeled into support for more punitive responses to crime (Hartnagel & Templeton, 2012; Johnson, 2009). The implication is that the media, and to some extent network experiences, feed people’s fears and fuel their outrage about crime.

This study was also among the few studies that explicitly examine Internet sources (Britto & Noga-Styron, 2014; Roche et al., 2016). Like those studies, Internet-based news was unrelated to the outcome variables, while Internet-based violence had only marginal effects on worry and anger. On the one hand, web-based news may overlap substantially with more traditional news sources, giving the appearance of more choices yet with the same content (Callanan & Rosenberger, 2011). On the other hand, people can be selective in accessing Internet content, be it news, programming, or social media, giving them latitude in framing their own stories about crime. If people are indeed creating their own stories, simple measures of media use cannot capture them nor can such measures account for the role of confirmation biases in what content becomes part of their stories.

Limitations

There were methodological limitations to this study. First, the survey sampled from one state, which is not necessarily representative of other populations. Second, although most survey techniques have their limitations, mail surveys conducted in the mode used here are considered more effective in reaching people than phone surveys (Dillman et al., 2014). Nonetheless, younger people may be less responsive to the survey, which could impact media usage reports in particular. Third, people in states with higher degrees of urbanization, racial diversity, political diversity, or objectively higher crime rates may have more diverse social networks as well as different experiences with and perceptions of crime. Because of their lower proximity and exposure to crime relative to other states (e.g., Nebraska ranked 33rd in violent crime rates in 2014; FBI, 2015), Nebraskans may be more dependent on the media as a window for viewing crime and justice, so that the media may be a more salient source. Future research should consider other locations and nationally representative samples to explore this topic in greater detail. Fourth, the mean age of the sample, about 54 years, may be biased toward TV use. Mitchell, Gottfried, Barthel, and Shearer (2016) noted that 72% of 50- to 64-year-olds and 85% of those 65 and older often get news from TV, whereas about half of 18- to 49-year-olds often get news online. As research on media effects moves forward, more sophisticated measures are needed to understand who uses what kind of medium, for what purpose, and to what end.

Fifth, the network measures could be refined. For example, beyond police and correctional officers, people may know other actors in the justice system (e.g., lawyers and probation officers). The measure used in this study intentionally captured family, friends, and acquaintances. However, how frequently people actually discuss crime and justice issues with those in their networks and the closeness of those relationships, unmeasured in this study, may be important for developing one’s point of view.

Sixth, the media measures could also be refined. The media reliability measure did not ask about specific sources, but other research illustrates people are more trusting of local and national news organizations than they are of social media (Mitchell et al., 2016). Although television news may routinely deal with crime, the more selective nature of the other news sources means that people can avoid reading or seeing crime-related content. Asking respondents how they access news and whether they looked at crime content may yield greater precision in future work. The study also did not incorporate the range of programming types that present messages about crime and justice, such as those presented as infotainment (Kort-Butler & Sittner Hartshorn, 2011). Recent research also suggests that social media play an increasingly important role in shaping beliefs about crime (Elsass, Schildkraut, & Stafford, 2014). Where social networks and media intertwine, the lines between the real and the virtual when it comes to crime may be further blurred. An examination of social media consumption and social networks may be a fruitful area for future research on attitudes about crime and justice.

Finally, the cross-sectional nature of the survey does not allow for statements of causality. For instance, people who are already fearful or angry about crime may be more likely to seek out violent media, which reinforce their concerns. Due to the historical and ongoing saturation of crime in the media, it may be difficult to establish definitively whether media consumption has a causal impact on attitudes about criminal justice. Likewise, given both the static (e.g., family members) and fluid (e.g., acquaintances) nature of social networks, it may be difficult to determine their causal impact. Future researchers could consider longitudinal designs or social network analysis to assess the directionality and specificity of the relationships observed here.

Conclusions

Given these limitations, this study demonstrated, albeit conservatively, that media consumption mattered when it came to worry and anger about crime and support for the criminal justice system. These results provided evidence for the cultivation thesis as it relates to program characteristics that emphasize violence. The results also provided support for the problem frame thesis in that people’s trust in the media when it came to crime intensified their anger about crime but nonetheless reinforced their support for the criminal justice system. Personal victimization, as well as victimization of others in one’s network, also influenced attitudes about crime, suggesting that what happens to people in the realworld matters, and those effects may be intensified by viewing violence in the media. As the media continue to evolve, research should continue to explore why people view crime-related programming or access crime-related information, and how they interpret such information in light of their experiences and the experiences of others in their social networks. Ultimately, such research can lead to a better understanding of how people gain knowledge about crime and develop or shift their attitudes about crime and justice issues.

