January 2013 Trends in Bullying and Peer Victimization David Finkelhor NATIONAL CRIME VICTIMIZATION SURVEY (DEPARTMENT OF JUSTICE) The annual National Crime Victimization Survey (NCVS) shows that between 1992 and 2010 for youth 12-18 school related violent victimizations declined 74% (Figure 1) and school re- lated thefts declined 82% (Figure 2).1 The declines were fairly linear over this period and included a drop of 50% in school related violence and 45% in theft in the most recent period from 2007 to 2010. These declines in the overall NCVS were roughly confirmed by the specific school crime supplement of the NCVS conducted among 12-18 year old youth at less frequent intervals. The school crime supplement showed a decline from 1995 to 2009 of 53% in violent school victimization and 61% in school theft.1 The large decreases were across the board, affecting youth of both genders, all races and in urban, suburban and rural set- tings. They included small declines from 2007 to 2009 (12% for violent victimization). The school crime supplement also asked about being the tar- get of hate-related words at school. The trend showed a de- cline of 29% from 2001 to 2009. There were declines for whites and blacks but not Hispanics.1 Finally, the NCVS school supplement began to ask a specific question about bullying in 2005 that was repeated in 2007 and 2009. The rate rose from 2005 to 2007 and then declined from 2007 to 2009 (from 28% to 32% and back to 28%). Figure 1. Trend in Violent Victimization at School, Past Year, Youth 12-17 70 Rate per 1000 60 50 40 1992 - 2010 74% decline 30 20 10 0 *Source: National Crime Victimization Survey, U.S. Department of Justice, Bureau of Justice Statistics, Indicators of School Crime Safety, 2011 Figure 2. Trend in Theft Victimization at School, Past Year, Youth 12-17 120 100 Rate per 1000 Highly publicized cases of bullying in recent years and new attention to the problem in schools have created the impres- sion among some observers that the problem has been on the rise. But at the same time, crime and violence, including youth violence, have been decreasing overall in the US. Is this true for bullying and peer victimization? In this bulletin, we will summarize the trends, from youth sur- veys that have tracked bullying specifically, and also those that have tracked closely related phenomena such as school assaults, school thefts, school fighting and school hate speech. The surveys that reflect change over the longest time periods, going back to the early 1990s, consistently show declines in bul- lying and peer victimization, some of it remarkably large. The more recent trends, since 2007, show some declines, but less consistently. 80 1992 - 2010 82% decline 60 40 20 0 *Source: National Crime Victimization Survey, U.S. Department of Justice, Bureau of Justice Statistics, Indicators of School Crime Safety, 2011 Page 2 Trends in Bullying and Peer Victimization YOUTH RISK BEHAVIOR SURVEY (CENTER FOR DISEASE CONTROL) The Youth Risk Behavior Survey (YRBS) also has conducted large national student surveys (9th-12th graders) going back to the early 1990s. Between 1991 and 2011, the survey docu- mented declines in youth physical fighting (down 23%) and in fighting on school property (down 26%) (Figure 3).2 In the more recent time period 2007 to 2011, changes were slight in both: physical fighting (down 7%) and fighting on school prop- erty (down 3%). The National YRBS began reporting specifically about bullying on school property in 2009 and found no difference between 2009 and 2011. However, in the YRBS survey conducted in Massachusetts a bullying question has been asked since 2003 and the rate there has declined 22% from then until 2011.3 Figure 3. Trend in Fighting on School Property, Past Year, Students Grades 9-12 17 Percent 16 15 1993 - 2009 26% decline 14 13 12 11 10 *Source: Youth Risk Behavior Survey, U.S. Department of Justice, Bureau of Justice Statistics, Indicators of School Crime Safety, 2011 HEALTH BEHAVIOR IN SCHOOL AGE CHILDREN STUDY As part of an international project on the health of children ages 11, 13 and 15, representative national samples were sur- veyed on multiple occasions in 27 countries including the US. A question using the specific term "bullying" along with a defi- nition was asked to inquire about victimization and perpetra- tion. The longest span in the US compared 1997-8 with 2005-6 and broke down trends by gender. For boys perpetration dropped 15.5% (occasional bullying) and 37.4% (chronic bully- ing) and victimization dropped 24.7% (occasional) and 27.0% (chronic). For girls the declines were less dramatic, with perpe- tration down 4.1% (occasional) and 16.7% (chronic), and vic- timization down 0.7% (occasional) and 3.6% (chronic).4 OTHER SURVEYS Another state, Minnesota, has a long-standing student survey that asked 6th, 9th and 12th graders every 3 years about "pushing, grabbing and shoving" on school property. The 6th graders, who had the highest rates overall, had a 19% decline from 1995 to 2010 in what was a linear trend over time. The 9th graders had an 18% decline mostly concentrated between 2001 and 2010. The 12th graders had very little overall decline and most of it wasn't until the period from 2007 to 2010. The National Survey of Children Exposed to Violence (NatSCEV) gathered information about victimizations of rep- resentative samples of children in 2006, 2009 and 2011. Be- tween 2006 and 2009 it