1 DAILY DOSE OF CRIME: THE MEDIAS ROLE IN BLAMING PARENTS By ASHLEY R. P. KOLNES A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTERS OF ARTS UNIVERSITY OF FLORIDA 2008
2 2008 Ashley R. P. Kolnes
3 To my husband, parents and brother who ha ve supported my academic endeavors, and encouraged me to dream, ma king this milestone possible
4 ACKNOWLEDGMENTS I would lik e to thank my chair, Eve Brank, and additional committee members, Jodi Lane and Lora Levett, for their knowledge and guidance. Additional thanks go to Caitlin Parry, Katie Greene, Rachel Arroyo and Chris King for their hard work and long hours as research assistants. To Thor Kolnes, Becca Grant, Becky Hayes-Sm ith, Caroline Martin and Sven Smith, much thanks for proofreading and editing. I thank my family and husband for their motivation and support, which encouraged me throughout my study.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ..........7 ABSTRACT.....................................................................................................................................8 CHAP TER 1 INTRODUCTION..................................................................................................................10 2 LITERATURE REVIEW.......................................................................................................13 Parental Responsibility........................................................................................................ ...13 General Media Effects............................................................................................................16 Media and Fear of Crime................................................................................................. 19 Just Media Hype?............................................................................................................ 21 Pretrial Publicity.....................................................................................................................23 Theoretical Framework.......................................................................................................... .26 Attribution Theory........................................................................................................... 26 Framing Theory............................................................................................................... 27 Agenda Setting Theory.................................................................................................... 29 Applying Media Theories to Notions of Parental Responsibility ................................... 29 3 METHODS.............................................................................................................................32 Study I, Phase I: New York Times Content Analysis ..............................................................32 Hypotheses......................................................................................................................33 Strategy and Implementation........................................................................................... 33 Study I, Phase II: Globe and Mail Content Analysis ..............................................................35 Hypotheses......................................................................................................................35 Strategy and Implementation........................................................................................... 36 Study I, Phase III: National Comparison................................................................................37 Hypotheses......................................................................................................................38 Strategy and Implementation........................................................................................... 39 Study II, Phase I: F lorida Times Union Content Analysis..................................................... 39 Hypotheses......................................................................................................................39 Strategy and Implementation........................................................................................... 40 Study II, Phase II: London Free Press Content Analysis ....................................................... 40 Hypotheses......................................................................................................................41 Strategy and Implementation........................................................................................... 41 Study II, Phase III: Local Comparison...................................................................................41 Hypotheses......................................................................................................................42 Strategy and Implementation........................................................................................... 42 Intercoder Reliability System................................................................................................. 42
6 4 RESULTS...............................................................................................................................49 Study I, Phase I: New York Times Content Analysis ..............................................................49 Study I, Phase II: Globe and Mail Content Analysis ..............................................................50 Study I, Phase III: National Comparison................................................................................51 Study II, Phase I: F lorida Times Union Content Analysis..................................................... 53 Study II, Phase II: London Free Press Content Analysis ....................................................... 54 Study II, Phase III: Local Comparison...................................................................................55 5 DISCUSSION.........................................................................................................................60 6 CONCLUSION..................................................................................................................... ..71 LIST OF REFERENCES...............................................................................................................73 BIOGRAPHICAL SKETCH.........................................................................................................79
7 LIST OF TABLES Table page 3-1 Search terms and productivity of national content analyses.............................................. 45 3-2 Examples of coding definitions used for national content analyses .................................. 46 3-3 Search terms and productivity of local content analyses ................................................... 47 3-4 Examples of coding definitions used for local content analyses ....................................... 48 4-1 Frequencies distributions of the national content analyses ................................................ 57 4-2 Reference to PR law and stance crosstab results for New York Times US .......................57 4-3 Reference to PR law and stance crosstab results for Globe and MailCanada ................. 57 4-4 Comparison pre-YOA to post-YOA for Globe and Mail Canada ....................................58 4-5 Comparison between New York Times and Globe and Mail National ............................ 58 4-6 Frequencies distributions of the local content analyses ..................................................... 58 4-7 Comparison between pre-curfew & post-curfew Florida Tim es Union-Local..................59 4-8 Comparison between Florida Times Union and London Free Press Local .....................59
8 Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Arts in Criminology, Law and Society DAILY DOSE OF CRIME: THE MEDIAS ROLE IN BLAMING PARENTS By Ashley R. P. Kolnes August 2008 Chair: Eve M. Brank Major: Criminology, Law and Society Numerous studies have examined media effects related to the publics perception of crime. Guided by theoretical framework and well establishe d areas of research, in cluding fear of crime and pretrial publicity, the current study seeks to establish the role the media has in reporting parental responsibility and blame for acts of juvenile delinquency. To accomplish this goal, the current study examines the frequency and context of parental responsibilit y in local and national print media, via four content analyses. Our research included two studi es. Study I details parental responsibility in the national media. First, a national U.S. newspaper, the New York Times was coded. Methods were replicated and used to analyze a Canadian national newspaper, the Globe and Mail To conclude study I, a national comparison was conducted, comparing findings from the Globe and Mail to the New York Times results. Study II examined parental inclusion in local juvenile crime reports. First, a U.S. local newspaper, the Florida Times Union was coded and methods were replicated in the analysis of a lo cal Canadian newspaper, London Free Press Findings from the Florida Times Union and London Free Press were then compared. Results demonstrate that the national media sources depict support for parental responsibility in the majority of coded news stories, yet the s upport decreases slightly when a
9 story references a specific pare ntal responsibility law. The lo cal sources analyzed specific juvenile crime stories and found little mention of parents. However, when parents were included in the story, the majority of references were irrele vant to blame. Finally, the U.S. media is more likely to assign blame to parents than the Cana dian media sources, illu strating a greater support of parental responsibility in the United States press.
10 CHAPTER 1 INTRODUCTION On Nove mber 3, 1998, an eight-year-old girl named Maddie Clifton vanished from her home, creating a media frenzy as television st ations and newspapers raced to cover the disappearance. Over the next week, teams of people desperately looked for the missing girl. The search ended tragically five days later as the mother of 14-year-old Joshua Phillips found the decomposing body of Maddie Clifton under her son s waterbed. Joshua Phillips, a teenage boy described by many to be soft-spoken and extremely kind, claimed that after accidentally hitting Maddies head and making her scream, he hit her in the head again and then repeatedly stabbed her to keep her quiet. He told the police that he then placed Maddies body under his bed to hide the crime because he was scared to tell his father he had someone over while his parents were not at home (Candiotti, 1999; Kohn, 1999; Shorstein, 1999; World, 1999). It is difficult for the public to grasp that a child could commit such a heinous act, and the media seemed to try to provide other factors that may have contributed to the juveniles actions. Specifically, the media reports suggested the father of Joshua Phillips was also to blame. Joshuas attorney hired a psychologist to evaluate his mental status, conf irming that Joshua was afraid of his father. Joshua claims that wh en he did something wrong, his father had a short temper. At times, he never knew what his father would do. It was this fear, Joshua said, that caused him to kill Maddie Clifton. Me dia outlets, including BBC, CBS, CNN and the local newspaper all included statements implicating Jos hua Phillips father as a reason for Cliftons death (Candiotti, 1999; Kohn, 1999; Shorstein, 1999; World, 1999). One of the most accusatory statements was made by the victims mother in a local newspaper article. She claimed she was doing her job as a parent to Maddie and asked w hy the parents across the street were not doing
11 theirs. She said that because of the Phillip s neglect, her family has to live with the consequences of Joshuas actions (Sweeney, 1999). Researchers recently have begun to examine wh ether parents like Joshua Phillips father should be held wholly or partially responsible for their childs crimin al actions. Parental responsibility laws hold parents legally responsible for the cr imes their children commit and have gained recognition in recent years. Most of the empirical research in this area has been limited to public support of thes e laws. Very little work has been done on the medias influence of the support toward parental responsibility, de spite the claim by some le gal scholars that the laws are nothing more than media sensationali sm (Shepherd, 1999). Borr owing from fear of crime and pretrial publicity research, the curr ent study examines the t ype and prevalence of parental responsibility concepts in the media. To test the id ea of media sensationalism, the current study employs the media theories of attr ibution, framing and agen da setting to examine parental blaming in the media. To understand how frequently the media blam es parents in juvenile delinquency and juvenile crime stories like the one described above, the current study examines four print media sources. First, a content analysis of a Un ited States national newspaper was conducted to establish the frequency and context of parental blame/responsibility in the media. Next, a Canadian national paper replicat ed the methods used in the US national paper to examine the portrayal of parental responsibility in a country. After unders tanding the frequency and context of parental responsibility in the national me dia sources, two local content analyses were conducted. The US and Candian local newspapers were analyzed to unde rstand the extent of parental inclusion in specific juvenile crime re ports. Media coverage and framing of a crime story, with statements such as the one made by Joshua Phillips and the victims mother, may
12 produce major social and legal implications by increasing the public demand for parental responsibility concerning a juve nile delinquents behavior.
13 CHAPTER 2 LITERATURE REVIEW Parental Responsibility Am erican citizens are looking for someone to blame as the public perception of juvenile crime rates and news stories of these crimes soar (Thurman, 2003; Tomaszewski, 2005; Tyler & Segady, 2000). Within the juvenile justice syst em, a conflicting movement is emerging on how to classify a juvenile offender. The system is more often asserting that juveniles must be held accountable for their actions, yet st ate and city laws require that the parent become involved and responsible for the delinquent acts of their children (Brank, Hays, & Weisz, 2006; Brank, Kucera, & Hays, 2005; Davidson, 1996; Dimitr is, 1997; Tyler & Segady, 2000). Although multiple areas of research support the concept that children are malleable and their moral development may be influenced by their pare nts, (Bahr, Marcos & Maughan, 1995; Bianchi & Robinson, 1997; Burton, Cullen & Evans, 1995; Kuendig & Kuntsche, 2006; Zhang, Welte & Wieczorek, 1999), there is no empirical research th at shows that punishing parents will have a positive effect on that influence. Despite the lack of empirical support for their e ffectiveness, all states within the US have either civil or criminal parental responsibility laws that hold a parent legally responsible for his or her childs actions. Brank et al. (2005) categorizes these laws into three different forms. The first is the civil liability versi on that holds parents civilly liable if their child intentionally harms someones person or property. Thes e statutes provide an option to the difficult to win negligent supervision case within common law, but they have relatively low recovery limits. The statutes that involve contributing to the delinquency of a minor are the sec ond form and they can apply to any adults. Under these laws, a parent will be held responsible when he or she directly contributes to or encourages a delinquent act. The third form varies greatly across the
14 jurisdictions, and the laws are generally referred to as parental involvement statutes. These laws encourage or require a variety of activities by th e parents such as part icipating in the court hearing, paying for court costs, participating in community service or attending parenting classes. Some legal scholars support these statutes (Davidson, 1996; Dimitris, 1997), while others do not find parental responsibility laws necessary, effective or fair (Alexander, 1948; Evans, Evans, & Meredith, 1993; Kenny & Kenny, 1961; Tomaszewski, 2005; Tyler & Segady, 2000). Even in cases where parental punishment seems mo re appropriate, such as instances in which a parent is present during commissi on of a crime, some scholars beli eve it is possible that society may be losing sight of solving the problem and simply reacting and seeking retribution (Tomaszewski, 2005; Tyler & Segady, 2000). Some legal scholars posit that assigning blame to the parents allows the community to eliminate it s own level of responsibility in contributing to delinquency (Kenny & Kenny, 1961). Other scholar s believe it is possible for parental responsibility laws to be more effective if they are accompanied by the proper support programs (Davidson, 1996; Dimitris, 1997). Nonetheless, not all scholars are convinced that parental liability laws alone will decrease the level of juve nile delinquency or that the laws will increase the level of parental responsib ility (Tyler & Segady, 2000). When community members were asked by Kenny and Kenny (1961) if a parent should be held responsible for a juvenile s delinquency, an overwhelming 89 percent said parents should bear the responsibility of their childs actions. Alt hough the Kenny and Kenny study may seem outdated, similar results were found in a recent study conducted by Brank and Weisz (2004). In this study 70% of respondents said parents, as well as the juve nile, were responsible when a delinquent act was committed. Additionally, accord ing to a 1999 Gallup poll, parents were said
