Actions Speak Louder Than Words: Examining the Relationship Between Violent Behaviors and Bullying Victimization Among Adolescents

Abstract

Incidences of traditional bullying and cyberbullying have received national attention and have increased concerns of law makers and education officials regarding prevalence and the potential impacts on adolescent victims, specifically regarding the relationship to violent behaviors. However, research that specifically examines the relationship between the co-occurrence of traditional- and cyber-victimization and the potential association with violent behaviors are scarce. This research examines the potential influence of traditional and cyberbullying victimization on counts of violent behaviors in a nationally representative adolescent sample. Data are from the Youth Risk Behavior Surveillance System, for which a national sample of 15,624 adolescents aged 12–18 years, in grades 9 through 12 were surveyed during the 2015 school year. Negative binomial regression models were created to examine the relationship between victimization and violent behaviors among adolescents. Results indicate that victimization, particularly the combination of traditional and cyberbullying, significantly impacts counts of violent behaviors among adolescents. The findings show that further action should be taken to understand the negative outcomes stemming from the co-occurrence of traditional and cyber-victimization and to strengthen programs created to reduce rates of traditional and cyberbullying among adolescents in schools and homes.

Introduction

Bullying has been defined by the U.S. Centers for Disease Control and Prevention (CDC) as “any unwanted aggressive behavior(s) by another youth or group of youths who are not siblings or current dating partners that involves an observed or perceived power imbalance and is repeated multiple times or is highly likely to be repeated” (Gladden et al. 2014). While this traditional form of bullying typically occurs within middle- and high-schools, the ongoing advances in technology have contributed to the creation of a new type of electronic bullying, referred to as “cyberbullying” (Patchin and Hinduja 2010; Willard 2004). Hinduja and Patchin (2009, p. 5) have defined cyberbullying as the “willful and repeated harm inflicted through the use of computers, cell phones, and other electronic devices.” National estimates report that ∼20% of adolescents have reported experiencing traditional bullying victimization, while a slightly smaller proportion, 15%, report having experienced cyberbullying victimization (National Center for Education Statistics 2016). Research has established that cyberbullying is a unique form of bullying as it can occur anywhere and anytime, and can also be used in conjunction with traditional bullying (Mehari and Farrell 2018). Currently, there are no national estimates of the amount of adolescents who report experiencing both types of bullying victimization combined. This study seeks to bring attention to the importance of this population.

Existing literature on traditional bullying victimization and cyberbullying victimization has established that each of these types of behaviors can have negative outcomes for adolescent individuals. For instance, traditional bullying victimization has been linked to adolescents experiencing higher levels of anxiety, depression, and lower levels of self-esteem (Salmivalli et al. 1999), while cyberbullying victims have been found to engage in risky sexual behavior, vandalizing of property, misuse prescription drugs and over-the-counter drugs, increased risk for suicide ideation, and suicide attempts (Elgar et al. 2014; Litwiller and Brausch 2013). Moreover, a vast amount of research has addressed the association between traditional bullying victimization and violent behaviors, while others have examined the relationship between cyber-victimization and violence. For instance, Nansel and colleagues (2003) found that being traditionally bullied was consistently related to violence-related behaviors, such as weapon carrying in and out of schools and physical fighting, for both male and female adolescents. In addition, research has found that bullying victimization is significantly associated with high levels of aggression (Duggins et al. 2016) as well as extending violent behaviors into adulthood (Kim et al. 2011).

While research has established that each type of victimization on its own has a variety of negative outcomes, particularly related to violent behaviors, there are few studies that examine the co-occurrence of these types of victimization and the outcomes that may follow. Reed and colleagues (2015) found that both traditional and cyberbullying victimization can lead to an increase in antisocial behavior as well as being put at risk for possibly developing violent behavior. However, they do not go into detail regarding specific types of violent crime and their relationship to traditional and cyberbullying. Mehari and Farrell (2018) examined aggression behaviors among individuals who use traditional and cyber bullying, however, their sample size was not generalizable to the larger population as a whole and did not examine specific forms of violence. This study seeks to further this discussion and fill the gaps regarding the co-occurrence of traditional and cyberbullying victimization and the relationships that might occur regarding violent behavior, such as carrying a weapon or engaging in physical fights. Understanding these relationships could provide a deeper insight into the ways in which bullying impacts its victims, and could provide officials and policy makers with more information to guide programming and legislature regarding school safety.