Variables	Mean	Standard Deviation	Range
Age	54.08	16.81	20/92
Female	0.59	0.49	0/1
Non-White	0.09	0.29	0/1
Education	0.45	0.50	0/1
Urban	0.50	0.50	0/1
Conservative	2.26	0.91	0/4
Media reliability	2.27	0.76	1/4
TV news viewing	0.67	0.47	0/1
Newspaper	0.48	0.50	0/1
Web news	0.49	0.50	0/1
Radio news	0.43	0.50	0/1
Violence on TV news	0.63	0.48	0/1
Violence on TV shows	0.57	0.50	0/1
Violence in movies	0.72	0.45	0/1
Violence on the Internet	0.41	0.49	0/1
Employee in network	2.63	2.83	0/10
Involvement in network	0.69	1.24	0/5
Personal victimization	0.08	0.27	0/1
Victimization in network	0.45	0.50	0/1
Worry about crime	2.49	0.78	1/5
Angry about crime	2.64	0.87	1/4
Support for the justice system	2.27	0.57	1/4

	Model 1	Model 2	Model 3	Model 4
Media reliability	.00 (.04)	—	—	.01 (.04)
TV news viewing	.05 (.07)	—	—	.05 (.07)
Violence on TV news	.15 (.07)***	—	—	.14 (.07)***
Violence on TV shows	.07 (.07)	—	—	.06 (.07)
Violence in movies	.03 (.07)	—	—	.03 (.07)
Violence on the Internet	.09 (.07)*	—	—	.08 (.07)
Employee in network	—	.05 (.01)	—	.01 (.01)
Involvement in network	—	.06 (.03)	—	−.02 (.03)
Personal victimization	—	—	.14 (.11)***	.12 (.11)**
Victimization in network	—	—	.13 (.06)**	.10 (.07)*
Age	.00 (.00)	−.02 (.00)	.01 (.00)	.02 (.00)
Female	.39 (.06)***	.41 (.06)***	.41 (.06)***	.40 (.06)***
Non-White	−.02 (.11)	−.03 (.11)	−.03 (.10)	−.03 (.11)
Education	−.05 (.06)	−.05 (.06)	−.05 (.06)	−.04 (.06)
Urban	.08 (.06)	.11 (.07)**	.09 (.06)*	.08 (.07)*
Conservative	.11 (.03)**	.10 (.03)*	.10 (.03)**	.11 (.03)**
F	12.54***	14.86***	18.60***	10.83***
Adj. R ²	.21	.17	.20	.23

	Model 1	Model 2	Model 3	Model 4
Media reliability	.12 (.05)**	—	—	.12 (.05)**
TV news viewing	.03 (.09)	—	—	.02 (.09)
Violence on TV news	.13 (.08)**	—	—	.12 (.08)**
Violence on TV shows	.11 (.08)*	—	—	.09 (.08)*
Violence in movies	.06 (.09)	—	—	.05 (.08)
Violence on the Internet	.09 (.08)	—	—	.08 (.08)
Employee in network	—	.16 (.01)***	—	.13 (.01)**
Involvement in network	—	.07 (.03)	—	.01 (.03)
Personal victimization	—	—	.01 (.14)	−.03 (.13)
Victimization in network	—	—	.12 (.08)**	.06 (.08)
Age	.14 (.00)**	.12 (.00)**	.12 (.00)**	.15 (.00)
Female	.10 (.07)*	.12 (.07)**	.11 (.07)**	.11 (.07)**
Non-White	−.04 (.13)	−.04 (.13)	−.04 (.13)	−.05 (.13)
Education	−.10 (.08)*	−.10 (.08)*	−.10 (.08)*	−.10 (.08)*
Urban	.09 (.08)*	.12 (.08)**	.06 (.08)	.12 (.08)**
Conservative	.15 (.04)***	.10 (.04)*	.12 (.04)**	.14 (.04)***
F	6.86***	6.33***	4.70***	6.11***
Adj. R ²	.12	.07	.05	.13