15 to be the next most responsible party in a juve nile crime, after the j uvenile offender (Brank & Weisz). In Brank et al. (2006), a survey to m easure support for parental blame assignment was completed by 168 undergraduates. Participants were asked general/global questions about parental responsibility, given a specific juvenile crime case a nd asked to answer specific questions about parental responsib ility relevant to the crime story. The study demonstrated that while global support for parental responsibility exists, there may be less support for blaming the parent when a specific scenario is presented (Brank et al., 2006). Some scholars believe that parental responsibility, as a form of blame assignment, has gained public attention and support due to high pr ofile media coverage that emphasizes the role parents play in juvenile crimes (Tomaszewski, 20 05). After the extensive media coverage of the Columbine High School shootings, the victims parents expressed thei r desire to hold the shooters parents responsible in civil court, thru sting the concept of parental responsibility into the media spotlight (Tomaszewski, 2005). A dram atic influx in coverage of juvenile crimes, such as the media frenzy surrounding similar hi gh school shootings, may strengthen support for parental responsibility laws. (Chiricos, Eschhol z, & Gertz, 1997; Dorfman, Woodruff, Chavez, & Wallack, 1997). Recently, Book and Perala-Littunen (2008) content analyzed a Finnish newspaper to establish various definitions and categories of parental responsi bility and how the concept is being portrayed in the national media. An analysis of stories printed in select months of 2002 in Helsingin Sanomat a Finnish national daily paper, yielded a sample of 18 letters to the editor. From the analysis, three dimensions of parental responsibility were extracted. These concepts included beginning responsibility (from the viewpoint of exp ectant parents), diminished responsibility (sense that parents do not take responsibility for their children or feel that the role
16 is theirs) and obligating responsib ility (a parent must or sh ould be responsible for their child). Although the stu dy utilized a small sample, it provided a qualitative categorization of parental responsibility in the media and de monstrated a need for additional and more systematically complete content analyses focusing on this topic. Regardless of the effectiveness, support and me dia attention toward pa rental responsibility laws, some people question the constitutionality and fairness of holding a person responsible for an act someone else committed (Evans et al., 19 93; Tomaszewski, 2005). To some, parenting is a private matter and parents should be given the ri ght to raise their children in the manner they choose (Evans et al.; Tomaszweski, 2005). However, others feel there may need to be a limit on this freedom in order to prot ect communities (Dimitris, 1997). Ty ler and Segady (2000) discuss how Americans have developed a common assumpti on that bad kids are the product of bad parents. It is possible, however, for responsib le, caring parents to have a delinquent child and this creates the potential that parents may become victims of th e system when punished for their childs behavior (Tomaszewski, 2005; Tyler & Segady, 2000). The current study examines these different viewpoints by measuring the frequency and context of parental blame in th e local and national media. Th e establishment and understanding of the medias portrayal of pare ntal responsibility a nd juvenile delinquency is critical. Once establishing the presence and construct of parental blame in the media, it will become possible for future researchers to ques tion how the media might contribute to public opinion of parental responsibility. The following section details th e potential influences the media may have on public opinion. General Media Effects The m edia is said to paint an artificial sens e of reality (Heath & Gilbert, 1996; Lowry, Nio, & Letner, 2003; Romer, Jamieson, & Aday, 2003). In other words, the media provide the public
17 with a descriptive view of the world and create a specific perception of events the public cannot directly witness (Iyenga r, Peters, & Kinder, 1982 ; Heath & Petraitis, 1987). Violence and crime are media favorites. Stories of vi olence often dominate local news coverage, from the front page of the newspaper to the evening news broadcast (Dorfman et al., 1997). For instance, University of California researchers observed all Los Angeles newscasts from a single station for a year and found that crime was the focus of 51% of the ne ws broadcasts (Budiansky, Gregory, Schmidt, & Bierck, 1996). Of the crime stories shown, 78% involved violent crime and 27% of the stories were about murder. With murders only accounti ng for 2% of felony charges in Los Angeles during this same time period, it is evident that news stations still follow the adage, If it bleeds, it leads (Budiansky et al., p.63; Yanich 2005, p. 104). Previous studies have shown television news casts have a considerable effect on public perception (Heath & Gilbert, 1996; Lowry et al., 2003; Romer et al ., 2003). One noted effect of television viewing is the degree of importance viewers assign to current issues (Iyengar et al., 1982). The more people watch television, the more likely they are to identify crime as a major problem in their community (Heath & Gilbert, 1 996; Lowry et al.; Romer). Even though other problems and risks may be mentioned in the newscast, crime is still perceived by the viewers as a more significant issue compared to other issues mentioned (Romer). Media attention to juvenile crime is al so high. A 1997 report by Butts and Snyder, on behalf of the Office of Juven ile Justice and Delinquency Prevention, examined increased media attention of juvenile crime. Th e report noted that the media sensa tionalizes juvenile crime, often repeating the details of the crime and the story multiple times. In a content analysis of 214 hours of California television news, re searchers found that over two-thir ds of violent news stories involved juveniles and over half the stories th at featured children al so included violence
18 (Dorfman et al., 1997). Dramatic influxes in juve nile crime coverage could lead Americans to identify juvenile crime as a pressing and rising is sue in society, even though actual crime rates have remained relatively constant (Butts & Snyder 1997; Chiricos et al). Some have argued that the over-represented media depictions of juvenile offenders and juvenile crime have made public attitudes toward juvenile offenders more pun itive (e.g., Ghetti & Redlich, 2001) and that the media attention appears to have in creased the publics fear of crime. A persons view of reality is oftentimes shaped more by the mass media than reality itself (Lowry et al., 2003; Yanich, 2005). In fact, ne ws media exposure can create a fear of crime within a viewer that is indepe ndent of actual crime rates (Romer et al., 2003). Utilizing the Vanderbilt TV News Index, Lowry et al. (2003) found that factors in television news have four times the effect on peoples fear than actual police statistics. Many people depend on the media for various types of information, yet the mass media has a motivation beyond just presenting information. In order to increase circulation or viewing, distortion of news facts may take place leading to audiences not rece iving accurate portrayals of reality (Yanich, 2005). As mentioned above, one such media distortion of reality is the focus on crime. Crime is a central topic of news broadcasts and may lead viewers to misj udge the problem of crime while ignoring other less publicized issues (Iyengar et al., 1982; Ro mer et al., 2003). By placing a greater degree of importance on the problem of crim e, viewers are more likely to perceive crime as the most important concern facing their community (Rome r) even when crime rates are relatively low. It is even possi ble that viewers experience indirect victimization due to the fear generated by their media consum ption (Lane & Meeker, 2003). Although the current project does not directly test for eff ects of media on public opinion, research conducted on the medias influence of fear of crime and on jurors attitudes (i.e., pretrial
19 publicity effects) provides an example of how the medias focus might influence notions of parental responsibility. These two areas will be described in turn: 1) the impact media distortion may have on fear of crime, and 2) the impact of pretrial publicity on jurors attitudes about a case. Media and Fear of Crime Clearly, the m edia focuses much of its atte ntion on crime and violence. Literature demonstrates that fear of crim e may be highly correlated to th e exposure to crime information through the media. Some scholar s argue that peoples perception of crime is a significant problem in the United States and may be just as important to examin e as actual crime rates (Clemente & Kleiman, 1977). Although fear can socia lly inhibit individuals, it is functional in conditioning people to protect themselves and their property (Garofalo, 1981; Stafford & Galle, 1984). Fear of crime can come from various sources including personal experience, social communication and the mass me dia (Garofalo, 1981). Although many scholars agree that th e media impacts fear of crime, there is less agreement on the specific effects of television versus newspapers. Even though fewer Americans are reading the news, newspapers se t the standards for journalists and often serve as a frame for television news broadcasts (McManus & Dorfma n, 2002). Over time, television has become the dominant medium chosen for informational news (Romer et al., 2003). Network television news not only influences which problems viewers decide ar e important, but the stati ons also are able to shape public agendas (Iyengar et al., 1982; Lowry et al., 2003). This vast influence has led researchers to question the effect s television news has on fear of crime. In a study by Romer et al. a telephone survey was administered to over 1,000 people to determ ine the participants perceived personal risks and exposures to multip le news mediums. Results suggested that watching television news increased perceived risk of crime. Ch iricos et al. (1997) similarly
20 found that television news influences public perception, when controlling for gender, age, race, prior victimization and other va rious views of crime. The re sults of their telephone survey indicated a significant positive association between television viewing and an increase in fear of crime. Newspapers often influence the direction a nd content of television news broadcasts (McManus & Dorfman, 2002); however, some schol ars have found that unlike television news, newspapers do not have a signifi cant effect on fear of crime. Utilizing a random phone survey, Chiricos et al. (1997) found fre quency of reading the newspaper did not significantly increase fear of crime among respondents. In a series of studies, Romer et al. (2003) also found that even frequent reading of newspape rs did not significantly in crease fear of crime. Regardless of the medium, many scholars be lieve specific demographic factors can influence the effect media have on a viewers per ception and fear of crime. For example, gender is believed to have a significant effect on fear of crime. Women viewers and readers are more likely than men to be fearful of crime (Chi ricos et al., 1997; Gross & Aday, 2003; Heath, 1984; Heath & Petraitis, 1987). Some studies express that age also affects fear le vels; although whether age is negatively or positiv ely associated with fear is not consistent throughout prior fear research (Chiricos et al.; Gro ss & Aday, 2003). Finally, race app ears to be a factor in media effects on fear; with minorities typically feeling more fearful th an whites (Chiricos, Escholtz, and Gertz, 1997; Gross & Aday, 2003; Lane & Meeker, 2003). In addition to the demographic factors, divers e social factors can also partially determine effects of the mass media on fear of crime. Preconceived attitudes of the medium and media message may also influence pe rsonal interpretati ons of the observed media message, which could elevate fear levels in particular individuals (Os good & Tannenbaum, 1955). Finally,
21 previous experience with crime and victimizatio n is one of the greatest predictors when estimating fear of crime (Gross & Aday, 2003). Resonance is the concept that media messages of violence would resonate with those who have b een or are at a high risk of victimization, and it is a common interpretation of reason for fear (Chiricos, et. al, 1997; Doob & Macdonald, 1979). Chiricos et. al (1997) found some evidence of resonance in the results of their survey. Consistent with resonance, white women who had been recently victimized and were living in low income neighborhoods demonstrated that television news consumption was significantly related to fear of crime. However, there was no evidence that black participants or men, populations most likely to be victimized, demonstr ated an increased level of fear based on media consumption. Location of the story is a final element that ma y contribute to an increase in fear. Whether a described crime is local or dist ant may be more influential in pr edicting levels of fear than previously mentioned factors. Vi ewers are more likely to demonstrate an increase of fear if a reported crime is in close proximity to th eir neighborhood (Heath, 1984; Heath & Petraitis, 1987). Conversely, if a crime is distant from pa rticipants homes, they are more likely to experience reassurance in their own neighbor hoods safety levels (Doob & Macdonald, 1979; Heath, 1984; Heath & Petraitis, 1987). Regardless of how gruesome and disturbing a crime story may be, if the crime occurred in a different city, it will still bolster peoples feeling of personal security (Heath, 1984). Just Media Hype? Although the vast m ajority of re search has demonstrated a media effect on fear of crime, some studies have had conflicting results. Gro ss and Aday (2003) report that local television news was not related to an increase in fear in their study, even though results indicated that media consumption increased participants likelihood of naming crime as an important
22 community issue. After administering a door-todoor survey in selected areas of Toronto, Doob and Macdonald (1979) found the amount of television viewed was related to fear of victimization or crime. However, once the researchers cont rolled for actual incidents of neighborhood crime, the fear effect disappeared. A major con cern with the Doob and Macdonalds study, and a possible explanation for the ra re finding, is the year of publication. Since 1979, technology, monumental crime stories and media reporting in general have evolved. Furthermore, more recent studies have demonstrated the impact of the media on individuals fear of crime (Heath, 1984; Heath & Petraitis, 1987; Heat h & Gilbert, 1996; Lowry et al ., 2003; Romer et al., 2003). Studies that illustrate the lack of media effects are rare. Researchers who conducted studies and found contradictions to the common idea that media c onsumption influences fear of crime are typically not claiming the media does not have any effect on its consumer. Instead, the articles tend to explain the complexity of a nd the many additional factors that lead to a heightened level of fear of crime. Ditton, Ch adee, Farrall, Gilchrist and Bannister (2004) found no significant relationship between fear and fr equency of media consumption. Instead, they (Ditton, et al., 2004), explained how their study rev ealed a connection between fear of crime and peoples perception of the media message. They fu rther detailed that one of the most important factors influencing fear of crim e was a storys perceived relevan ce to an individual consumers life. The conflicting studies demonstrate the im portance of continuing to examine the messages the media sends. The media effects explained in fear of crime literature demonstrate that public perception and opinion can be shaped by fre quent exposure to news. The media is capable of influencing society through the manipulation and presentation of information. Fear of crime is just one aspect of media influence; public attitudes about other issues may also be affected by the medias