Materials and Methods

Data for this study came from the 2015 National High School Youth Risk Behavior Surveillance System (YRBSS), which assesses and monitors health risk behaviors that greatly contribute to the leading causes of social problems, disability, and death among adolescents in the United States (USDHHS 2013). This survey fits well with this research as it has a specific section devoted to adolescents' experiences with traditional bullying victimization and cyberbullying victimization. In addition, it has a specific set of survey questions that deal with adolescents engaging in violent behaviors, such as carrying a weapon, carrying a gun, and being involved in a physical fight.

The YRBSS obtains a nationally representative sample of students in grades 9–12 across the United States by using a “three-stage, cluster sample design” (USDHHS 2013). All students in grades 9–12 in both public and private schools are included in the target population. Data are collected from all 50 states and the District of Columbia; however, the U.S. territories are excluded from the sample (USDHHS 2013). Following the three-stage, cluster sample design, schools are first sorted into primary sampling units (PSUs) in accordance with the size of the county in which they are located. Next, schools are selected from PSUs and divided into “whole” schools, meaning the school has all four high school grades 9–12, or “fragment” schools, meaning they have any other set of grades. Schools with more than 25 students in each grade are considered to be “large” schools, and schools with lower enrollment numbers are considered to be “small.” Approximately 25% of PSUs are selected for “small school sampling,” and out of each of these PSUs, one “small school” and three “large schools” are selected with probability relative to the school's enrollment size (USDHHS 2013). Finally, one or two entire classes in each school are chosen randomly from grades 9 to 12. All students in the sampled classes are chosen as participants.

For the national survey, data collectors are sent to each participating school to administer the questionnaire to students (USDHHS 2013). In addition, data collectors record information, such as the grade level and the number of students enrolled in the sample classes. By recording this information, researchers are later able to verify sample selection and weight data. The school response rate for the 2015 survey was 69%, and the student response rate was 86% (USDHHS 2013). No subsample was created from the sample obtained by the YRBSS. All respondents were included in the analysis (N = 15,624).

Violent Behaviors Index

The measure Violent Behaviors Index was created by combining six survey question regarding violent behaviors. The six questions that were included asked (1) “During the past 30 days, on how many days did you carry a weapon such as a gun, knife, or club?” (2) “During the past 30 days, on how many days did you carry a gun?” (3) “During the past 30 days, on how many days did you carry a weapon such as a gun, knife, or club on school property?” (4) “During the past 12 months, how many times were you in a physical fight?” (5) “During the past 12 months, how many times were you in a physical fight in which you were injured and had to be treated by a doctor or nurse?” (6) “During the past 12 months, how many times were you in a physical fight on school property?” (CDC 2013). Each of these questions was dummy coded 0 = no and 1 = yes. The Violent Behaviors Index was then created by combining all six questions together to create an index variable with a range from 0 to 6, with 0 = engaged in no violent behaviors and 6 = engaged in all violent behaviors listed. The Cronbach's alpha for this scale was α = 0.72.

Bullying Victimization

The measure Bullying Victimization was created by combining two survey questions regarding experiencing bullying victimization. To measure cyberbullying victimization, the survey asked respondents, “During the past 12 months, have you ever been electronically bullied? (Count being bullied through email, chat rooms, instant messaging, websites, or texting.)” (CDC 2013). The survey question regarding traditional bullying victimization asked “During the past 12 months, have you ever been bullied on school property?” (CDC 2013). These questions were then combined to create a categorical variable Bullying Victimization where 1 = experienced no victimization, 2 = experienced cyberbullying victimization only, 3 = experienced traditional bullying victimization only, and 4 = experienced both cyberbullying and traditional bullying victimization.

Sex was measured by asking participants, “What is your sex?” with the choices of female and male (CDC 2013). This measure was dummy coded so that 0 = female and 1 = male.