found a decline in physical harass- ment by peers but no decline in psychological harassment.5 From 2009 t0 2011 there was no change. CYBERBULLYING Because of concern that aggressive behavior and bullying have transferred to electronic media, surveys have begun to assess this problem as well. The Youth Internet Safety Survey (YISS) asked a representative sample of US Internet-using youth about experiences of being harassed online. Rates in- creased from 6% in 2000 to 9% in 2005 to 11% in 2010, an increase of 83% over the decade.6 INTERNATIONAL TRENDS Rigby & Smith7 reviewed trend information for bullying on the international level and concluded that most of the data pointed toward declines. "All nine of the data sets noted ... showed some decrease in reported victimization, although for Norway there was evidence of a recent resurgence. In the international data set for boys and girls combined, 19 of the 27 reported cases showed a significant decrease in occasional victimization and 21 in chronic victimization. From this it ap- pears that the prevalence of bullying among young people is generally decreasing." SUMMARY Four US national data sets show substantial declines in face- to-face bullying and peer related victimizations at school from the 1990s to recent years. Some of these are quite large. In general, the declines are broad across demographic groups. One study reports less decline in bullying for girls, but other indicators do not replicate this gender difference. Being the target of hate words may not have declined for Hispanics. The trends for more recent years, beginning in 2006 or 2007, are fairly consistent but less dramatic. The NCVS violence measures since 2007 showed a large continuing decline. The NCVS school supplement and YRBS showed declines, includ- ing a small decline in a specific bullying question from 2007 to 2009 (although it rose from 2005 to 2007). NatSCEV found a decline for physical, but not psychological bullying. On the other hand, Internet harassment appears to have in- creased in the 2000s. Could peer victimization have simply been displaced from the school to the electronic environ- ment? Two factors argue against a simple displacement model. First, the declines in peer victimization date from the early 1990s, some years before the social media and mobile device revolution gained its dominion among youth in the early 2000s. Second, the main declines documented in the surveys are for physical assaults and property crimes that do Page 3 Trends in Bullying and Peer Victimization not transfer to the electronic environment. Rather than a dis- placement, the increase in online harassment is probably best seen simply as growth in the usage of electronic media for all kinds of socialization including its negative forms.8 The strong overall conclusion from the available data is that there have been fairly substantial declines in face-to-face bullying and peer victimization in school since the 1990s with indications that the declines have continued in recent years. EXPLANATIONS What are the likely explanations of these trends? Although the intensity of attention to bullying has risen in very recent years, the issue of violence and crime in school has been a problem of longstanding interest to authorities in education and law en- forcement. One of the main innovations in school safety pro- grams was the School Resource Officer (SRO) program, which delegates police to work in schools and try to develop and pro- mote violence and crime reduction programs. The SRO grew out of Community Oriented Police (COPS) funding in the mid to late 1990s. One estimate was that, "in 1999, 30 percent of local police departments, employing 62 percent of all officers, had about 9,100 full-time school resource officers assigned to schools."9 At the same time, curriculum developers created a wide array of bullying and violence prevention programs for school age children, some of which have proven effective in scientific evaluation.10 Surveys suggest that considerable portions of school age populations in the US have been exposed to such programs. Given this mobilization since the 1990s, it may be that efforts of school personnel and school curricula on pre- venting violence and crime have had some role in reducing its prevalence. At the same time that bullying apparently has been declining, other indicators of crime and youth deviance have also been improving. Homicide and suicide rates among a broad age spectrum of the population have been decreasing. Most kinds of crime involving offenders at most ages have also gone down. This may reflect active crime prevention efforts at many levels, but it is also possibly related to some general social changes that might affect crime, violence and deviance more generally. There are several possible candidates. One has been the advent of electronic media and mobile communication. These tech- nologies may have dampened crime and bullying by providing more ways of summoning help, more forms of social surveil- lance, and engrossing activities that undermine forms of alien- ation that lead to crime. Another possible candidate is psychi- atric medication, which has promoted an increased access to mental health services, particularly among males and less edu- cated segments of the population including youth who were reluctant to engage in counseling therapy. These medications may have eased some of the forms of depression, anxiety