23 power. For instance, research has illustrated that publicity before a trial can influence a jurors attitude about a case. The followi ng section will detail some of th e literature concerning pretrial publicity effects. Pretrial Publicity The First Amendment guarantees the free dom of press while the Sixth Amendment secures a persons right to a fa ir trial. Scholars question whether these two amendments can coexist successfully (Costant ini & King, 1980; Kramer, Kerr, & Carroll, 1990; Otto & Penrod, 1994; Steblay, Besirevic, Fulero, & Jimenez-Lo renta, 1999). The difficulty in achieving the needed balance between these f undamental rights is illustrated in nationally publicized cases (Bruschke & Loges, 1999), such as the Laci Pe terson murder. The intense media coverage and scrutiny of Scott Peterson, the man suspected of killing his wife and unborn child, made it easy for the public to develop a preconceived opini on of Mr. Petersons guilt. Negative media influences are not limited to such national scenar ios, but local cases are at similar risk of potential pretrial publicity effects (G reene, 1990; Steblay et al., 1999). There are likely to be other influences that lead to preconceived j udgments about a trial, but pretrial publicity may be the most signifi cant in shaping peoples thoughts (Costantini & King, 1980). Pretrial publicity studies have demonstrated the power of the media on a persons decision making processes (Carroll, Kerr, Alfi ni, Weaver, MacCoun, & Feldman, 1986; Greene, 1990). Previous pretrial publicity research relevant for the curre nt proposal examines the types of pretrial publicity a nd the effects of these on jurors d ecisions prior to and during a trial (Kramer et al., 1990; Wilson & Bornstein, 1998). Negative pretrial publicity has a significant effect on jurors judgment (Greene, 1990; Otto, Penrod, & Dexter, 1994; Steblay et al., 1999). Specific categories of pretrial publicity have been
24 developed, with each type affecting attitudes in di fferent ways. The areas most relevant to the current study will be information on factual, emotional and recalled pretrial publicity. Factual, case-specific pretrial publicity contains specific incr iminating material about the defendant in a particular case (Kramer et al ., 1990; Wilson, 1998). Two good examples of factual pretrial publicity from the example case descri bed above are that Scott Peterson refused a polygraph, and he was having an affair at the tim e his wife was murdered. One of the strongest effects of the factual pretrial publicity is the negative charac terization of a defendant if discriminatory information or the prior record of the defendant is released (Otto et al., 1994). Exposure to negative case facts, re gardless of accuracy, is likely to cause potential jurors to develop a pro-prosecution st ance (Kramer et al.; Wilson & Bornstein, 1998). Emotional publicity contains sensational, vivid elemen ts of a case that are noninformational but likely will solicit emotional re actions in potential jurors (Kramer et al., 1990; Wilson, 1998). The following facts, again from th e Laci Peterson case described above, are good examples of emotional publicity. At the time of his arrest, Scott was in possession of his brothers ID, Viagra and $15,000. Days after his wifes disappearance, Scott added pornography channels to his television cable package. Scott was also seen laughing at his wifes funeral. While emotional pretrial publicity may not contain specific facts about a defendants guilt in a case, the coverage may have longer lasting consequences be cause emotional arousal is memorable (Kramer et al.). Emotional and fact ual pretrial publicity are constructed using different levels of emotional triggers, elements that stimulate the information recipient. Both types of publicity possibly produce the same level of bias (Wilson & Bornstein, 1998). In an experiment conducted by Wilson and Bornstein (1998), the ideas of factual and emotional pretrial publicity were tested. To manipulate factual pub licity, information of the case
25 was included for some participants. For example, some participants read statements such as Kelly fled the scene of the crime and Kelly tried to kill her brothe r Brad a year earlier (p.590). Emotional publicity was manipulated vi a statements like Kelly [the defendant] ran back to the bedroom and started to crush her dead mothers skull w ith a baseball bat and Kelly would frequently flush her moms heart medication down the toilet (p.590). The results of the experiment did not provide sufficient evidence that factual pretri al publicity was less prejudicial than emotional. Instead, it provided evidence that both types of pretrial publicity created very similar levels of bias. No significant differen ce was found between the type s of pretrial publicity (emotional vs. factual) or the format of the messages (print vs. video). Some researchers believe the key to understand ing pretrial publicity is determining what the public and potential jurors remember and how this memory may influence future reasoning in court (Honess, Charman, & Levi, 2003). The con cept of recalled pretrial publicity is not based on the contents of the media coverage, but rather the viewers recollectio n and interpretation of such exposure (Honess et al., 2003). Those who are able to recall great amounts of information from pretrial publicity may be more likely to assu me a defendant is guilty than those people who do not remember the media coverage (Steblay et al., 1999). Pretrial publicity can affect initial judgment s of a defendant based solely on information gathered from media reports (K overa, 2002; Otto et al ., 1994; Wilson & Bornstein, 1998). In an experiment conducted by Otto et al., 262 participants were exposed to a videotaped version of a case and fictional newspaper articles that di splayed five degrees of information about the defendant. Participants exposed to negative pr etrial publicity pieces involving the defendants character were more likely to administer a guil ty verdict than those w ho were not exposed to character-relevant information. Th ese findings were confirmed in a meta-analysis by Steblay et
26 al. (1999), which revealed that the majority of research has found exposure to negative pretrial publicity is more likely to result in a guilty verdict than little or no exposure to media information (Otto et al.; Steblay et al.). The fact that both fear of crime and pret rial publicity research demonstrate a media influence on attitudes is no surprise; media studies theories such as attribution, agenda setting and framing are all based on the notion that the media can and does influence public opinion. The next section will detail each of these theori es and describe how they can guide a systematic study of the medias message. Theoretical Framework Attribution Theory Attribution theory was born out of the field of social psychology and was developed by Fritz Heider, Harold Kelley, Edward E. Jones an d Lee Ross. The theory refers to a persons interpreted cause of an event or behavior (Bu ss, 1978; Hastie, 1984; Kelley & Michela, 1980). It is important to emphasize that th e attributions are derived from informational observations; ways in which people gather information and use th e observation to learn about society around them (Bassili & Regan, 1977). These proposed causes or explanations then become the basis for attribution of blame or responsibility for the behavior. It is comm on for people to explain behavior by their perceived cause s and for those explanations to generate a reaction to the behavior (Kelley & Michela). Pe ople are likely to assign responsibility to the first causal factor that they think of or at least the most logical explanation ob served at the time the assignment of blame arises (Kelley & Michela). For instance if the media describes a presidents foreign policies as a reason for higher gas prices, then the public may be more likely to attribute the blame of high gas prices on the president. The mo re severe the consequences of an event, the more likely the forced reaction will be a strong ne ed to assign responsibil ity to the appropriate
27 party (Walster, 1966). Building on th e prior example, if the media reports that the average price of gas is expected to increase from $4 to $5 per gallon, the publics desire to find the responsible party will be exacerbated. The perceiver will co mplete the attribution process by searching for specific bits of information and applying the su itable information to the scenario (Bassili & Regan, 1977). Attribution of res ponsibility has a strong connection with social change because whom society holds accountable for social pr oblems will shape the development of social solutions (Coleman & Thorson, 2002). Framing Theory News, in tim e, can develop into a media real ity (Yanich, 2005). Understanding framing is crucial to understanding the me dias power. Framing theory is the idea of constructing and formatting the news in order to develop news broadcasts or newspaper layouts that will be disseminated to consumers with manipulated inform ation. According to th e theory, the viewers understanding of an issue can be distorted or limited due to the pr esentation of the story. For example, including or excluding ce rtain facts can significantly alte r how a viewer will interpret an issue. Framing assumes mass communication ha s the ability to direct peoples attention to certain problems or issues (Heath & Petraiti s, 1987; Iyengar, 1990; McCombs & Shaw, 1972) by creating certain salient facts wh ile ignoring others (Entman 1993; Iyengar, Peters, & Kinder, 1982; McManus & Dorfman 2002). If a news outlet wa nted to develop a story detailing the Iraq war in a positive light, it may leave out any details of casualties. Conversely, if the news source wanted to construct the same story in a negative light, it may chose to exclude all facts of progress and instead focus on increasing loss of innocent civilians and U.S. soldiers. One possible side-effect of framing is a distorted or alternative reac tion to a specific event by various audience members (Entman).
28 Certain crime stories provide a clear example of framing. In order to create interest in a crime story, reporters personalize, dramatize and segment the material (Yanich, 2005). The degree of frequency and dramatic framing of crim e stories often leads to an increased perception of vulnerability that becomes a false reality to vi ewers (Romer et al., 2003). For example, with the emergence of high profile school shootings an d juvenile murder cases, the understanding of juvenile crime is overtaken by sensational images (Yanich). The method in which a story is constructed can determine how viewers attribute blame or responsibility for a specific act (Coleman & Thor son, 2002). By controlling who is portrayed as the most responsible character in a violent cr ime story, the news frame can heavily influence who is blamed (Coleman & Thorson; Rodgers & Thorson, 2001). In a study by Rodgers and Thorson, the contents of various Los Angeles Times crime stories were analyzed to determine the presence or absence of blame frames. These bl ames frames were identified by story facts that included elements of blame such as a father bein g held responsible for a car accident because he knowingly allowed his intoxicated t eenage son to drive or that th e accident victim died because he was not wearing his seat belt. Blame frames were found to be the most frequently utilized news frames, particularly in crime reporting wher e stories define blame as person X was killed because person Y stabbed him (Rodgers & Thorson). Framing theory has two distinct frame type s, each with differing effects on blame assignment. Episodic framing, in which a specif ic person or event is highlighted, leads to blaming the individual while thematic coverage, which describes a broad so cial issue, leads to societal blame (Iyengar, 1990). A content anal ysis conducted by Dorfman et al. (1997) found episodic framing occurred five times more frequen tly than thematic fram ing in reports of youth violence.
29 Agenda Setting Theory Sim ilar to cultivation theory, which describes the influence media has on peoples fear of crime, agenda setting theory describes the medias ability to tell consumers what issues are important. Agenda setting theory is based on two basic assumptions: 1) the press and media do not reflect reality and 2) the media concentrates on select issues leading the public to consider those issues as important. Recently, the media has given ample attention to going green or helping improve the environment and conserve re sources. However, often the media encourages its viewers to go green by buying eco-friendly products. Therefor e, this media coverage is leading people to believe that by purchasing eco-friendly products they are significantly contributing to the environmental cause without ha ving to make a major lifestyle changes. For many viewers, perceptions of reality portrayed by the mass media are ofte n stronger than reality itself (Lowry et al., 2003). It is clear that media influence our perceptions of social problems, victims, and perpetrators. Soci ety utilizes the media as a primary source of understanding crime and justice; yet it is becoming apparent thr ough research that the media does not necessarily serve the public with an accurate depiction of its community (Yanich, 2005). The extent of media influence leads to the conclusion that media not only tells us what to think about, but it also tells us what to think (Iyengar, et al., 1982; Kovera, 2002; Lo wry et al.). The strength of agenda setting grows as the public fa ils to question what they see and hear in the media (Iyengar, Peters, & Kinder, 1982). Applying Media Theories to Noti ons of Parental Responsibility The aforem entioned theories attempt to predic t the influence of th e media on society and detail why the content and deliv ery of media messages is impor tant to understand. Content analyses are a well-established method of expl oring and quantifying the information delivered via the media. However, other than the Finni sh content analysis by Book and Perala-Littunen
30 (2008) few studies have been conducted to anal yze the content of relevant print media for mentions of juvenile crime. A content analysis conducted by Humphrie s (1981) highlighted crime in a major metropolitan area, but was not specifically focused on juvenile crime. Humphries coded the New York Post for routine crime stories published in 1951 and 1968. These two years were selected to represent the heighten ed crime level of the Post-World Wa r II era. From the annual stories, 5% were selected from each year, yielding 126 ar ticles. Each story was recorded for factual crime data, location and severity. The results pr oduced categories of yout h criminal activity and depicted one-third of offenders and 40% of vict ims as juveniles. Humphries commented that adolescence is a problematic period for our yout h and an illustration of societal failures. In Canada, a study conducted by Sprott (1996) co mpared juvenile crime content from three Canadian newspapers to content found in offi cial crime reports of delinquent youth acts. Newspaper stories were selected from a two month period in 1995, yielding 113 articles. Each story was coded for crime facts, including offense and effects/results of the crime. These articles were then compared to crime descriptions from official crime reports. Within the newspaper stories, 94% of juvenile crimes involved viol ence, while less than one -fourth of actual youth crime reports included violence. The compara tive content analysis al so found that newspaper reports rarely included a juveniles court disposition and instea d only focused on the facts and effects of the crime. Sprott emphasized that mo st people do not read the official crime reports, but instead are relying on newspapers to provide the crime reports, which may not always be completely accurate. Guided by establish areas of research, includ ing fear of crime and pretrial publicity, the current study will seek to strengthen the exis ting literature on parental responsibility. The
31 current study will examine how the media portray s parental responsibility and will expand the limited content analyses that focus on juvenile delinquency. In addition to previous literature, attribution, framing and agenda setting theory guide the current studys hypotheses and content analysis structure.