	Model 1	Model 2	Model 3	Model 4
Media reliability	.43 (.03)***	—	—	.43 (.03)***
TV news viewing	.05 (.05)	—	—	.05 (.05)
Violence on TV news	−.01 (.05)	—	—	.00 (.05)
Violence on TV shows	−.09 (.05)*	—	—	−.07 (.05)
Violence in movies	−.03 (.05)	—	—	−.03 (.05)
Violence on the Internet	−.04 (.05)	—	—	−.02 (.05)
Employee in network	—	.01 (.01)	—	.03 (.01)
Involvement in network	—	−.12 (.02)**	—	−.08 (.02)
Personal victimization	—	—	−.11 (.09)**	−.13 (.08)**
Victimization in network	—	—	−.11 (.05)*	−.04 (.05)
Age	−.10 (.00)*	.01 (.00)	−.01 (.00)	−.13 (.00)**
Female	−.02 (.05)	−.05 (.05)	−.05 (.05)	−.02 (.05)
Non-White	.00 (.08)	.02 (.09)	.01 (.08)	.02 (.08)
Education	.12 (.05)**	.09 (.05)	.09 (.05)*	.11 (.05)*
Urban	−.02 (.05)	−.02 (.05)	−.01 (.05)	−.02 (.05)
Conservative	.13 (.02)***	.10 (.03)*	.10 (.03)*	.13 (.02)**
F	11.89***	2.48*	3.78***	10.26***
Adj. R ²	.20	.02	.04	.22

	Worry	Angry	Support for Criminal Justice System
Media reliability	.01 (.04)	.13 (.05)**	.32 (.03)***
TV news viewing	−.08 (.11)	.02 (.09)	.06 (.05)
Violence on TV news	.23 (.07)***	.23 (.08)**	.00 (.05)
Violence on TV shows	.10 (.07)	.18 (.11)	−.01 (.07)
Violence in movies	.04 (.07)	.09 (.08)	−.05 (.05)
Violence on the Internet	.12 (.07)	.16 (.08)	−.01 (.05)
Employee in network	.01 (.02)	.07 (.02)**	.00 (.01)
Involvement in network	−.10 (.05)*	.12 (.05)*	.01 (.03)
Personal victimization	.44 (.18)*	.01 (.23)	−.53 (.14)***
Victimization in network	.06 (.11)	−.15 (.12)	.100 (.08)
TV News × Employed	.00 (.02)	—	—
TV News × Involvement	.14 (.06)*	—	—
TV News × Personal Victimization	−.17 (.23)	—	—
TV News × Network Victimization	.16 (.14)	—	—
Violent Shows × Employed	—	−.04 (.03)	.01 (.02)
Violent Shows × Involvement	—	−.16 (.07)*	−.07 (.04)
Violent Shows × Personal Victimization	—	−.12 (.28)	.42 (.17)*
Violent Shows × Network Victimization	—	.40 (.16)*	−.23 (.10)*
F	9.31***	5.54***	9.134***
Adj. R ²	.24	.15	.24

Footnotes

Appendix

Table A2.

Ordinary Least Squares Regression of News Consumption on Worry About Crime.

	Model 1	Model 2	Model 3	Model 4	Model 5
	β (SE)	β (SE)	β (SE)	β (SE)	β (SE)
TV news viewing	.10 (.07)*	—	—	—	.11 (.08)*
Newspaper	—	.00 (.07)	—	—	−.03 (.07)
Web news	—	—	.00 (.06)	—	−.03 (.06)
Radio news	—	—	—	−.02 (.06)	−.03 (.06)
Age	−.07 (.00)	−.03 (.00)	−.03 (.00)	−.03 (.00)	−.07 (.00)
Female	.40 (.06)***	.40 (.06)***	.41 (.06)***	.40 (.06)***	.39 (.06)***
Non-White	−.02 (.11)	−.02 (.11)	−.02 (.11)	−.02 (.11)	−.03 (.11)
Education	−.06 (.06)	−.06 (.06)	−.06 (.06)	−.06 (.06)	−.05 (.07)
Urban	.09 (.06)*	.09 (.06)*	.09 (.06)*	.10 (.06)*	.10 (.06)*
Conservative	.10 (.03)*	.10 (.03)*	.10 (.03)*	.11 (.03)**	.10 (.03)*
F	17.05***	16.07***	16.17***	15.93*	11.60***
Adj. R ²	.17	.16	.16	.16	.16

Note. Standardized coefficients presented. SE = standard error.

*p < .05. **p < .01. ***p < .001.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Notes

References

Altheide

D. L.

(1997). The news media, the problem frame, and the production of fear. Sociological Quarterly, 38, 647–668.