Race was measured by asking, “What is your race?” Racial categories included the following: American Indian or Alaska Native, Asian, black or African American, Native Hawaiian or other, and white. Race was recoded as (1) white, (2) black, and (3) other/multiracial. The demographic control variables include age and grade in school. Age had a range of 1–7 and was measured by asking respondents, “How old are you?” with the choices of (1) 12 years or younger, (2) 13 years old, (3) 14 years old, (4) 15 years old, (5) 16 years old, (6) 17 years old, and (7) 18 years old or older. Grade in school had a range of 1–4 and was measured by asking, “In what grade are you?” with the choices of (1) 9th, (2) 10th, (3) 11th, and (4) 12th grade.

Analytic strategy

Using STATA SE 15 (StataCorp 2015), frequency and descriptive tests were run on all dependent, independent, and control variables to provide characteristics of the sample as well as to check for any errors in the data. The YRBSS uses a complex sample design, and data are weighted to adjust for school and student nonresponse, which makes the data more representative of the population of students from which the sample was drawn. Data were appropriately weighted in STATA SE 15, following the YRBSS technical manual. Multiple imputation was used to address issues of missing data. Due to the presence of overdispersion in the dependent variable, negative binomial models were run to examine the association of bullying victimization on counts of violent behaviors. In the models, bullying victimization was entered as a categorical predictor variable with no victimization as the reference category. Similarly, in the models, sex and race were entered as a categorical predictor variable with female and white as the reference categories.

Results

Table 1 provides the descriptives for all measures included in the analysis. The average age of the sample was 16 years old; 49.6% of the sample was male and 49.6% of the sample was female. The sample comprised 72.8% whites, 17.7% blacks, and 9.5% other/multiracial individuals. Approximately 75% of the adolescent sample reported that they had not experienced any type of bullying victimization within the 12 months before the survey being conducted. Adolescents who reported experiencing only traditional bullying victimization comprised 9.6% of the sample, and those who reported experiencing only cyberbullying victimization made up 5.2% of the sample. Furthermore, 9.2% of the adolescent sample reported experiencing both traditional and cyberbullying victimization.

Table 1.

Sample Characteristics (N = 15,624)

	N	Total %
Gender
Male	7749	49.6
Female	7757	49.6
Race
White	6849	72.8
Black	1667	17.7
Other	890	9.5
Bullying Victimization Index
Did not experience victimization	11,649	74.6
Cyberbullying victimization only	808	5.2
Traditional victimization only	1506	9.6
Both types of victimization	1439	9.2
Violent behaviors
Carried a gun in past 30 days
Yes	713	5.4
No	12,550	94.6
Carried a weapon in past 30 days
Yes	2526	16.2
No	11,897	76.1
Carried a weapon on school property in past 30 days
Yes	704	4.6
No	14,764	95.4
Engaged in physical fight in past 12 months
Yes	2881	18.4
No	10,243	65.6
Engaged in physical fight in past 12 months that required medical attention
Yes	397	2.5
No	12,943	82.8
Engaged in physical fight in past 12 months on school property
Yes	1253	8
No	14,079	90.1
Violent Behavior Index
Engaged in zero behaviors	8917	57.1
Engaged in one behavior	1743	11.2
Engaged in two behaviors	1249	8
Engaged in three behaviors	486	3.1
Engaged in four behaviors	210	1.3
Engaged in five behaviors	71	0.5
Engaged in six behaviors	46	0.3

	Mean	Range
Age	16	12–18
Grade	10–11	9–12

2015 National Youth Risk Behavior Surveillance System.

Of the adolescent sample, 5.4% reported having carried a gun within 30 days before the survey being conducted, and 16.2% reported having carried a weapon, such as a knife or club. About 5% reported having carried a weapon on school property within the last 30 days. Approximately 18.4% of adolescents reported that they had engaged in a physical fight in the last 12 months, and about 3% reported that they had engaged in a physical fight that required medical attention. Furthermore, 8.0% reported engaging in a physical fight on school property within the 12 months before the survey being conducted. While 57.1% of the sample reported that they had not engaged in any violent behaviors, 11.2% reported engaging in one violent behavior, 8.0% reported engaging in two violent behaviors, and 5.2% reported engaging in three or more violent behaviors.