32 CHAPTER 3 METHODS The current study em ploys four content analys es. The main functions of this methodology are to provide a replicable desc ription of content and to create inferences about the context in regards to its classified, deta iled subject matter (Smith, 1988). Like all methods, content analyses have strengths and weaknesses. Accord ing to Krippendorf (1986), content analysis is a strong tool because it is unobtrusive and content se nsitive, accepts unstructured material and is capable of handling large levels of data. A major concern with the process is that problems can emerge when large quantities of information ar e condensed into fewer categories (Weber, 1990). For instance, personal biases could influence the coding of information. To avoid issues within the current study, extensive pla nning and training was instituted and a systematic intercoder process was applied to each content an alysis (described in detail below). For each content analysis, the researcher care fully assessed the coding of multiple stories on the same crime/incident. The first story detailing a juvenile crime was always coded. If the independent coder found subseque nt stories detailing the same incident, the following would have to be considered: if the subsequent story provided new information (additional accomplices, parental involvement, explanation for the act, court decision, specificati on of charge/ruling) it would be coded as an additional story, but if the subsequent story was simply a reiteration of the primary story, and did not provide any new in formation, it was not coded for purposes of analysis. Study I, Phase I: New York Times Content Analysis The first phase of the current study is designe d to establish the existence of parental blam ing in the media through a longitudinal conten t analysis. To develop a broad understanding
33 of parental issues in juvenile delinquency, a national newspaper was selected for analysis as it mainly examines issues and opinions rather than specific juvenile crime stories. Hypotheses The study in Phase I tests multiple hypotheses to measure the frequency and context of parental responsib ility within the New York Times. H1: There is a positive slant in support of parental responsibility in the national media (framing theory). H2: The context of the majority of stories in the national media will be thematic rather than episodic (Adapted from Dorfma n, et al., 1997; Iyengar, 1990). H3: If a story contains mention of a specifi c parental responsibili ty law (including the application of such a law), the story is less lik ely to support parental bl ame than stories that do not reference a parental respons ibility law (Adapted from pub lic opinion studies by Brank & Weisz, 2004; Brank, et al 2006; Kenny & Kenny, 1961). Strategy and Implementation The prim ary content analysis was designed to establish the existence of parental responsibility in print media, wh ile gauging both the frequency and context of parental blame. To accomplish this goal, the New York Times was chosen, as it is the largest newspaper in the United States, according to the Audit Bureau of Circulation. The newspaper is well known for its coverage of public opinion, business, and brea king news. This portion of the study was not created to examine in-depth juvenile crime c overage. Instead, the purpose was to provide a broad, general examination of public opinion and so cietal reaction to pare ntal responsibility and juvenile delinquency. National stories were analyzed for the incl usion of parental re sponsibility or legal developments holding parents le gally responsible for juvenile delinquency. These types of
34 stories appeared mainly in articles (n = 501) and letters to the editor (n = 40). Therefore, the national search was limited to these two categories of stories to exclud e irrelevant categories such as real estate transactions classified ads and stock quotes. The articles (coded as 0) were likely to examine juvenile behavi or, policy reviews and legal info rmation. Letters to the editor (coded as 1) contained opinion pieces and re actions to crime probl ems and legal action. Stories were analyzed from 1899 through 2004 to demonstrate how the medias portrayal of parental responsibility has changed throughout time. The beginning year of 1899 is significant as it marks the inception of the first j uvenile justice system in the United States. Originally, the analysis was to only include a 100-year span of parental responsibility in the media, ending the analysis in the 1990s. However, a number of school shootings involving juvenile perpetrators occurred in the 1990s. To avoid ending the analysis within the time frame of these events, the analysis was expanded an additional five years to include stories through 2004. Six search terms (Table 3-1) were used to search the New York Times for references to juveniles and their parents. Basic citation information was recorded, including date of publication, article title, section/pa ge and author. Searches were done within the Letters to the Editor and articles, eliminating advertisements obituaries, marriage/birth announcements and editorial cartoons. Three elements of the stories were coded. First, the stance of the article was recorded as (1) supports parental blame, (2) neutral on parental blame or (3) against parental blame (See Table 3-2 for examples of article stan ce). Next, the article was examined for context and documented to include (1) a thematic referen ce to parenting and pare ntal effects on children or (2) an episodic reference to parental acts or encouragement cont ributing to juvenile delinquency (See Table 3-2 for examples of thematic /episodic context). Fi nally, the mention of a
35 parental responsibility law was r ecorded for each story, coded (0) fo r absent and (1) for present. Examples of coding definitions can be found in Tables 3-2. Study I, Phase II: Globe and Mail Content Analysis After m easuring the presence of parental responsibility in US national newspapers, the current phase expanded to Canadian media sources to determine if a similar slant and context could be observed in other countries. It seemed appropriate to select Ca nada as a country for analysis, as the movement of j uvenile-specific curfews originated in Canada and extended into the US in the mid 1890s (Baldw in, 2002). Also, the juvenile justice systems of the United States and Canada are parallel, allowing the juvenile crime storie s of both countries to be easily compared. Similar to the first phase of the current st udy, the current phase is designed to establish the existence of parental responsibility in Ca nada through an additiona l longitudinal content analysis. Again, a national newspaper was chosen as it primarily provides general/thematic representation of current issues. Hypotheses The Globe and Mail content analysis tests m ultiple hypotheses, which are mirrored from the New York Times phase, to measure the frequency and context of parental responsibility within the Canadian national newspaper. H1: There is a positive slant in support of parental responsibility in the national media (framing theory). H2: The context of the majority of stories in the national media will be thematic rather than episodic (Adapted from Dorfm an, et al., 1997; Iyengar, 1990). H3: If a story contains mention of a specifi c parental responsibili ty law (including the application of such a law), the story is less lik ely to support parental bl ame than stories that do
36 not reference a parental respons ibility law (Adapted from pub lic opinion studies by Brank & Weisz, 2004; Brank, et al ., 2006; Kenny & Kenny, 1961). Strategy and Implementation Acting as a com parison group to the primary New York Times content analysis, the Globe and Mail analysis was also designed to measure the frequency and context of parental responsibility in the national media. Comparable to the New York Times in popularity and structure, Globe and Mail was chosen for analysis. The pa per is the third largest national newspaper in Canada, with a daily circul ation of approximately 417,000 and a weekend circulation rate of 330,000. Again, the goal of this phase was not to establish parental components in specific juvenile cr ime stories. Instead, the goal wa s to establish contexts, trends and slants within national news stories i nvolving juvenile delinquency and parental responsibility. The main focus for the current study was to examine the frequency and framework of general parental responsibility and legal developments holding parents responsible for juvenile delinquency. Again, stories were extracted from articles (n = 142) and letters to the editor (n = 2) to encompass feature stories, opinion pieces and societal reactions. Stories were analyzed from 1978 through 2007. The time frame for the Globe and Mail analysis is limited, compared to the New York Times because the researcher only had access to stories published after 1977. The history of the Canadian Juvenile Justic e System began in 1908. In addition to the creation of an official juvenile justice system, several significan t shifts in the purpose of the Canadian system should be noted. According to the Canadian Department of Justice, the juvenile system began in 1908 under the Juvenile Delinquents Act to esta blish the state as a parent for misguided children. A major shift in the system arose in 19 84 with the passage of the Young Offenders Act. The Act holds that, in stead of acting as a parent to a misguided
37 juvenile, the state held juveniles responsible for their actions, and began to assume a punitive role in the juveniles lives. The current study begins with newspaper coverage from 1978, capturing the years leading up to the implementation of the Young Offenders Act. The most recent change was the Youth Criminal Justice Ac t of 2003. With this newest act, the purpose of the juvenile justice system was to reduce the punitive approach and rely less on incarceration of juvenile delinquents. Again, Canada shifted to a rehabilitative system, which remains in place today (International Cooperation Group, 2004). Th e historical changes mentioned will be important to consider when analyzing the results of the Globe and Mail content analysis. Therefore, findings within the Globe and Mail will be compared prior to (0) and after (1) the Young Offenders Act, to determine if the act had any effect on new stories of parental responsibility and juvenile delinquency. Using the same six search terms used to search the New York Times and one additional search term (Table 3-1), basic citation information was recorde d, including date of publication, article title, section/page a nd author. Next, as in the New York Times content analysis, the stance, context and presence of a law/fine were recorded. First, the stance of the article was recorded as (1) supports parental blame, (2) neutral on parental blame or (3) against parental blame. Next, the article was examined for c ontext and documented to include (1) a thematic reference to parenting and parent al effects on children or (2) an episodic reference to parental acts or encouragement contributing to juvenile delinquency. Fi nally, each story was coded to determine if a story contained a reference to or application of a parental responsibility law (0 = absent, 1 = present). Examples of coding definitions can be found in Table 3-2. Study I, Phase III: National Comparison Once each n ational paper content analysis was completed, a comparison of the two papers was conducted to determine if differences exist be tween Canada and the United States in reports
38 of parental responsibility. In an article deta iling the political and soci al changes surrounding the evolution of the Canadian juvenile justice syst em, Hartnagel (2004) deta ils the public insistence that Canadian juveniles be held responsible for their delinquent actions. Schissel (1997) discusses how Canadian society demands stern actions against j uvenile delinquents and how the youth have become the primary scapegoat for many societal problems. In the United States, conflicting opinions/objectives emerge between pub lic opinion and juvenile justice aims. Some scholars argue that U.S. citizens are looking for someone to take responsibility for juvenile delinquency (Thurman, 2003; Tomaszewki, 2005; Tyler & Segady, 2000), oftentimes placing blame on the parents. However, many studies re cognize that public demand often conflicts with the goals of the U.S. juvenile justice system (Brank, Hayes, & Weisz, 2006; Brank, Kucera, & Hays, 2005; Davidson, 1996; Dimitris, 1997; Ty ler & Segady, 2000), which seeks to hold juveniles responsible for their actions and prevent patterns of delinquency. Together, the research from Canada and the U.S. were adapted to develop the hypotheses for the New York Times and Globe and Mail comparison phase of the current study. Hypotheses H1: The New York Times will hav e more thematic and fewer episodic context stories on parental responsibility than the Globe and Mail (Adapted from Sprott, 1996). H2: The New York Times will have more stories that s upport holding parents responsible for their childs actions than the Globe and Mail (Adapted from Hartnage l, 2004; Schissel, 1997; Thurman, 2003; Tomaszewki, 2005; Tyler & Segady, 2000). H3: The New York Times will be less likely to include a reference to or application of a parental responsibility law than the Globe and Mail (Adapted from Sprott, 1996; Thurman, 2003; Tomaszewki, 2005; Tyler & Segady, 2000)
39 Strategy and Implementation The two national content analyses we re merged to create a single data set. Utilizing crosstabs, the context, stance, and reference to a parental responsibility law of each paper was compared to the other to measure whether the tw o papers differed in each dependent variable. Study II, Phase I: Flor ida Times Union Content Analysis Upon establishing a general/broad existence of parental blame in Study I, the next focus was to explore the presence of parental mention in specific juvenile crime stories. Study II is comprised of additional content analyses c onducted to detail juvenile crime accounts and establish the frequency and context in which the media includes parents in these stories. For the current phase of Study II, a content analysis of the Florida Times Union, a local US paper, was selected for analysis. The Florida Times Union was chosen for examination because it represented a large, local pa per ranking in the top 100 US newspapers based on measured circulation by the Audit Bureau of Circulation an d because of a recent enactment of a city curfew ordinance that that holds parents responsible for thei r childs violation (O rd. 2006-889-E, 2). Unlike Phase I, this section of the current study was created to examine in-depth coverage of juvenile crime. Instead of seeking broad ideas of parental responsibility, instances of parental mention within juvenile crim e stories were recorded. Hypotheses H1: Parents are likely to be m entioned in the majority of juvenile crime stories (Adapted from Chiricos, Eschholtz, & Gertz, 1997; Dorf man, Woodruff, Chavez, & Wallack, 1997; Sprott, 1996). H2: If a parent is mentioned in a juvenile crime story, the major ity of references are likely to be relevant to parental blame (Adapted from Book & Perala-Littunen, 2008; Sprott, 1996).
40 Strategy and Implementation The content analysis include d local juvenile crim e stories from January 2003 through December 2007. This time period was chosen to encompass the five year time period before the implementation of the city curfew ordinance related to parental responsibility and the year after its implementation. Findings within the Florid a Times Union will be compared before (0) and after (1) the curfew implementation, to determine if the or dinance influenced parental inclusion in news stories about juvenile delinquency. Using six search terms (Table 3-3), basic cita tion information was reco rded, including date of publication, article title, sect ion/page and author. Specific ca se details were also coded to disguise the true purpose of the local content analysis from the i ndependent coder. These details included the number of offenders/victims (numeric ), age (numeric) and sex of offenders/victims (0 = male, 1 = female), crime type (specific ch arge was recorder), and location (i.e., name of business, address of residence, sc hool). Finally, the presence of parental mention was recorded (0 = absent, 1 = present). A br ief detail statement of parental mention/involvement is included (Table 3-4). Study II, Phase II: Lo ndon Free Press Content Analysis To continue Study II, the current pha se extends the findings from the Florida Times Union to a large local Canadian newspaper, the London Free Press Again, the purpose of this study will be to examine the inclusion of parental co mponents in specific crime stories. For this content analysis, the London Free Press was chosen for its parallel ci rculation and format to the Florida Times Union. The newspaper is the 13th largest paper in Canada with a circulation rate of 115,000 daily and reports local and national stories to the residents of London, Ontario. Similar to Phase II, this section of the current study was created to examin e in-depth coverage of
41 juvenile crime. Instead of seeking broad ideas of parental responsibility, instances of parental mention within a juvenile crime story were recorded. Hypotheses H1: Parents are likely to be m entioned in the majority of juvenile crime stories (Adapted from Chiricos, et al, 1997; Dorfman, et al., 1997; Sprott, 1996). H2: If a parent is mentioned in a juvenile crime story, the major ity of references are likely to be relevant to parental blame (Adapted from Book & Perala-Littunen, 2008; Sprott, 1996). Strategy and Implementation The content analysis include d local juvenile crim e stories from January 2003 through December 2007. Using six search terms (Table 3-3), basic citation information was recorded, including date of publication, arti cle title, section/page and aut hor. Also, specific case detail, including number of victims and offenders (numer ic), ages of all persons involved (numeric), gender of victim/offender (0 = male; 1 = female ) and the crime in question (specific charge; assault, arson, etc), was record ed. Most relevant to the st udy were the two components measuring whether or not parents were mentioned in the story (0 = absent, 1 = present) and the qualitative description of the inclusion. Study II, Phase III: Local Comparison After m easuring the difference between the two national papers, an additional comparison analysis was conducted to measure differences between elements in the Florida Times Union and London Free Press. Sprott (1996) found juvenile crime reports in Canada rarely contained information of the juvenile offenders circumstance or the legal reasoning behind the juveniles disposition. Instead, many of the Canadian juvenile crime reports detailed the seriousness of the offense and victimization eff ects of the crime. Sprott also noted that the local newspaper stories rarely in clude dispositions and instead fo cused on the facts and effects of
42 the crime, allowing little print space for superf luous material. These findings were used to generate the hypotheses for comparing the Florida Times Union to the London Free Press Hypotheses H1: The London Free Press will have fewer counts of parental m ention in juvenile crime stories than the Florida Times Union (Adapted from Book and Pe rala-Littunen, 2008; Sprott, 1996; Thurman, 2003; Tomaszewki 2005; Tyler & Segady, 2000). H2: The London Free Press will have fewer general /thematic stories than the Florida Times Union (Adapted from Sprott, 1996) H3: The London Free Press will have fewer blame related references to parents than the Florida Times Union (Adapted from Sprott, 1996; Thurma n, 2003; Tomaszewki, 2005; Tyler & Segady, 2000) H4: The London Free Press will have fewer references to or application of a parental responsibility law than the Florida Times Union (Adapted from Thurman, 2003; Tomaszewki, 2005; Tyler & Segady, 2000) Strategy and Implementation Findings from the two local papers w ere merged into one dataset. Each story element, measured in the individual news paper content analysis, was tested to determine if statistically significant differences between the London Free Press and Florida Times Union existed. Intercoder Reliability System A consistent system for testing intercoder reliability was utilized throughout the studies. After each newspaper was coded by the primary re searcher, statistical software was used to randomly select a subset of 20% of the stories fo r intercoder reliability. An independent coder, unaware of study hypotheses, was trained to reco gnize and code the relevant elements of each selected story. Once the 20% was coded by the independent researcher, a crosstab test was
43 conducted to measure the Kappa score for each coded variable. The intercoder reliability statistical standard util ized in this study was a kappa score of .80 (Krippendorf, 1986). After the kappa was calculated, the primary researcher and independent c oder discussed any discrepancies and made minor adjustments to reach agreement. The kappa results for each content analysis are described below. For the New York Times a subset of 121 articles was ra ndomly selected and recoded to measure intercoder reliability for each variable. Context measured at .92, stance reliability was .90 and reference to a parental re sponsibility law was scored w ith a kappa of .83. All of the variables surpassed the predetermined signifi cance level of a kappa score higher than .80. Intercoder reliability for the Globe and Mail was measured on a randomly selected subset of 35 articles. Context of the st ories, either thematic or episodic, was measured at .80, stance (supporting, neutral, or against parental blaming) was .85 and references of a parental responsibility law was scored w ith a kappa of .89. All of the variables met or surpassed the predetermined significance level of a kappa score of .80. When measuring intercoder reliability of the Florida Times Union content analysis, a randomly selected subset of 172 articles was re coded. Parental mention received a kappa score of .96. When measuring whether or not a story included a thema tic or episodic reference to parental responsibility, 100% agreement between re searchers was recorded. Type of parental mention produced a kappa score of .84, while the me ntion of a law or fine was unable to obtain a kappa score due to all values be ing constant. Both variables e ligible to be measured by the kappa statistic surpassed th e desired kappa score of .80. To ensure reliability within the London Free Press content analysis, these stories were measured for intercoder reliability. A randomly selected group of 76 stories was selected for
44 testing. The inclusion of pare nts received a kappa score of .81. When determining whether a storys context was thematic or episodic in nature, the kappa score was 1.00. The type of parental inclusion was measured by a kappa scor e of .90. Finally, the presence of a parental responsibility law was tested to obtain a kappa score. 100% c oder agreement was achieved when measuring the presence of a reference to a parental responsibility law. All variables of interested received a score above the desired level of .80.