Altheide

D. L.

Michalowski

R. S.

(1999). Fear in the news: A discourse in control. Sociological Quarterly, 40, 475–503.

Barak

(1994). Media, society, and criminology. In Barak

(Ed.), Media, process, and the social construction of crime: Studies in newsmaking criminology (pp. 3–48). New York, NY: Garland.

Barlow

M. H.

Barlow

D. E.

Chiricos

T. G.

(1995). Mobilizing support for social control in a declining economy: Exploring ideologies of crime within crime news. Crime and Delinquency, 41, 191–204.

Bjornstrom

E. E.

Kaufman

R. L.

Peterson

R. D.

Slater

M. D.

(2010). Race and ethnic representations of lawbreakers and victims in crime news: A national study of television coverage. Social Problems, 57, 269–293.

Britto

Noga-Styron

K. E.

(2014). Media consumption and support for capital punishment. Criminal Justice Review, 39, 81–100.

Buckler

Griffin

Travis

L. F.

(2008). Criminologists as benchwarmers: The exclusion of egghead from crime news discourse. Southwest Journal of Criminal Justice, 5, 4–21.

Buckler

Wilson

Hartley

Davila

(2011). Racial and ethnic perceptions of injustice: Does prior personal and vicarious incarceration experience alter the racial/ethnic gap in perceptions of injustice? Criminal Justice Review, 36, 269–290.

Callanan

V. J.

Rosenberger

J. S.

(2015). Media, gender, and fear of crime. Criminal Justice Review, 40, 322–339.

10.

Callanan

V. J.

Rosenberger

J. S.

(2011). Media and public perceptions of the police: Examining the impact of race and personal experience. Policing & Society, 21, 167–189.

11.

Cavender

(2004). Media and crime policy: A reconsideration of David Garland’s the culture of control. Punishment and Society, 6, 335–348.

12.

Cavender

Deutch

S. K.

(2007). CSI and moral authority: The police and science. Crime, Media, Culture, 3, 67–81.

13.

Cavender

Fishman

(1998). Television reality crime programs: Context and history. In Fishman

Cavender

(Eds.), Entertaining crime (pp. 1–15). New York, NY: Aldine de Gruyter.

14.

Chermak

(1994). Crime in the news media: A refined understanding of how crimes become news. In Barak

(Ed.), Media, process, and the social construction of crime: Studies in newsmaking criminology (pp. 95–129). New York, NY: Garland.

15.

Chiricos

Eschholz

Gertz

(1997). Crime, news and fear of crime: Toward an identification of audience effects. Social Problems, 44, 342–357.

16.

Chiricos

Padgett

Gertz

(2000). Fear, TV news, and the reality of crime. Criminology, 38, 755–786.

17.

Custers

Van den Bulck

(2011). The relationship of dispositional and situational fear of crime with television viewing and direct experience with crime. Mass Communication and Society, 14, 600–619.

18.

Dillman

D. A.

Smyth

J. D.

Christian

L. M.

(2014). Internet, phone, mail, and mixed-mode surveys: The tailored design method. Hoboken, NJ: John Wiley.

19.

Ditton

Bannister

Gilchrist

Farrall

(1999). Afraid or angry? Recalibrating the ‘fear’ of crime. International Review of Victimology, 6, 83–99.

20.

Ditton

Chadee

Farrall

Gilchrist

Bannister

(2004). From imitation to intimidation a note on the curious and changing relationship between the media, crime and fear of crime. British Journal of Criminology, 44, 595–610.

21.

Dotter

(2002). Creating deviance: Scenarios of stigmatization in postmodern media culture. Deviant Behavior, 23, 419–448.

22.

Dowler

(2003). Media consumption and public attitudes toward crime and justice: The relationship between fear of crime, punitive attitudes, and perceived police effectiveness. Journal of Criminal Justice and Popular Culture, 10, 109–126.

23.

Doyle

(2006). How not to think about crime in the media. Canadian Journal of Criminology and Criminal Justice, 48, 867–885.

24.

Elchardus

De Groof

Smits

(2008). Rational fear or represented malaise: A crucial test of two paradigms explaining fear of crime. Sociological Perspectives, 51, 453–471.

25.

Elsass

H. J.

Schlidkraut

Stafford

M. C.

(2014). Breaking news of social problems: Examining media consumption and student beliefs about school shootings. Criminology, Criminal Justice, Law and Society, 15, 31–42.