Table 2 displays the results of the negative binomial regression model examining bullying victimization and the impact on counts of violent behaviors. Adolescents who reported that they had been traditionally bullied engaged in 55.6% more counts of violent behaviors than did their nonbullied counterparts. In addition, those who reported experiencing only cyberbullying victimization engaged in 58.6% more incidents of violent behaviors than those who were not bullied. Furthermore, those who reported experiencing both traditional and cyberbullying victimization engaged in ∼85% more counts of violent behaviors than those who did not experience any type of bullying. Male adolescents engaged in 177% more incidents of violent behaviors than their female counterparts. In addition, black adolescents were found to engage in 29.8% more count of violent behaviors than white adolescents, whereas individuals who were categorized as other (including American Indian, Alaska Native, Asian, Native Hawaiian, or Other) were found to engage in 21.1% fewer counts of violent behaviors compared with their white counterparts. As an individual's grade level increased, they participated in 15.3% fewer counts of violent behaviors, whereas if the individual's age increased, they were found to participate in 15.2% more counts of violent behavior.

Table 2.

Negative Binomial Regression Predicting Counts of Violent Behaviors

	IRR	Confidence intervals
Traditional bullying victimization	1.56^***	1.39–1.75
Cyberbullying victimization	1.59^***	1.39–1.80
Both cyber and traditional bullying victimization	1.85^***	1.59–2.14
Male	2.77^***	2.53–3.04
Black	1.30^***	1.16–1.46
Other	0.79^***	0.68–0.92
Age	1.15^***	1.07–1.24
Grade	0.85^***	0.78–0.92
Intercept	0.18^***	0.14–0.23

p < 0.05; ^**p < 0.01; ^***p < 0.001.

IRR, incidence rate ratios.

Discussion

Currently, violence among American youth is a national concern. The U.S. CDC (2017) reports that youth violence is a leading cause of injury and death among adolescents and young adults. Research has demonstrated that adolescents who experience bullying victimization are at higher risk for engaging in violent behaviors and increased aggression (Duggins et al. 2016; Hinduja and Patchin 2008; Reed et al. 2015). Furthermore, existing literature has linked traditional bullying victimization with an extreme form of violence in schools–school shootings (Patchin 2002; Vossekuil et al. 2004). According to the (Vossekuil et al. 2004), 71% of school shooters reported being bullied before their attack. In regard to cyber-victimization, Hinduja and Patchin (2008) found that cyber-victimization among adolescents increased the likelihood of the victim engaging in “offline problem behaviors,” such as assaulting a peer, assaulting an adult, damaging property, and carrying a weapon. This study fills a gap as it also examines the association between violent behaviors and the co-occurrence of traditional and cyber-victimization.

From this study, we see that all types of bullying victimization significantly increase the risk of adolescents engaging in violent behaviors, which is consistent with existing literature (Duggins et al. 2016; Hinduja and Patchin 2008; Reed et al. 2015). Results show adolescents who experienced the co-occurring victimization engaged in higher counts of violent behaviors, than those who only experienced one type of bullying victimization. Adolescents who experience both in-person and online victimization could engage in higher counts of violent behaviors as they feel there is no way for them to “escape” their victimization, which supports existing research (Peleg-Oren et al. 2012; Waasdorp and Bradshaw 2015; Ybarra and Mitchell 2004). In addition, males reported higher counts of violent behaviors than females, which is constant with existing literature (Wasserman et al. 2005). Finally, blacks were found to have higher counts of violent behaviors than white adolescents, while those grouped into Other were found to have fewer counts of violent behaviors than whites. While prior research has found mixed results regarding the role of race and ethnicity and bullying (Carlyle and Steinman 2007; Estell et al. 2007; Fitzpatrick et al. 2007; Goldweber et al. 2013; Schuster et al. 2012; Shetgiri et al. 2012; Wang et al. 2009), the current study finds support to demonstrate higher rates of offending among black students.

The current study helps fill a large gap in the bullying literature. The majority of studies on bullying focus on either traditional or cyber bullying, while not examining the co-occurrence of the two. The limited studies that do examine the types together have small and ungeneralizable sample sizes. The current study provides a much broader scope to understanding the bullying problem, which will ultimately allow school officials and practitioners better insight into how to handle the problem. Bullying should no longer be considered a normative behavior among children and adolescents, rather it should be seen as a potential precursor for more serious violent behaviors, such as weapon carrying, physical fighting, and fighting-related injuries.