45 Table 3-1. Search terms and productivity of national content analyses Search term NYT retrievals NYT applicable G&M retrievals G&M applicable liability AND parent 1401 96 1451 41 p arent AND contrib.* 8713 189 4 0 parent AND encourage* 7059 85 5853 ^ parent* pre/5 responsib* AND juvenile 217 149 0 --p arent pre/5 involv* AND deliquen* 11 6 0 --juvenile pre/5 crime pre/10 parent 20 16 12 0 parent AND deliquen* ----374 94 Notes: Includes original arti cles only. No duplicates are included in these numbers; Mom and dad were substituted for parent but yielded unbeneficial results. ^ = This search term was deleted after 500 stories were read for anal ysis and lacked any applicable articles. The term was replaced with parent AND delinquen*. New search term did not contribute to the New York Times analysis and therefore was not included. Time frames: NYT 1899-2004; G&M 1978-2007
46 Table 3-2. Examples of coding definiti ons used for national content analyses Element Example Stance Supports parental blame Separated parent s contribute their fu ll quota to a childs delinquency.a Neutral parental blame When a youngster gets ticketed, the police send parents a notice Parents must appear, too.b Against parental blame Only society can meet the challenge of the hour in planning for understanding parents of the future and providing a safe and adequate community in which children can be reared.c Context Thematic We can make it clear to men who father children out of wedlock but dont want to take care of them, the responsibility of the child is not that of the mother, it is not that of the stated Episodic In referencing a specific family and their children, the following comment was include d: They go out and say, Why thats a Spic, and the little child is only a carbon copy of the parent.e Reference to PR law Present After a jurys verdic t of guilty, a judge fined the Provenzinos $100 each and ordered each to pay $1,000 in court costs for their failure to properly supervise their son.f Absent --Notes: examples taken from NYT stories a nd utilized throughout both national content analyses; a = Juvenile Delinquency: A Variety of Views; Causes Cures; May 29, 1955; Frances Rodman; 143. b = Danger on Two Wheels, Mar 19, 1978, Edward R. Walsh, LI20. c = The Real Delinquents -Parents or Soci ety? Feb 16, 1947; Nochem S. Winnet, SM15. d = A Strong Warning That Moral Decay Is Basic Trouble Facing the Nation; 'We need people to stand up to the special-interest groups. Jan 14, 1988.; Pat Robertson. p. A20 e = Youth Crime Tied to Racial Hatred; Oct 12, 1959; Leibowitz Blames; 39. f = Parents Convicted for a Youth's Misconduct; May 10, 1996; Robyn Meredith, A14.
47 Table 3-3. Search terms and produc tivity of local content analyses FTU retrievals FTU applicable FTU duplicates LFP retrievals LFP applicable LFP duplicates Juvenile AND crime 854 365 --443 167 --Juvenile AND accused 247 74 51 136 20 23 Juvenile AND arrest 696 177 261 215 23 83 Teen AND crime 854 164 165 751 71 89 Teen AND accused 287 34 116 223 17 32 Teen AND arrest 754 46 542 415 17 83 Notes: child was substituted for juvenile and teen for each search term but yielded unbeneficial results (majority of stories involved child pornogra phy or child abuse); Several stories also included 18 and 19 year-o lds as teens/juveniles yet they are not eligible for juvenile status; Dates rang ed from January 1, 2003 through December 31, 2007 for both papers
48 Table 3-4. Examples of coding definiti ons used for local content analyses Element Example Context Thematic Why are parents in most cases, single parents, which is another part of the problem letting their young teenagers roam around the neighborhoods at those hours? a Episodic The children, who were not iden tified because they are juveniles, said their mother would take the narcotic medications and give them any pills she did not want, investigators said.b Parental Responsibility Type Omission "Never leave your child because a child needs both a mother and a father," 17-year-old "Kyle," who is ch arged with murder, told Burney. "I know my life would be different if my father was around. I have not seen him in three years. When he was with my mother I was a straight-A student. He left when I was 7 and I had to become the man of the house. ... A mother cannot play the male ro le. I started smoking weed when I was 7 years of age -my cousin and my brother got me started ..."c Commission Police said two girls, 14 and 15, were fighting on a school bus at 7:15 a.m. The parent of the 15-year-old ha d come to the school to confront the younger girl about harassing her daughter, police said. When the mother saw the fight as the bus pulled in, she got on board and confronted the 14-year-old. Police said the woman hit the younger girl and shoved her into a seat as other students and the bus driver watched.d GeneralSupports PB "Parents need to be involved in what their kids do," the detective said. "They need to check out their [children's] activities outside the home."e GeneralNeutral on PB Some members of the community say the parents are to blame, while others say there is no way the parents could have prevented this tragedy.f Irrelevant to PB A 14-year-old Glynn County boy has been charged as an adult in Saturday night's shooting death of his stepfather.g Workman's mother declined comment.h Reference to PR law Present Thornal said the parents of the two youths could be responsible for paying for the damages.i Absent --Notes: examples taken from FTU and utilized through both local content analyses. PB = Parental blame. a = TeenagersParents should be in charge. Letters from the Readers. July 31, 2007; The Times-Union b = Police say mom used kids to get her drugs; She's accused of having them lie to doctors so she could get prescription narcotics. January 30, 2007; Gordon Jackson, A-1. c = CITY SOJOURN: Jail tales aim to help readers get involved. November 2, 2005. Tonyaa Weathersbee. d = School fights brought in hand. December 18, 2004. Dan Scanlan. e = Teenagers' fun can escalate into crimin al behavior. April 16, 2005. Deborah Cearnal. f = Example created by researcher. The FTU did not have any stories coded as general/neutral on PB. g = Teen faces charges as adult in shooting. December 17, 2007. Teresa Stepzinski. h = Youth faces 40 years in prison for mu rdering woman in 2005. January 5, 2007. Paul Pinkham. i = 14-year-old boys charged in middle school vandalism. October 6, 2006. Gordon Jackson.
49 CHAPTER 4 RESULTS Study I, Phase I: New Y ork Times Content Analysis Results of the New York Times content analysis are detailed in Table 4-1. The following variables were measured in each coded story: cont ext (thematic or episodic reference to parental responsibility), stance (supporting, ne utral, or against parental blam e), and reference to a parental responsibility law (present or absent). Exampl es of each coding definition are provided in the forthcoming paragraphs and equivalent defini tions were applied to each newspaper content analysis. The New York Times content analysis yielded 541 rele vant articles, approximately 5 stories per year, with the most stories from 1947 and 1954 (n = 23 per year). In each story, the level of support for parental blame assignment was measured. The stance was coded as support, neutral or agains t parental blame. Parents should be held responsible when a child that young commits such a vicious crime would be an example of support for parental blame. However, a statemen t such as It is not fair to hold parents responsible for delinquent acts that do not occur in their presence is an example of a statement against parental blame. Stories coded as neutral provided both si des of the parental responsibility argument or simply described a law without any opinionated statements. Hypothesis 1 indicated support for parental blame was likely in the ma jority of coded stories. A majority (73%, n = 395) of the stories demonstr ated support for blaming parents in cases of juvenile delinquency, 21% (n = 111) were neutra l and 7% (n = 35) included a stance against parental blame, thus corrobor ating hypothesis 1. Another element analyzed was the context of the story, coding each either episodic/specific or thematic/general reference to parental respons ibility. An example of thematic context would be parents should be more cautious of the peers their children associate with to prevent juvenile
50 delinquency and drug use. Episod ic framing would be included via a statement such as the boys mother admitted she did not pay attention to the friends her son associated with, and had no knowledge of his involvement with drugs. H ypothesis 2 predicted the context of the stories will be more likely to be thematic than episodic in the national media. More than 93% (n = 505) of the stories included thematic/broad references to parent al responsibility and the role or parenting in juveniles deli nquent behaviors, thus supporti ng hypothesis 2. The remaining stories (n = 36) included an episodic refere nce to a delinquent act committed by a minor. Of those stories analyzed, a reference to a specific parental res ponsibility law was only present in approximately 20% (n = 106) of the stories and absent in the other 80% of the stories. Hypothesis 3 stated that a story co ntaining reference to or applica tion of a parental responsibility would more likely be against parental blame than a story where a law was not included. A chisquare analysis was conducted to test for significance. The pr oportion of stories that included a reference to a parental responsib ility law and also in cluded statements supporting parental blame was .65 whereas .75 of the stories supported pare ntal blame when there was no mention of a parental responsibility law ( X2 = 6.53, p < .05, df = 2; See Table 2-2). Study I, Phase II: Globe and Mail Content Analysis Table 4-1 details the results of the co ntent analysis of the Globe and Mail Replicating the New York Times phase, context, stance, and reference to a parental responsibility law were measured in each story. The Globe and Mail content analysis yielded 144 articles, approximately 5 stories per year, with 1999 having the most stories (n = 15), The first hypothesis stated the majority of stories would be in support of parental blame. When examining the stance of each story, 13% (n = 18) of the stories were against parental blame, 30% (n = 43) were neutral, and 58% (n = 83) supporte d some type of parental respons ibility for juveni le behavior. Within the analyzed stories, approximately 42% (n = 61) contained a reference to a parental
51 responsibility law, while the remaining 58% (n = 83) did not have a reference present. Due to the fact that more than 57% of the stories supported some type of parental blame, hypothesis 1 was supported. Hypothesis 2 stated the majority of stories would have thematic context. When examining the context, 82% (n = 118) of the stories included thematic/broad references to parental responsibility a nd the role or parenting in juveniles delinquent behaviors, while 18% (n = 26) provided an episodic reference to pare ntal contribution, thus supporting hypothesis 2. Replicating the New York Times content analysis, the effects of the presence of parental responsibility law reference on th e stance of a story was tested using Chi-square analysis. Hypothesis 3 stated if a story cont ained a reference to a parental responsibility law or fine, there was less likely to be support for parental blam e than if no law holding parents responsible was present in the story. The proporti on of stories that included statements supporting parental blame and a reference to a parental responsibility law was .30 comp ared to a proportion of .78 supporting parental blame when a reference to a parental responsibility law was not included in the story, thus supporting hypothesis 3 ( X2 = 36.63, p < .01, df = 2; See Table 4-3). Finally, researchers examined whether or not the introduction of the Youth Offenders Act in 1984 influenced the context, stance or presence of a parental responsibility law in the coded Globe and Mail news stories. The results of the Globe and Mail were analyzed using Chi-square analysis, comparing coded articl es from pre-1984 to coded articl es published during or after 1984. No statistically significant differences we re found, suggesting the Youth Offender Act did not significantly influence the content of pare ntal responsibility stories. Details of the comparison can be seen in Table 4-4. Study I, Phase III: National Comparison To com pare findings from the New York Times in the United States to Globe and Mail in Canada, two statistical measures were employed. First, to measure the differences between
52 coded elements of the two papers, a multi-sample chi-square test was employed. For categories that did not have 5 or more obs ervations, Fishers Exact Test wa s utilized instead of the chisquare. This test adjusts for small samples and approximates chi-square as the frequencies increase. Completed details of the comparison betw een the two papers can be seen in Table 4-5. Hypothesis 1 predicted that the New York Times would have a greater number of thematic stories on parental responsibility and a fewer number of stories with episodic context than the Globe and Mail The difference between the two nationa l papers was statistically significant when examining story context (t hematic or episodic). The New York Times utilized thematic references to the parenting role and effects on juve nile behavior in 93% (n = 505) of the stories, compared to 82% (n = 118) of stories in the Globe and Mail supporting the predictions in hypothesis 1. Fishers Exact Test was utilized in this analysis because all frequencies did not exceed 50. The test illustrated non-random associ ations between the newspaper and context was significant (p < .01). To compare the stances of the two national newspapers, Hypothesis 2 stated that the New York Times would have more stories that s upported parental blame than the Globe and Mail The proportion of stories that included statem ents supporting parental blame was .73 in the New York Times compared to a proportion of .58 supporting parental blame in the Globe and Mail ( X2 = 13.61, p < .01, df = 2; See Table 4-5). Convers ely, 13% (n = 18) of stories in the Globe and Mail were against parental blame, while 7% (n = 35) of New York Times stories were against blaming parents in cases of juvenile deli nquency. Consistent with hypothesis 2, the New York Times stories supported parental blame in juvenile behavior significantly more often than the Globe and Mail
53 Finally, hypothesis 3 proposed that the New York Times was less likely to include a reference to or application of a pare ntal responsibility law than was the Globe and Mail Again the Fishers Exact Test was used to test for non-random associati ons between the reference of a parental responsibility law and the two newspape rs, and resulted in a si gnificant association ( p < .01). The Globe and Mail included a law or fine in 58% (n = 61) of its stories on parental responsibility. The New York Times mentioned a law or fine in 20% (n = 106) of stories on parental responsibility. Hypot hesis 3 was supported, as the New York Times had significantly fewer stories that detailed a parental responsibility law compared to the Globe and Mail Study II, Phase I: Flor ida Times Union Content Analysis Detailed results of the Florida Times Union content analysis are displayed in Table 4-6. In addition to specifi c information about the offender a nd crime, researchers recorded detailed information about pa rental inclusion in the j uvenile crime reports. The Florida Times Union content analysis yielded 860 app licable articles that had at least one of the search terms present, meaning per year there was an average of 172 juvenile crimes articles. Hypothesis 1 predicted that parents were likely to be menti oned in the majority of specific juvenile crime stories. A parent was mentione d in 17% (n = 144) of the stor ies, thus hypothesis 1 was not supported. Of those with parental inclusion, 2% were thematic re ferences to parents, while 98% (n = 713) provided an episodic reference to one or more of the defendants parents. The type of parental inclusion was also detailed. Hypothesis two stated that if a parent was mentioned in the story, the majority of parental references would be relevant to parental blame. The majority of stories, 86%, (n = 124) included a reference to a parent th at was not relevant to parental blame; therefore, hypothesis 2 was not supported. Irrelevant fact s included information such as a parent picked up the delinquent from j uvenile detention, the parent was a victim of the crime, the parent was present in the courtroom or the parent failed to comment on the situation.