26.

Enns

P. K.

(2016). Incarceration nation. New York, NY: Cambridge University Press.

27.

Eschholz

(1997). The media and fear of crime: A survey of the research. Journal of Law and Public Policy, 9, 37–59.

28.

Eschholz

(2003). Crime on television: Issues in criminal justice. Journal of the Instituted of Justice and International Studies, 2, 9–18.

29.

Eschholz

Chiricos

Gertz

(2003). Television and fear of crime: Program types, audience traits, and the mediating effect of perceived neighborhood racial composition. Social Problems, 50, 395–415.

30.

Eschholz

Mallard

Flynn

(2004). Images of prime time justice: a content analysis of NYPD Blue and Law & Order. Journal of Criminal Justice and Popular Culture, 10, 161–180.

31.

Federal Bureau of Investigation (FBI). (2015). Crime in the United States. Accessed March 1, 2017 from https://ucr.fbi.gov/crime-in-the-u.s/2015/crime-in-the-u.s.-2015

32.

Frost

N. A.

(2010). Beyond public opinion polls: Punitive public sentiment & criminal justice policy. Sociology Compass, 4, 156–168.

33.

Frost

N. A.

Phillips

N. D.

(2011). Talking heads: Crime reporting on cable news. Justice Quarterly, 28, 87–112.

34.

Gerbner

(1970). Cultural indicators: The case of violence in television drama. Annals of the American Academy of Political and Social Science, 388, 69–81.

35.

Gerbner

Gross

(1976). Living with television: The violence profile. Journal of Communication, 26, 172–194.

36.

Granovetter

(1983). The strength of weak ties: A network theory revisited. Sociological Theory, 1, 201–233.

37.

Green

D. A.

(2009). Feeding wolves: Punitiveness and culture. European Journal of Criminology, 6, 517–536.

38.

Gruenewald

Chermak

S. M.

Pizarro

J. M.

(2013). Covering victims in the news: What makes minority homicides newsworthy? Justice Quarterly, 30, 755–783.

39.

Habecker

Dombrowski

Khan

(2015). Improving the network scale-up estimator: Incorporating means of sums, recursive back estimation, and sampling weights. PLoS One, 10, e0143406. doi:10.1371/journal.pone.0143406

40.

Hartnagel

T. F.

Templeton

L. J.

(2012). Emotions about crime and attitudes to punishment. Punishment & Society, 14, 452–474.

41.

Heath

Gilbert

(1996). Mass media and fear of crime. American Behavioral Scientist, 39, 379–386.

42.

Iannacchione

V. G.

(2011). The changing role of address-based sampling in survey research. Public Opinion Quarterly, 75, 556–575.

43.

Johnson

(2009). Anger about crime and support for punitive criminal justice policies. Punishment & Society, 11, 51–66.

44.

Karstedt

(2002). Emotions and criminal justice. Theoretical Criminology, 6, 299–317.

45.

Kleck

Jackson

D. B.

(2016). Does crime cause punitiveness? Crime & Delinquency. doi:0011128716638503

46.

Kort-Butler

L. A.

Sittner Hartshorn

K. J.

(2011). Watching the detectives: Crime programming, fear of crime, and attitudes about the criminal justice system. Sociological Quarterly, 52, 36–55.

47.

Lowry

D. T.

Nio

T. C. J.

Leitner

D. W.

(2003). Setting the public fear agenda: A longitudinal analysis of network TV crime reporting, public perceptions of crime, and FBI crime statistics. Journal of Communication, 53, 61–73.

48.

Lundman

R. J.

(2003). The newsworthiness and selection bias in news about murder: Comparative and relative effects of novelty and race and gender typifications on newspaper coverage of homicide. Sociological Forum, 18, 357–386.

49.

Lyons

Scheingold

(2000). The politics of crime and punishment. In LaFree

(Ed.), The nature of crime: Continuity and change, criminal justice 2000 (Vol. 1, pp. 103–149). Washington, DC: U.S. Department of Justice.

50.

McCarty

Killworth

P. D.

Bernard

H. R.

Johnsen

E. C.

Shelley

G. A.

(2001). Comparing two methods for estimating network size. Human Organization, 60, 28–39.

51.

Mitchell

Gottfried

Barthel

Shearer

(2016). The modern news consumer: News attitudes and practices in the digital era. Pew Research Center. Retrieved December 10, 2016, from http://www.journalism.org/2016/07/07/the-modern-news-consumer/

52.