and hyperactivity that fueled bullying, crime and other deviance. IMPLICATIONS One implication concerns how bullying and peer victimiza- tion are discussed in both the media and the public. Given that surveys continue to show high rates of peer victimiza- tion and bullying, and that youth are the most criminally victimized segment of the population, strong concern about this exposure is clearly warranted. But these concerns should not be framed as a recent increase in the problem. Moreover, reporting about the high rates should be tem- pered with information about the declines. What should media and researchers say about the size of the decline? Though the various studies point to declines, the magnitudes do vary among the studies. For example the NCVS showed drops of over 50% in school related violence since the 1990s, but other studies like the YRBS have de- clines more like 25%. The NCVS is clearly the oldest and most methodologically rigorous study, but it is not con- ducted in schools, and its definitions of violence are narrow. If room is not available to cite specific studies, we suggest a formulation such as the following: "Studies tracking school violence and bullying since the early 1990s through 2010 show declines of between 25% and 75% for problems such as violent peer victimization, fighting in school, bullying, and school theft." Another implication concerns how to look at prevention efforts. The trends suggest that something has been help- ing to reduce the toll of violence and bullying in schools. While we cannot conclude that everything that schools have been doing is contributing positively to this decline, the im- provement at least counsels caution about abandoning pro- grams and approaches that have been utilized in the recent past. The decline also has implications for research and evalua- tion. In doing program evaluation, it is not sufficient simply to track bullying rates before and after the implementation of a program. If bullying is declining most places because of some larger social factors, the decline showing up in the evaluation may be due to those factors, not the program being implemented. The proper design is to make sure that the evaluation tracks bullying in some comparison group that is not receiving the program, so the effects for the pro- gram group can be distinguished from general overall trends. Page 4 Trends in Bullying and Peer Victimization CONCLUSIONS Bullying and peer victimization appear to be declining since the 1990s. This is good news. But it should not be interpreted as the problem having been solved. First, the rates are still incredibly high. For example, more than one in 10 high school students said they were in a physical fight on school property in the last year. We would not tolerate a level of work place danger that was so high, nor should we. Second, there is no reason to assume that the trend will continue downward. It may decline further as we continue to promote violence pre- vention. But it may also be that the early declines were the easiest, preventing the most readily preventable cases. New strategies and more intensive work may be needed to prevent the parts of the problem that remain. But advocates and young people should feel inspired. Change can happen and it can get better. REFERENCES 1. Robers, S., J. Zhang, J.L. Truman, and T.D. Snyder, Indica- tors of school crime safety: 2011. 2012, Bureau of Justice Statistics: Washington, DC. p. 1-203. 2. Centers for Disease Control & Prevention. Trends in the prevalence of behaviors that contribute to violence - National YRBS: 1991 - 2011. [Survey]. 2012. Available from: http://www.cdc.gov/healthyyouth/yrbs/pdf/ us_violence_trend_yrbs.pdf. 3. Milligan, C. Massachusetts Youth Risk Behavior Survey. 2012. Available from: http://www.doe.mass.edu/cnp/hprograms/yrbs/. 4. Molcho, M., W. Craig, P. Due, W. Pickett, Y. Harel-Fisch, M.D. Overpeck, and HBSC Bullying Writing Group, Cross- national time trends in bullying behaviour 1994-2006: Find- ings from Europe and North America. International Journal of Public Health, 2009. 54: p. S225-S234. 5. Finkelhor, D., H.A. Turner, R. Ormrod, and S.L. Hamby, Trends in childhood violence and abuse exposure: Evidence from two national surveys. Archives of Pediatrics & Adoles- cent Medicine, 2010. 164(3): p. 238-242. 6. Jones, L.M., K.J. Mitchell, and D. Finkelhor, Trends in youth Internet victimization: Findings from three youth internet safety surveys 2000-2010. Journal of Adolescent Health, 2012. 50: p. 179-186. 7. Rigby, K. and P.K. Smith, Is school bullying really on the rise? Social Psychology of Education, 2011. 14(4): p. 441- 456. 8. Jones, L.M., Mitchell, K.J., Wolak, J., & Finkelhor, D. (in press). Online harassment in context: Trends from three Youth Internet Safety Surveys (2000,2005, 2010), Psychol- ogy of Violence. 9. Finn, J.D. and J. Owings, The adult lives of at-risk students: The roles of attainment and engagement in high school. 2006: U.S. Department of Education, Washington, DC. 10. Ttofi, M.M. and D.P. Farrington, What works in preventing bullying: Effective elements of anti-bullying programs. Jour- nal of Aggression, Conflict and Peace Research, 2009. 1(1): p. 13-24. ACKNOWLEDGEMENTS Thanks to Kelly Foster, Toby Ball and Lisa Jones for help in the preparation of this bulletin. This work is supported in part with funds from the US Department of Justice, Office of Juvenile Justice and Delinquency Preventi0n.