54 However, 5% (n = 7) detailed parental omission in the crime while 5% (n = 7) of the stories detailed the parent being invol ved in the commission of the crime. An example of omission would be statements such as the mother had no t spoken to her son in 2 days and was not aware he and his friends were selling dr ugs or Katie said had her fath er been more involved in her life, the crimes may not have occurred. C onversely, parental mention would be coded as commission if statements such as the daughter and mother duo were arrested and charged with armed robbery or the father and son were caught selling drugs were incl uded in the story. In addition to the specific references, the general re ferences to parents generated 5% (n = 7) of stories which supported parental bl ame. Finally, a reference to a parental responsibility law was present in only 2% (n = 3) of the juvenile crime stories. Researchers examined whether or not the intr oduction of the juvenile curfew ordinance, a law implemented in 2007 that ho lds parents responsible when their child breaks the citys curfew, had any influence on reports of j uvenile delinquency. The results of the Florida Times Union were analyzed using Chi-square analysis, comparing coded articles from pre-curfew implementation (printed prior to 2007) to stor ies published post-curfew implementation (printed in 2007). The analysis did not yi eld any statistically significant results, leading researchers to conclude the curfew ordinance did not have a strong influence on the content of news stories focusing on juvenile delinquency cases. Details of the comparison can be seen in Table 4-7. Study II, Phase II: London Free Press Content Analysis Table 4-6 contains detailed results from the London Free Press content analysis. The London Free Press content analysis was designed to replicate the content analysis of the Florida Times Union Again, parental inclusion, thematic versus episodic context of stories, details of parental mention and reference to a pare ntal responsibility law were coded. The London Free Press content analysis yielded 315 articles, approximately 63 juvenile crime stories per year.
55 Hypothesis 1 again predicted that a parent was likely to be mentione d in the majority of juvenile crime stories. Parents were mentioned in 21% (n = 65) of the juvenile delinquency stories, while almost 80% (n = 250) did not have parental inclusion, rejecting pred ictions put forth in hypothesis 1. Of the stories with parental mention, 91% (n = 59) corresponded to an episodic reference to a parent. The type of parental descripti on varied with 8% (n = 5) involving omission and 8% (n = 5) including details of commission in a crime. Positive support for parental blame was illustrated in 14% (n = 9) of the stories. The second hypothesis stated that when parents were included in stories, the majority of the parental references would be relevant to respons ibility/accountability of the parent. More than half (67.7%, n = 44) of th e stories mentioned a parent but had no relevance to parental involvement/responsibility in the crime, rejecting the predictions in hypothesis 2. For example, these 44 stories either included the parent as the victim, discussed the parents presence in court or noted the parent failed to comment. When a parent was linked to the juveniles delinquent behavior, 97% (n = 63) of the stories did not have any reference to a law that was or could be imposed on the parent. Study II, Phase III: Local Comparison Again, m ultiple sample chi-square tests and Fishers Exact Tests were employed to compare the frequencies of the Florida Times Union to London Free Press. Analyses were conducted to compare the presence of parental men tion (present or absent), context (thematic vs. episodic), type of parental inclusion (omission, co mmission, general or irrelevant to blame) and presence of reference to a pare ntal responsibility law (present or absent) between the two local newspapers (Table 4-8). Statistically significant associations be tween the two papers were not observed for parental inclusion or reference to or application of a parental responsibility law; thus, hypotheses 1 and 4 were not supported.
56 When testing the relationships between the news papers story context, Fishers Exact Test were found to be significant ( p < .05). There is a statistically significant difference in the papers reporting context. Within the Florida Times Union 98% (n = 143) of the stories included an episodic reference to parents, while th e 91% (n = 59) of the stories in the London Free Press used an episodic context. This finding is the opposite of hypothesis 2 that stated the London Free Press would have fewer thematic context stories than the New York Times Chi-square analyses revealed that the type of parental in clusion differed between the two local papers, yet assumptions for the Chi-square analysis were violat ed and therefore this significance can not be reported. However, the conclusions that can be drawn from this analysis should be noted. The largest difference can be seen in the newspapers inclusion of parents without any relevance to responsib ility in the crime. In the Florida Times Union 86% (n = 124) of the stories mentioned parents, but did not include any relevance to parental responsibility, compared to 68% (n = 44) of the stories in the London Free Press. This finding is inconsistent with hypothesis 3, which stated the London Free Press would have a larger portion of irrelevant parental inclusions than the Florida Times Union
57 Table 4-1. Frequencies distributions of the na tional content analyses Story Element NYT-freqNYT-%G&M-freq G&M-% Context Thematic 50593.3118 81.9 Episodic 366.726 18.1 Stance Supports parental blame 39573.083 57.6 Neutral on parental blame 11120.543 29.9 Against parental blame 356.518 12.5 Reference to PR law Present 10619.661 42.4 Absent 43580.483 57.6 Table 4-2. Reference to PR law and stance crosstab results for New York Times US Crosstab results Frequency Adj. res. Percent Reference to a parental re sponsibility lawpresent*** Supports blaming of parents 69-2.0 65.1 Neutral blaming of parents 250.9 23.6 Against blaming of parents 122.3 11.3 Reference to a parental responsibility lawabsent*** Supports blaming of parents 3262.0 74.9 Neutral blaming of parents 860.9 19.8 Against blaming of parents 23-2.3 5.3 *** p < .05 Notes: ChiSquare = 6.53 df = 2 p < .05 Table 4-3. Reference to PR law and stance crosstab results for Globe and MailCanada Crosstab results Frequency Adj. res. Percent Reference to a parental re sponsibility lawpresent* Supports blaming of parents 18-5.9 29.5 Neutral blaming of parents 335.4 54.1 Against blaming of parents 101.2 16.4 Reference to a parental responsibility lawabsent* Supports blaming of parents 655.9 78.3 Neutral blaming of parents 10-5.4 12.0 Against blaming of parents 8-1.2 9.6 p < .001 Notes: ChiSquare = 36.63 df = 2 p < .001
58 Table 4-4. Comparison pre-YOA to post-YOA for Globe and Mail Canada Type of parental detail PRE-freq Adj res % POST-freq Adj res % Context Thematic reference 23 0.5 85.2 95 -0.5 81.2 Episodic reference 4 -0.5 14.8 22 0.5 18.8 Stance Supports blaming of parents 18 1.1 66.7 65 -1.1 55.6 Neutral blaming of parents 8 0 29.6 35 0 29.9 Against blaming of parents 1 -1.5 3.7 17 1.5 14.5 Reference to a parental responsibility law Present 7 -1.9 25.9 54 1.9 46.2 Absent 20 1.9 74.1 63 -1.9 53.8 Notes: No elements were significant at any accepted level of significance. Table 4-5. Comparison between New York Times and Globe and Mail National Type of parental detail NYT-freq Adj res % GM-freq Adj res % Context* Thematic reference 505 4.2 93.3 118 -4.2 81.9 Episodic reference 36 -4.2 6.7 26 4.2 19.1 Stance** Supports blaming of parents 395 3.6 73.0 83 -3.6 57.6 Neutral blaming of parents 111 -2.4 20.5 43 2.4 29.9 Against blaming of parents 35 -2.4 6.5 18 2.4 12.5 Reference to a parental responsibility law* Present 106 -5.7 19.6 61 5.7 57.6 Absent 435 5.7 80.4 83 -5.7 42.6 ** p < .01 p < .001 Notes: ContextFish ers Exact Test p < .001; StanceChiSquare = 13.61 df= 2 p < .01; Reference to a PR Law Fishers Exact Test p < .001. Table 4-6. Frequencies distributions of the lo cal content analyses Story element FTU-freqFTU-%LFP-freq LFP-% Parental Mention Yes 14416.865 20.6 No 71383.2250 79.4 Context Thematic 32.1 6 9.2 Episodic 14397.959 90.8 Parental Responsibility Type Omission 74.8 5 7.7 Commission 74.8 5 7.7 GeneralSupports parental blame 74.8 9 13.8 General-Neutral on parental blame -----2 3.1 Irrelevant to parental blame 12485.544 67.7 Reference to PR law Present 32.1 2 3.1 Absent 14297.963 96.9
59 Table 4-7. Comparison between pre-curfew & post-curfew Florida Times Union-Local Type of parental detail PRE-freq Adj res % POST-freq Adj res % Parental mention Yes 120 -0.2 16.7 24 0.2 17.3 No 598 0.2 83.3 115 -0.2 82.7 Context Thematic reference 2 -0.7 1.7 1 0.7 3.7 Episodic reference 117 0.7 98.3 26 -0.7 96.3 Parental responsibility type Omission 5 -0.7 4.2 2 0.7 7.4 Commission 4 -1.7 3.4 3 1.7 11.1 Generalsupports blaming of parent 5 -0.7 4.2 2 0.7 7.4 Irrelevant to parental blame 104 1.9 88.1 20 -1.9 74.1 Reference to a parental responsibility law Present 3 0.8 2.5 0 -0.8 0 Absent 115 -0.8 97.5 27 0.8 100 Notes: No elements were significant at any accepted level of significance. Table 4-8. Comparison between Florida Times Union and London Free Press Local Type of parental detail FTU-freq Adj res % LFP-freq Adj res % Parental mention Yes 144 -1.5 16.8 65 1.5 20.6 No 713 1.5 83.2 250 -1.5 79.4 Context*** Thematic reference 3 -2.4 2.1 6 2.4 9.2 Episodic reference 143 2.4 97.9 59 -2.4 90.8 Parental responsibility type Omission 7 -0.8 4.8 5 0.8 7.7 Commission 7 -0.8 4.8 5 0.8 7.7 Generalsupports blaming of parent 7 -2.3 4.8 9 2.3 13.8 Generalneutral blaming of parents 0 -2.1 0.0 2 2.1 3.1 Irrelevant to parental blame 124 3.0 85.5 44 -3.0 67.7 Reference to a parental responsibility law Present 3 -0.4 2.1 2 0.4 3.1 Absent 14 0.4 97.9 63 -0.4 96.9 *** p < .05 Notes: Context Fishers Exact Test p < .05; Parental responsibility type, Parental mention and Reference to a PR law were not si gnificant at any accepted p-value.
60 CHAPTER 5 DISCUSSION Societal expectations for successful parenting are not new, and it is no surprise parenting issues are being noted in the m edia. There ar e numerous learning and support mechanisms like books, how-to-guides, videos and networking groups solely devoted to good parenting. Most societies have general guidelines for being a go od parent. For example, the majority of people believe a parent should set a good example, be invol ved in his or her chil ds life, establish and enforce rules, teach morals and values and provide care/safety. Judea Christian belief systems instruct parents to Teach your children to choos e the right path, and when they are older, they will remain upon it (Proverbs 22:6). 1 Timothy 5: 8 states But if any provide not for his own, especially for those of his own house, he hath deni ed the faith, and is worse than an infidel. With parenting comes the social expectation of care, but more recently law has begun to define good parenting. Parental responsibility laws are becoming more popular, placing the blame on parents whose children break the law, and it is possible that the media is generating support and influence the creation of such legislation. Our study focused on the prevalence and type of media attention to parental responsibility issues. Through four content analyses, results de monstrated a presence of parental blaming in print media. Several details em erged about the context, stance, parental responsib ility type and presence of law/fines in these stories. In addition to these contextual elements, comparisons were made between United States and Canadian print media on both the local and national level. The following section discusses such findings and possible implications for this body of research. In both of the national content analyses, three hypotheses were tested. First, the current researcher predicted that there would be a positive slant in support of pare ntal responsibility.