Morgan

Shanahan

(2010). The state of cultivation. Journal of Broadcasting & Electronic Media, 54, 337–355.

53.

Phillips

N. D.

Strobl

(2006). Cultural criminology and kryptonite: Apocalyptic and retributive constructions of crime and justice in comic books. Crime, Media, Culture, 2, 304–331.

54.

Pickett

J. T.

Baker

(2014). The pragmatic American: Empirical reality or methodological artifact? Criminology, 52, 195–222.

55.

Pickett

J. T.

Mancini

Mears

D. P.

Gertz

(2015). Public (mis)understanding of crime policy: The effects of criminal justice experience and media reliance. Criminal Justice Policy Review, 26, 500–522.

56.

Pollak

Kubrin

C. E.

(2007). Crime in the news: How crimes, offenders and victims are portrayed in the media. Journal of Criminal Justice and Popular Culture, 14, 59–83.

57.

Roberts

S. G.

Dunbar

R. I.

Pollet

T. V.

Kuppens

(2009). Exploring variation in active network size: Constraints and ego characteristics. Social Networks, 31, 138–146.

58.

Roche

S. P.

Pickett

J. T.

Gertz

(2016). The scary world of online news? Internet news exposure and public attitudes toward crime and justice. Journal of Quantitative Criminology, 32, 215–236.

59.

Romer

Jamieson

K. H.

Aday

(2003). Television news and the cultivation of fear of crime. Journal of Communication, 53, 88–104.

60.

Rosenberger

J. S.

Callanan

V. J.

(2011). The influence of media on penal attitudes. Criminal Justice Review, 36, 435–455.

61.

Sacco

(1995). Media constructions of crime. Annals of the American Academy of Political and Social Science, 539, 141–154.

62.

Shanahan

(2004). A return to cultural indicators. Communications, 29, 277–294.

63.

Signorielli

Gerbner

Morgan

(1995). Standpoint: Violence on television: The cultural indicators project. Journal of Broadcasting and Electronic Media, 39, 278–283.

64.

Surette

(1998). Media, crime, and criminal justice: Images and realities. Belmont, CA: West/Wadsworth.

65.

Surette

(2003). The media, the public, and criminal justice policy. Journal of the Institute of Justice and International Studies, 2, 39–52.

66.

Unnever

J. D.

Benson

M. L.

Cullen

F. T.

(2008). Public support for getting tough on corporate crime: Racial and political divides. Journal of Research in Crime and Delinquency, 45, 163–190.

67.

Weitzer

Kubrin

C. E.

(2004). Breaking news: How local TV news and real-world conditions affect fear of crime. Justice Quarterly, 21, 497–520.

68.

Welsh

Fleming

Dowler

(2011). Constructing crime and justice on film: Meaning and message in cinema. Contemporary Justice Review, 14, 457–476.

	Worry	Angry	Support for Criminal Justice System
	b (SE)	b (SE)	b (SE)
Media reliability	.01 (.04)	.13 (.05)**	.32 (.03)***
TV news viewing	−.08 (.11)	.02 (.09)	.06 (.05)
Violence on TV news	.23 (.07)***	.23 (.08)**	.00 (.05)
Violence on TV shows	.10 (.07)	.18 (.11)	−.01 (.07)
Violence in movies	.04 (.07)	.09 (.08)	−.05 (.05)
Violence on the Internet	.12 (.07)	.16 (.08)	−.01 (.05)
Employee in network	.01 (.02)	.07 (.02)**	.00 (.01)
Involvement in network	−.10 (.05)*	.12 (.05)*	.01 (.03)
Personal victimization	.44 (.18)*	.01 (.23)	−.53 (.14)***
Victimization in network	.06 (.11)	−.15 (.12)	.100 (.08)
TV News × Employed	.00 (.02)	—	—
TV News × Involvement	.14 (.06)*	—	—
TV News × Personal Victimization	−.17 (.23)	—	—
TV News × Network Victimization	.16 (.14)	—	—
Violent Shows × Employed	—	−.04 (.03)	.01 (.02)
Violent Shows × Involvement	—	−.16 (.07)*	−.07 (.04)
Violent Shows × Personal Victimization	—	−.12 (.28)	.42 (.17)*
Violent Shows × Network Victimization	—	.40 (.16)*	−.23 (.10)*
F	9.31***	5.54***	9.134***
Adj. R ²	.24	.15	.24