Limitations

A potential limitation of this study is the data set that was used. The YRBSS conducted in 2015 version is only the second collection regarding cyber-victimization. Since the data were not collected from the same students over multiple time points, it is impossible to conclude that experiencing bullying victimization is a direct cause for engaging in violent behaviors. In addition, this dataset does not look at the frequency of bullying victimization, rather it only examines whether or not the respondent reports having experienced victimization. As frequency was not examined, we are unable to make a correlation between the frequency of victimization occurrences and the count of violent behaviors an individual took part in. While determining correlation among these variables is a good starting point, future research is needed. A third limitation of this data set is the self-report nature of the survey. Individuals may have underreported their experiences with all of the examined variables as there is a tendency for individuals to provide answers that are socially desirable (Brownfield and Sorenson 1993).

Conclusions

Based on the findings, this study provides evidence for the need to create and implement nationwide policies and programs targeting cyberbullying victimization at both the school and legal level. This study has demonstrated that not only is cyberbullying victimization harmful on its own, but also we now see the detrimental effects of the co-occurrence of cyber- and traditional victimization. Currently, many educational institutions address traditional bullying, yet take a “hands off” approach to cyberbullying. Moreover, many educational institutions do not have official policies in place, and most schools argue that if it does not occur on a computer that is school property, then it is not a school issue. However, this study shows the need to address cyber-victimization among adolescents. In addition, it is imperative that we continue to address the issue of bullying victimization empirically so that educational institutions can adjust their current policies/programs based on factual knowledge rather than on beliefs or anecdotal information. Educators may view bullying as a normative behavior among adolescents that does not have negative impacts on the victimized individual in the long run. However, this study shows that this is not the case.

Footnotes

Author Disclosure Statement

No competing financial interests exist.

References

Brownfield

, Sorenson

. (1993). Self-control and juvenile delinquency: Theoretical issues and an empirical assessment of selected elements of a general theory of crime. . Deviant Behav. 14, 243–264.

Carlyle

, Steinman

. (2007). Demographic differences in the prevalence, cooccurrence, and correlates of adolescent bullying at school. J Sch Health. 77, 623–629.

Centers for Disease Control and Prevention (CDC). (2013). 2013 Youth Risk Behavior Survey. . [Data file]. www.cdc.gov/healthyyouth/yrbs/pdf/questionnaire/2013_xxh_questionnaire.pdf (accessed June 2016 ).

Centers for Disease Control and Prevention (CDC). (2017). Youth violence: Facts at a glance. https://www..cdc.gov/violenceprevention/pdf/yv-datasheet.pdf (accessed February 2018 ).

Duggins

, Kuperminc

, Henrich

, et al. (2016). Aggression among adolescent victims of school bullying: Protective roles of family and school connectedness. Psychol Violence. 6, 205–212.

Elgar

, Napoletano

, Saul

, et al. (2014). Cyberbullying victimization and mental health in adolescents and the moderating role of family dinners. JAMA Pediatr. 168, 1015–1022.

Estell

, Farmer

, Cairns

. (2007). Bullies and victims in rural African American youth: Behavioral characteristics and social network placement. Aggress Behav. 33, 145–159.

Fitzpatrick

, Dulin

, Piko

. (2007). Not just pushing and shoving: School bullying among African American adolescents. J Sch Health. 77, 16–22.

Gladden

, Vivolo-Kantor

, Hamburger

, Lumpkin

. (2014). Bullying surveillance among youths: Uniform definitions for public health and recommended data elements, Version 1.0. (National Center for Injury Prevention and Control, Centers for Disease Control and Prevention, U.S. Department of Education, Atlanta, GA).

10.

Goldweber

, Waasdorp

, Bradshaw

. (2013). Examining associations between race, urbanicity, and patterns of bullying involvement. J Youth Adolesc. 42, 206–219.

11.

Hinduja

, Patchin

. (2008). Cyberbullying: An exploratory analysis of factors related to offending and victimization. Deviant Behav. 29, 129–156.