61 Next, it was hypothesized that the co ntext of the majority of stories would be thematic (general) rather than episodic (specific). Finally, if a story contained re ference of a specific parental responsibility law, the story w ould be less likely to support pare ntal blame. Results from the New York Times and Globe and Mail supported each proposed hypothesis. Next, the two national content analyses were compared. Three hypotheses were tested: the New York Times will be more thematic and fewer episodic context stories than the Globe and Mail, the New York Times will have more stories that support blaming parents than the Globe and Mail and the New York Times will be less likely to include a reference to or application of a parental responsibility law than the Globe and Mail All three hypotheses were supported. The local content analyzes, Florida Times Union and London Free Press, tested two hypotheses. First, it was expected that parents would be mentione d in the majority of juvenile crime stories. Next, the current researcher predic ted that if a parent was mentioned in a story, the majority of parental inclusions were likely to be relevant to pa rental blame. Neither hypothesis was supported in the US or Canadian local newspa per. Instead, very few stories included any element of parental mention. If a parental component was present in a st ory, it was irrelevant to blame a majority of the time. The final analysis compared the two local papers. The current research tested the following four hypotheses: the London Free Press will have fewer counts of parental mention in juvenile crime stories than the Florida Times Union, the London Free Press will have fewer general/thematic stories than the Florida Times Union the London Free Press will have fewer blame related references to parents than the Florida Times Union and the London Free Press will have fewer references to or application of a parental responsibility law than the Florida Times Union. None of the hypotheses were supported. No significant associations were seen when
62 examining parental inclusion or reference to a parental responsibility la w, failing to support two hypotheses. Tests comparing the context of both papers and the type of parental inclusion resulted in findings opposite from those predicted in the hypotheses. Agenda setting theory states that the medi a can have a signifi cant influence over its subscribers based on what stories it deems ne wsworthy and how much coverage the topic receives. Past research demons trates that the media has the power to influence what the public thinks about on a daily basis (Iyengar, et al., 1982; Kovera, 2002; Lowr y et al., 2003). The New York Times content analysis found that 93.3% of the pa rental responsibility stories were general references to parental influen ce without specific examples. It is possible (although not measured in the current study), that the coverage of parental responsibi lity could be gui ding public opinion and potential legislators, to recognize and accept the idea of parental responsibility, even when specific examples of parental blame are not in cluded, according to agenda setting and media research. At the very least, the current study conf irms that the media is depicting juvenile crime and parental responsibility as a newsworthy issue (Heath & Gilbert, 1996; Lowry et al.; Romer). At the national level, the New York Times content analysis produced several important findings. A great majority of st ories that included mention of a parent were also blaming the parents. The overwhelming support for parental bl ame in the national media confirms the idea that the United States, specifically the media, are eager to attribute blame to the parents in cases of juvenile delinquency (Thurman, 2003; Tomaszewki, 2005; Tyler & Segady, 2000). Interesting historical trends were found in the New York Times content analysis. This content analysis was the most e xpansive, covering more than 100 years of media coverage. Two major peaks were observed in the anal ysis of the coded stories from the New York Times The first and largest spike in storie s on parental responsibility and juvenile delinquency occurred in
63 the late 1940s and continued th rough the 1950s. Stories focu sed on the shifting roles of a mother and father during this time frame. Father s were leaving their fami lies and going to war, while mothers were beginning to pursue jobs outsi de the home. The idea of both parents leaving the home was linked to problems of juvenile deli nquency and was heavily discussed in the coded articles. The second peak can be seen in stor ies coded between the late 1980s through the mid 1990s. This period was filled with stories a bout the decline of youth in the 1990s and the growing concern of juvenile deli nquency linked to recent high school shootings. Each of the analyzed newspapers was examined for historical trends but did not ge nerate any significant findings. The Globe and Mail content analysis examined Canadian national media attention given to parental responsibility. Again, the majority of national stories (57%) demonstrated support for parental blame in juvenile behavior. Both the New York Times and Globe and Mail found support for parental blame, a result that coul d have possible implications based on similar research. Although the current study measures media coverage, surveys conducted by Kenny and Kenny (1961) and Brank and Weisz (2004) f ound similar results when measuring public opinion on the subject of parental responsibil ity. Kenny and Kenny (1961) found 89% of survey respondents thought parents shoul d accept the respons ibility for their childs delinquent behavior. Additional support for parental re sponsibility was found in a study conducted by Brank and Weisz (2004) where 70% of respondents he ld both the juvenile and parent responsible when a juvenile crime was committed. Once the national papers were analyzed i ndividually, the two national papers were compared. Within the national papers, significant differences were observed when comparing the context of the New York Times articles to the context of the Globe and Mail. When stories of
64 parental responsibility were observed, the United States was less likely than Canada to reference a specific act/encouragement or lack of parental guidance that lead to juvenile delinquency. One possible explanation for this findi ng could be that in Canada, public opinion stipulates juveniles must take responsibility for thei r actions and be accountable for the consequences of delinquent behavior (Hartnagel, 2004; Schi ssel, 1997). Conversely, in the U.S., the concept of parental responsibility may satisfy the need to blame someone for a societal problem (Thurman, 2003; Tomaszewki, 2005; Tyler & Segady, 2000) on a thema tic/general level. However, people are less likely to blame parents in episodic/specific crime st ories (Brank et al, 2006). Framing theory states that the way in which a story is presented or framed can structure the way the public interprets the prov ided information. This theory is important in understanding the possible implications as the current st udy reveals 73% of the stories in the New York Times and 57.6% in Globe and Mail were in support of blaming parents. If the article referenced a parental responsibility law, the stance of the story was more likely to be against parental blame in both countries. Conversely, if the stance was in fa vor of holding parents resp onsible, a reference to a specific parental responsibility law was more like ly to be absent from stories in both papers. This finding is consistent with Brank, et. al (2006), who found gene ral support for parental responsibility but did not find public support fo r punishing parents in specific juvenile delinquency scenarios. While Brank and he r colleagues tested pub lic opinion through constructed vignettes, the curren t study reinforces the finding of public support in general situations in national media articles. In additi on to strengthening the argument in Brank et al., the current findings generate a need for resear ch on the effect of media on public opinion of parental responsibility. The goal of examining medias influence on parental responsibility attitudes could be achieved with an experiment similar to the future plans described in
65 forthcoming paragraphs, that manipulates the parental blame element in a juvenile crime story and measures how likely a person is to blame a delinquents parent based on the context and format of the news story. Regardless of the presence of a parental resp onsibility law, it should be noted that the New York Times articles illustrated support of parental blame significantly more than the Globe and Mail The comparative analysis of the national pape rs revealed a significant difference in the presence or absence of a law/fine. The Globe and Mail provided greater accounts of holding parents legally accountable by surpassi ng the number of articles in the New York Times that had included reference to a parental responsibility law. Evident by the lower levels of support for parental blame, the Canadian media appears to be more cautious to place blame on parents (Hartnagel, 2004; Schissel, 1997) and therefore, unlike the United States, provides legal justifications to many of its stories on the subject. Previous literature found that the more freque nt juvenile crime stories exist, the more likely juvenile delinquency is to be considered a major social problem (Chiricos, Eschholz, & Gertz, 1997; Dorfman, Gregory, Schmidt, & Bi erck, 1997). To measure the frequency of juvenile crime reports and the relationship between parental responsibilitie s, local papers were also analyzed for accounts of delinquency. Less than 20% of stories about juvenile crime in the Florida Times Union mentioned any reference to parents. If a parent was mentioned in a story, the parental inclusion was irre levant to blame in 85.5% of the stories. Tomaszewski (2005) established that high profile crim es were likely to generate ex tensive media coverage, bringing attention to parental responsibility in juvenile crimes. The curre nt study finds that when stories are not high profile, and instead are local juvenile crime report s, parental re sponsibility and blame are not likely to be recognized.
66 To replicate the content analysis of the Florida Times Union the London Free Press a Canadian local paper, was analyzed. Again, onl y a small portion (20.6%) of the juvenile crime reports included a parental mention. Of those ar ticles that did include a reference to parents, 67% of the inclusions were irrelevant to blame/re sponsibility. This finding furthers the idea that while high profile crimes generates mass cove rage and support for parental responsibility (Tomaszewski, 2005), local crime reports do not often include references to parental responsibility. Finally, in the local reports of juvenile crime, the Florida Times Union was more likely to have irrelevant parental information than was the London Free Press Again, this finding adds to the concept that Canada is more cautious to bring attention to the parents of a juvenile delinquent unless they specifically contributed to or influe nced the delinquents be havior, because of the emphasis on the juveniles own responsibil ity (Hartnagel, 2004; Schissel, 1997). Several possible limitations are recognized in the current stu dy. First, a limited set of search terms were utilized in the current study and it is recognized that other useful terms could exist. The search terms for each content analysis were carefully and systematically selected for use and tailored for the goals of each search. The current researcher tested synonyms and alternative words to increase produc tivity prior to creating the final list of search terms for each analysis. Next, limitations arise when examining the selected media source s. Two local and two national papers were selected fo r analysis, and it is possible the results would vary if different titles had been examined. Additional newspapers could certainly be coded to extend/support the current findings. Another limitation is that comp arisons were limited to the United States and Canada to explore international differences in media portrayals of pa rental responsibility. Further work, however, could explore the media messages from other countries with parental
67 responsibility laws or contrast to the media messages from countries without parental responsibility laws. Also, the time frames for th e two national papers were not the same due to limited database access, which could slightly sk ew results. Finally, the current researcher utilized frequency counts to meas ure the results of the findings. Replications of the study should consider the proportion and not onl y frequency of relevant news stories, as this could a misleading representation of the newspapers content. Future research may be utilized to address th e current studys limitations. An examination of medias effects on parental re sponsibility legislation would be beneficial to understanding the growing number of parental responsibility laws. Different newspaper titles and other countries media may be of interest when replicating the current study. Most importantly, a test that establishes the effects of media on public opinion of parental blame should be conducted to test the notion that the publics apparent support of parental responsi bility is based on the medias cultivation of that attitude (Shepherd, 1999). The primary researchers future plans include employing an experimental design to test the media effects on fear of crime and parental blame assignment on public opinion, in both television and print media. To measure media eff ects on participants perceptions, 10 versions of an edited juvenile crime news story will be used One version of the crime story will be included in each experimental cell, and will vary in medium (print or video) and parental mention (excluded, quote, parental fact, parental encouragement, or comb ination of all three parental manipulations). One parental manipulation element will incorporate an emotional quote that will inquire as to how the parent could have le t the delinquency happen, and how the community member being quoted raised her child to know ri ght from wrong. This will be modeled after the quote made by the mother of murder victim Madd ie Clifton, the story de scribed at the beginning
68 of this thesis. Two parental manipulations were derived from the current studys content analyses results, exemplifying frequent trends seen in actual newspaper articles. One of these variations will provide information about lack of parental interest in the juveniles life, illustrating parents failure to attend parent-teach er conferences about juveniles behavior. The other manipulation includes parent al encouragement of delinquency, in the form of a statement the juvenile made to police. The final variat ion will combine all three elements of parental blame. Additional superfluous news stories will comprise the remaining stimuli in the experiment. The purpose of the stories will be to hide the purpose of the study in order to achieve the most accurate results. These stories will also be vari ed in television and newspaper format and will feature topics including life, trends, celebrity, sports and health. These additional stories are necessary because the part icipants will be advised that they are participating in a study about memory retention and the effects of television versus newspaper. The experimental study will utilize many elem ents of the current study to construct hypotheses and test for multiple factors that c ould influence public perception of parental responsibility; including th e research areas of fear of crime and pretrial publicity and attribution, cultivation, agenda setting and fram ing theories. First, utilizi ng cultivation th eory (Butts & Snyder, 1997; Gross & Aday, 2003; Romer et al., 2003) it is believed that participants exposed to a juvenile crime story will be more likely to express a higher level of fear of crime than those who were not exposed to a juvenile crime story. Past victimization, both direct and indirect, will be measured to test Gross & Adays (2003) findi ng that prior victimization is one of the most important predictors of fear. To test this idea, the experiment will test examine if participants reporting higher levels of prior vict imization are more likely to expe rience a greater level of fear
69 of crime than those participants without prior victimization. Due to the perceived increase in delinquency as a societal problem (Chiricos et al ., 1997), coupled with cultivation and attribution theory, it is hypothesized that f ear of crime will have a moderating influence on the relationship between media attribution of blame and participants blame assignment. Additionally, previous literatu re demonstrates television has a significant effect on viewers perception (Health & Gilbert, 1996; Lowry et al., 2003; Ro mer et al., 2003). Participants exposed to televisi on (video) juvenile crime stories are more likely to express a higher level of fear of crime than those who were exposed to the same level of story manipulation in print format. Multiple studies reveal television news has a greater effect on viewers than print medi a (Health & Gilbert, 1996; Lowry et al., 2003; Romer et al., 2003). Participants exposed to televisi on (video) juvenile crime stories may therefore be more likely to place blame on the juveniles parents than those e xposed to the same level of story manipulation in print format. Framing and cultivation theory describe the medias power to manipulate news information and the power this has on public per ception. Using the concepts in these theories, it is predicted that partic ipants not receiving a parental responsibility elem ent are less likely to blame a juveniles parent than t hose participants exposed to any of the parental manipulations. Based on framing and attribution theory, participants receiving the parental responsibility combination version of the juvenile crime story will be more likely to place blame on the juveniles parent than any other story format, controlling for medium type. Pretrial publicity research distinguishes between types of pretrial publicity, indicating emotional publicity has the greatest effect on the public because it is easiest to recall (Kramer et al., 1990). In addition to attribut ion theory, pretrial publicity effects shapes the hypothesis that
70 participants exposed to emotional mention of parental responsibility will be more likely to assign blame to juveniles parents than those participants e xposed to factual information or no mention of parental contribution. In orde r to test these hypotheses, simple effect tests of the parental blame assignment at the varying levels of media manipulation will be conducted. The parental responsibility dependent variab le question will be em bedded in a retention questionnaire asking participants Other than the offender, who else was responsible for the assault. Informed that different factors might infl uence retention rates, the participants next will be instructed to complete demographic (sex, age, race, etc.) beliefs (parents are responsible when their child breaks the law, childrens behavior problems usually have little to do with mistakes their parents make, a parent shoul d be punished when their child breaks the law, etc.), family composition (I contribute my success and achieveme nt to my, I blame my current struggles on my during high school my behavior was a reflect ion of my parents guidance, etc.), and fear of crime information (i.e. I have witnessed a non-vio lent crime, I have been a victim of a violent crime, at night I do not feel safe walking across campus alone, etc) questionnaires. The questions relevant to the expe rimental goals will be disperse d among superfluous questions to disguise the true nature of the study. The current study significantly contributes needed information to the field of parental responsibility. Findings demonstrat e the frequency and context of parental responsibility in the media. The forthcoming experiment will further these findings by experimentally testing the media effects on public opinion of parental responsibility.
71 CHAPTER 6 CONCLUSION The current study established the frequency and context of parental blame in the m edia. The findings reveal a clear presen ce of parental blaming in the media and further illustrating the type of parental responsibility media stories that are dissem inated to the public. The current findings also i ndicate a need to understand the medias effects on public opinion, parental responsibility legi slation, the juvenile ju stice system and the juveniles within it. Studies have demonstrated that the media depicts a false sense of reality to its viewers (Heath & Gilbert, 1996; Lowry, et al., 2003; Romer, et al., 2003), but it is just as critical to understand how these messages are being interpreted. Previous research on fear of crime explains how exposure to news coverage of crime can heighten ones fear of being vic timized or feeling unsafe. Thr ough framing, the media is able to manipulate the information it provides, thus in fluencing societal inte rpretations of these constructed messages. When the majority of news stories involve acts of violence and crime, agenda setting theory explains that crime is th e issue viewers will most likely interpret as important. Overall, researchers have demonstrat ed that frequent exposure to news can shape perceptions and public opinion. Pretrial publicity literature also confirms me dias effect on public opinion. It has been documented that initial judgments of a defenda nt can be determined solely on information provided through news reports (Kovera, 2002; Ott o, et al., 1994; Wilson & Bornstein, 1998). Based on this knowledge, exposure to case details via the media can highly impact court decisions. Again, elements of framing, attri bution and agenda setting theories provide confirmation that the media does influence public opinion in the legal system.
72 From our study we established that messages are being sent by the media on the topic of parental responsibility. Most na tional stories are including s upport for parental blame without episodic references to parental contribution. On the other hand, local st ories covering specific juvenile crimes often fail to incl ude any relevant refere nce to a parent and th eir responsibility in the crime. From this, the primary researcher re cognizes that it is important to question what effects media coverage of pare ntal responsibility ar e having on public opinion, if any. Results developed via the curren t study have many uses. The current study has analyzed more than 100 years of media (1899-2007) and pr ovided a look at the U.S. and Canadian media coverage on both a local and national level. Th is information will be influential to future researchers as it provides a longitudinal and in -depth look at the medias development and portrayal of parental responsibility. The findings illu strate that parental responsibility in juvenile crime is a current issue and one that requires attention from researchers and legal actors. The story of Maddie Clifton embodies the concept of parental responsibility in the media. On November 3, 1998 Maddie vanished and her body was later found under a neighborhood boys bed. The young boy, Joshua Phillips, claimed he killed the young girl after accidentally hitting her, in fear his father would punish him. Maddie Cliftons mother also blamed Joshuas parents when she said she was doing her job as a mother and wondered w hy Joshuas parents had not done the same. Maddie is gone, but the questi on of who is ultimately to blame, Joshua or his parents, will forever go unanswered for the Clif ton family. Should Joshuas parents also be punished for Maddies death, and to what extent has the media de veloped public support for this concept of parental responsibility?
73 LIST OF REFERENCES Alexander, P.W (1948). Whats this about punishing parents? Federal Probation 2329. Bahr, S. J., Marcos, A. C., & Maughan, S. L. (1995) Family, educational and peer influences on the alcohol use of female and male adolescents. Journal of Studies on Alcohol, 56, 457. Baldwin, P.C. (2002) Nocturnal habits and dark wisdom: The American response to children in the streets at night, 1880-1930. Journal of Social History, 35 (3), 593-611. Bassili, J.N., & Regan, D.T. (1977). Attributiona l focus as a determinant of information selection The Journal of Social Psychology 101, 113-121. Bianchi, S.M., & Robinson, J. (1997). What did you do today? Childrens use of time, family composition and the acq uisition of social capital. Journal of Marriage and Family, 59 332-344. Book, M.L. & Perala-Littunen, S. (2008). Childre n need their parents more than a pizza in the fridge! Parental responsib ility in a Finnish newspaper. Childhood 15(1), 74-88. Brank, E.M., Hays, S.A. & Weisz, V. (2006). A ll parents are to blame (except this one): Global versus specific attitudes rela ted to parental responsibility laws. Journal of Applied Social Psychology 36(11), 2670. Brank, E.M., Kucera, S.C. & Hays, S.A. (2005). Parental responsibility statutes: An organization and policy implications. Journal of Law & Family Studies 7 (1). Brank, E.M., & Weisz, V., (2004). Paying for th e crimes of their youth: Public support of parental res ponsibility laws. Journal of Criminal Justice, 32, 465-475. Bruschke, J., & Loges, W.E. (1999). Relations hip between pretrial publicity and trial outcome. Journal of Communication, 49(4), 104-120. Budiansky, S., Gregory, S., Schmidt, K.F., & Bierck, R. (1996). Local TV: Mayhem central. U.S. World and News ReportCulture & Ideas, 120 (9), 63. Burton, V.S., Cullen, F.T., & Evans, T.D. (1995) The Impact of Parental Controls on Delinquency. Journal of Criminal Justice, 23(2) 117-118. Buss, A.R. (1978). Causes and reason in at tribution theory: A conceptual critique. Journal of Personality and Social Psychology 36 (11), 1311-1321. Butts, J.A., & Snyder, H.N. (1997). The youngest delinquents: Offenders under age 15 Office of Juvenile Justice and Delinquen cy Prevention: Juven ile Justice Bulletin
74 Candiotti, S. (Contributing Correspondent). (199 9, July 8). Florida teen guilty of killing 8-year-old neighbor: Boy could get life w ithout parole. CNN. Retrieved October 23, 2007, from http://www.cnn.com/US/9907/08/fla.child.murder.03/index.html Carroll, J.S., Kerr, N.L., Alfini, J.J., Weav er, M.F., MacCoun, R.J., Feldm an, V. (1986). Free press and fair trial: The role of behavioral research. Law and Human Behavior, 10(3),187-201. Chiricos, T., Eschholz, S., & Gertz, M. (1997). Crime, news and fear of crime: Toward an identification of audience effects Social Problems, 44(3), 342-357. Clemente, F., & Kleiman, M. (1977). Fear of crime in the U.S.: A multivariate analysis. Social Forces, 56(2), 520-531. Coleman, R., & Thorson, E. (2002). The effect s of news stories that put crime and violence into context: Testing th e public health model of reporting. Journal of Health Communication, 7, 401-425. Constantini, E., & King, J. (1980). The pa rtial juror: Correla tes and causes of prejudgment. Law and Society Review, 15 (1), 9-40. Davidson, H. (1996). No consequences Re-e xamining parenting re sponsibility laws. Stanford Law & Policy Review 7(1), 23-37. Dimitris, J.E. (1997). Parental responsibility statutes a nd the programs that must accompany them. Stetson Law Review 27, 655-98. Ditton, J., Chadee, D., Farrall, S., Gilchrist, E., and Bannister, J. (2004) From imitation to intimidation: A note on the curious and changi ng relationship between the media, crime and fear of crime. The British Journal of Criminology, 44, 595-610. Doob, A.N., & Macdonald,G.E. (1979). Television viewing and fear of victimization: Is the relationship casual? Journal of Personality and Social Psychology 37 (2), 170-179. Dorfman, L., Woodruff, K. Chavez, V., & Wa llack, L. (1997). Youth and violence on local television news in California. American Journal on Public Health 87(8), 13111316. Entman, R.M. (1993). Framing: Towards clarification of a fractured paradigm Journal of Communication, 43(4), 51-58. Evans, T.D., Evans, M.A., & Meredith, C. W. (1993). The fallacy of parental responsibility. Individual Psychology, 49 (2), 231-234. Garofalo, J., (1981). The fear of crime: Causes and consequences. The Journal of Criminal Law and Criminology 72(2), 839-857.
75 Ghetti, S., & Redlich, A.D. (2001). Reacti ons to youth crime: Perceptions of accountability and competency Behavioral Sciences and the Law 19, 33-52. Greene, E. (1990). Media effects on jurors. Law and Human Behavior 14 (5), 439-450. Gross, K., & Aday, S. (2003). The scary wo rld in your living room and neighborhood: Using local broadcast news, neighborhood crime rates, and personal experience to test agenda setting and cultivation Journal of Communication, September, 411-426. Hartnagel, T.F. (2004). The rhetoric of youth justice in Canada. Criminal Justice, 4 (4), 355-374. Hastie, R. (1984). Causes and effects of casual attribution Journal of Personality and Social Psychology 46(1), 44-56. Heath, L. (1984). Impact of newspaper crime reports on fear of crime: Multimethodological Investigation Journal of Personality and Social Psychology 47 (2), 263-276. Heath, L., & Gilbert, K. (1996). Mass media and fear of crime. American Behavioral Scientist 39 (4), 378-386. Heath, L., & Petraitis, J. (1987) Television viewing and fear of crime: Where is the mean world? Basic and Applied Social Psychology 8(1 & 2), 97-123. Honess, T.M., Charman, E.A., & Levi, M. (200 3). Factual and affec tive/evaluative recall of pretrial publicity: Their relative influence on juror reasoning and ve rdict in a simulated fraud trial. Journal of Applied Social Psychology, 33 (7), 1404-1416. Humphries, D. (1981). Serious crime, news cove rage, and ideology: A content analysis of crime coverage in a metropolitan paper. Crime & Delinquency, 27 (2) 191-205. International Cooperation Group. (2004) The evolution of juvenile justice in Canada Department of Justice of Canada. Retrieved April 6, 2008 from http://www.justice.gc.ca/eng/pi/icg-gci/jj2-jm2/jj2-jm2.pdf Iyengar, S. (1990). Framing responsibility for political issues: The case of poverty. Political Be havior, 12(1), 19-40. Iyengar, S., Peters, M.D., & Kinder, D.R. ( 1982). Experimental demonstrations of the not-so-minimal consequences of television news programs. The American Political Science Review, 76(4), 848-858. Jacksonville City Ordinance. (2006). An ordinance regarding curfews for minors or children. Ordinance 2006-889.
76 Kelley, H.H., & Michela, J.L. (1980) Attribution theory and research Annual Review of Psychology, 31, 457-501. Kenny, J.A., & Kenny, J.V. (1961). Shall we punish the parents ? American Bar Association Journal, 47, 804-808. Krippendorf, K. (1980). Content analysis: An introduction to its methodology. Beverly Hills, CA: Sage Publications. Kohn, D. (Executive Produ cer). (1999, October 7). The search for Maddie: Children were playing in a quiet neighborhood. 48 Hours. (Television broadcast). New York: CBS. Kovera, M.B. (2002). The effects of general pretrial publicity on juror decisions: An examination of moderators and mediating mechanisms Law and Human Behavior, 26(1), 43-72. Kramer, G.P., Kerr, N.L., & Carroll, J.S. (1990). Pretrial publicity, judi cial remedies, and jury bias. Law and Human Behavior 14(5), 409-438. Kuendig, H., & Kuntsche, E. (2006). Family bonding and adolescent alcohol use: Moderating effect of living with excessive drinking parents. Alcohol & Alcoholism, 41(4), 464. Lane, J., & Meeker, J.W. (2003). Ethnicity, in formation sources, and fear of crime. Deviant Behavior: An Interdisciplinary Journal 24(1), 1-26. Lowry, D.T., Nio, T.C.J., & Le tner, D.W. (2003). Setting th e public fear agenda: A longitudinal analysis of network TV crime reporting, public perceptions of crime, and FBI crime statistics. Journal of Communications 53(1), 61-73. McCombs, M.E., & Shaw, D. (1972). The agenda-setting function of the mass media. Public Opinion Quarterly, 36, 176-187. McManus, J., & Dorfman, L. (2002). Youth vi olence stories focus on events, not causes Newspaper Research Journal, 23(4), 6-20. Osgood, C.E., & Tannenbaum, P.H. (1955). The principle of congruity in the prediction of attitude change. Psychological Review 62(1), 42-55. Otto, A.L., Penrod, S.D., & Dexter, H.R. (1994). The biasing impact of pretrial publicity on juror judgment. Law and Human Behavior 18 (4), 453-469. Rodgers, S., & Thorson, E. (2001). The reporti ng of crime and violence in the Los Angeles Times: Is there a public health perspective? Journal of Health Communication 6, 169-182.
77 Romer, D., Jamieson, K.H., & Aday, S. (2003). Television news and the cultivation of fear of crime. Journal of Communication, 53(1), 88-104. Schissel, B. (1997). Blaming Children. Halifax: Fernwood Publishing. Shepherd, R. E. (1999). Section III: Developments in the United States: Film at eleven: The news media and juvenile crime. Bridgeport Law Review/Q uinnipiac Law Review 18. Shorstein details Phillips confession. (1999, July 7). Florida Times-Union. Retrieved October 23, 2007, from http://www.jacksonville.com/special /phillipstrial/stories/1011.shtm l Smith, M.J. (1988). Contemporary communication research methods. Belmont, CA: Wadsworth Publishing Company. Sprott, J. (1996). Understanding public view s of youth crime and the Youth Justice System. Canadian Journal of Criminology, 38 (3), 271-290. Stafford, M.C. & Galle, O.R. (1984). Victimizati on rates, exposure to risk, and fear of crime. Criminology, 22(2), 173. Steblay, N.M., Besirevic, J., Fulero, S.M., & Ji menez-Lorenta, B. (1999). The effects of pretrial publicity on juror verd icts: A meta-analytic review. Law and Human Behavior 23(3), 219-233. Sweeney, K. (1999, November 3). A year of h ealing: Clifton family adjusts to life without Maddie. Florida Times-Union Retrieved October 23, 2007 from, http://www.jacksonville.com/tuonline/stories/110399/m et_1128983.html Thurman, T. (2003). Parental re sponsibility laws: Are they the answer to juvenile delinquency ? Journal of Law and Family Studies 5. Tomaszewski, A.L. (2005). From Columbine to Kazaa: Parental liability is a new world. University of Illinois Law Review 2, 573-600. Tyler, J.E., & Segady, T.W. (2000). Parental Liability laws: Rationale, theory, and effectiveness. The Social Science Journal 37(1), 79-96. Walster, E. (1966). Assignment of responsibility for an accident. Journal of Personality and Social Psychology, 3 (1), 73-79. Weber, R.P. (1990). Basic content analysis. Second Edition. Newberry Park, CA: Sage Publications.
78 Wilson, J.R., & Bornstein, B.H. (1998). Methodo logical considerati ons in pretrial publicity research: Is the medium the message? Law and Human Behavior, 22(5), 585-597. World: Americas 15-year-old guilty of murder. (1999, July 9). British Broadcasting Channel. Retrieved October 23, 2007, from http://news.bbc.co.uk/1/h i/world/americas/389777.stm Yanich, D. (2005). Kids, crime, and local television news Crime & Delinquency 51(1), 103-132. Zhang, L., Welte, J. W ., & Wieczorek, W. F. ( 1999) The influence of parental drinking and closeness on adolescent drinking. Journal of Studies on Alcohol, 60 245.
79 BIOGRAPHICAL SKETCH Ashley R. P. Kolnes was born and raised in Pensacola, Florida. W ith the support of her parents, Bruce and Cindy Peake, and her older brother, Gerrad, Ashley became an ambitious student. In 2004, Ashley graduate d with a BS in Communications Public Relations from the University of North Florida. With the comple tion of a masters degree in 2008, she is continuing her graduate career at the Univer sity of Florida, pursuing her doctorate in Criminology, Law and Society. In addition to these degrees, Ashley is currently earning a Certificate of Death Investigation through UFs Departme nt of Forensic Science. Ashley married Thor Kolnes on July 14, 2007. Th ey currently live in Gainesville, Florida with their cat and two Welsh Pembroke Corgis. For enjoyment, Ashley is an active member of the Delta Gamma Alumni Association, avidly attends UF sporting events, and engages in exercise and outdoor activities. Her faith and fam ily are guiding factors in her life and have been influential in her successes. The focus of Ashleys doctoral research will in clude cold case homicide investigations and protocol, with special attention given to the effects these unsolved crimes have on a victims survivor family members. Other areas of special interest include juvenile justice, law and society and crime in the media