12.

Hinduja

, Patchin

. (2009). Bullying Beyond the Schoolyard: Preventing and Responding to Cyberbullying. (Sage Publications, Thousand Oaks, CA).

13.

Kim

, Catalano

, Haggerty

, Abbott

. (2011). Bullying at elementary school and problem behaviour in young adulthood: A study of bullying, violence and substance use from age 11 to age 21. Crim Behav Ment Health. 2, 136–144.

14.

Litwiller

, Brausch

. (2013). Cyberbullying and physical bullying in adolescent suicide: The role of violent behavior and substance use. J Youth Adolesc. 42, 675–684.

15.

Mehari

, Farrell

. (2018). Where does cyberbullying fit? A comparison of competing models of adolescent aggression. Psychol Violence., 8, 31–42.

16.

Nansel

, Overpeck

, Haynie

, et al. (2003). Relationships between bullying and violence among US youth. Arch Pediatr Asolesc Med. 4, 348–353.

17.

National Center for Education Statistics. (2016). Student Reports of Bullying: Results from the 2015 School Crime Supplement to the National Crime Victimization Survey. (U.S. Department of Education, Washington, DC).

18.

Patchin

. (2002). Bullied youths lash out: Strain as an explanation of extreme school violence. Caribbean J Criminol Soc Psychol. 7, 22–43.

19.

Patchin

, Hinduja

. (2010). Traditional and nontraditional bullying among youth: A test of general strain theory. Youth Soc. 20, 1–25.

20.

Peleg-Oren

, Cardenas

, Comerford

, Galea

. (2012). An association between bullying behaviors and alcohol use among middle school students. J Early Adolesc. 32, 761–775.

21.

Reed

, Nugent

, Cooper

. (2015). Testing a path model of relationships between gender, age, and bullying victimization and violent behavior, substance abuse, depression, suicidal ideation, and suicide attempts in adolescents. Child Youth Serv Rev. 55, 128–137.

22.

Salmivalli

, Kaukiainen

, Kaistaniemi

, et al. (1999). Self-evaluated self-esteem, peer-evaluated self-esteem, and defensive egotism as predictors of adolescents' participation in bullying situations. Pers Soc Psychol Bull. 25, 1268–1278.

23.

Schuster

, Elliott

, Kanouse

, et al. (2012). Racial and ethnic health disparities among fifth-graders in three cities. N Engl J Med. 367, 735–745.

24.

Shetgiri

, Lin

, Avila

, Flores

. (2012). Parental characteristics associated with bullying perpetration in U.S. children aged 10 to 17 years. Am J Public Health. 102, 2280–2286.

25.

StataCorp. (2015). Stata Statistical Software: Release 14. (StataCorp LP, College Station, TX).

26.

U.S. Department of Health and Human Services (USDHHS), Centers for Disease Control and Prevention. (2013). Methodology of the Youth Risk Behavior Surveillance System. Retrieved from www.cdc.gov/mmwr/pdf/rr/rr6201.pdf

27.

Vossekuil

, Fein

, Reddy

, Borum

, Modzelski

. (2004). The final report and findings of the safe school initiative: Implications for the prevention of school attacks in the United States. Retrieved from https://www2.ed.gov/admins/lead/safety/preventingattacksreport.pdf

28.

Waasdorp

, Bradshaw

. (2015) The overlap between cyberbullying and traditional bullying. J Adolesc Health. 56, 483–488.

29.

Wang

, Iannotti

, Nansel

. (2009). School bullying among adolescents in the United States: Physical, verbal, relational, and cyber. J Adolesc Health. 45, 368–375.

30.

Wasserman

, McReynolds

, Ko

, et al. (2005). Gender differences in psychiatric disorders at juvenile probation intake. Am J Public Health. 95, 131–137.

31.

Willard

. (2004). An Educator's Guide to Cyberbullying and Cyberthreats. Retrieved from http://cyberbully.org/docs/cbcteducator.pdf

32.

Ybarra

, Mitchell

. (2004). Youth engaging in online harassment: associations with caregiver-child relationships, Internet use, and personal characteristics. J Adolesc. 27, 319–336.